- Devel - libvirt List Archives

[libvirt] [PATCH 00/10] tests: More XML 2 XML tests and cleanups
by Cole Robinson 21 Oct '09

21 Oct '09

In attempting to refactor some storage pool XML parsing, I wanted some unit tests to ensure I wasn't causing regressions. So, the following series adds XML 2 XML tests for storage pools (and virtual networks and storage vols). Also included are several cleanups in tests/, and bug fixes needed to make the new tests actually pass. Cole Robinson (10): tests: Fix text output for interface XML 2 XML tests: Document VIR_TEST_DEBUG environment variable. tests: Break out duplicate schema verification functionality. storage: Fix generating iscsi 'auth' xml schema: Update storage pool schema. tests: Add XML 2 XML tests for storage pools. network: Fix reading 'forwardDelay' attribute. schema: Update network schema. tests: Add network XML to XML tests. tests: Add storage volume XML 2 XML tests. HACKING | 10 ++ docs/schemas/network.rng | 41 ++++++- docs/schemas/storagepool.rng | 51 ++++++++- src/conf/network_conf.c | 10 ++- src/conf/storage_conf.c | 2 +- tests/.gitignore | 3 + tests/Makefile.am | 33 +++++- tests/capabilityschematest | 34 +----- tests/domainschematest | 33 +----- tests/interfaceschematest | 31 +----- tests/interfacexml2xmltest.c | 4 +- tests/networkschemadata/netboot-network.xml | 12 -- tests/networkschematest | 33 +----- tests/networkxml2xmlin/forward-delay-network.xml | 13 ++ tests/networkxml2xmlin/isolated-network.xml | 10 ++ tests/networkxml2xmlin/nat-network.xml | 13 ++ tests/networkxml2xmlin/netboot-network.xml | 14 +++ tests/networkxml2xmlin/routed-network.xml | 11 ++ tests/networkxml2xmlout/forward-delay-network.xml | 13 ++ tests/networkxml2xmlout/isolated-network.xml | 10 ++ tests/networkxml2xmlout/nat-network.xml | 13 ++ tests/networkxml2xmlout/netboot-network.xml | 14 +++ tests/networkxml2xmlout/routed-network.xml | 11 ++ tests/networkxml2xmltest.c | 97 +++++++++++++++ tests/nodedevschematest | 31 +----- tests/schematestutils.sh | 39 ++++++ tests/statstest.c | 2 +- tests/storagepoolschemadata/pool-dir.xml | 17 --- tests/storagepoolschemadata/pool-disk.xml | 22 ---- tests/storagepoolschemadata/pool-fs.xml | 19 --- tests/storagepoolschemadata/pool-iscsi.xml | 10 -- tests/storagepoolschemadata/pool-logical.xml | 19 --- tests/storagepoolschemadata/pool-netfs.xml | 20 --- tests/storagepoolschematest | 33 +----- tests/storagepoolxml2xmlin/pool-dir.xml | 18 +++ tests/storagepoolxml2xmlin/pool-disk.xml | 22 ++++ tests/storagepoolxml2xmlin/pool-fs.xml | 19 +++ tests/storagepoolxml2xmlin/pool-iscsi-auth.xml | 17 +++ tests/storagepoolxml2xmlin/pool-iscsi.xml | 16 +++ tests/storagepoolxml2xmlin/pool-logical-create.xml | 20 +++ tests/storagepoolxml2xmlin/pool-logical.xml | 19 +++ tests/storagepoolxml2xmlin/pool-mpath.xml | 12 ++ tests/storagepoolxml2xmlin/pool-netfs.xml | 20 +++ tests/storagepoolxml2xmlin/pool-scsi.xml | 15 +++ tests/storagepoolxml2xmlout/pool-dir.xml | 18 +++ tests/storagepoolxml2xmlout/pool-disk.xml | 19 +++ tests/storagepoolxml2xmlout/pool-fs.xml | 19 +++ tests/storagepoolxml2xmlout/pool-iscsi-auth.xml | 20 +++ tests/storagepoolxml2xmlout/pool-iscsi.xml | 19 +++ .../storagepoolxml2xmlout/pool-logical-create.xml | 22 ++++ tests/storagepoolxml2xmlout/pool-logical.xml | 19 +++ tests/storagepoolxml2xmlout/pool-mpath.xml | 17 +++ tests/storagepoolxml2xmlout/pool-netfs.xml | 20 +++ tests/storagepoolxml2xmlout/pool-scsi.xml | 18 +++ tests/storagepoolxml2xmltest.c | 102 ++++++++++++++++ tests/storagevolschemadata/vol-logical.xml | 20 --- tests/storagevolschemadata/vol-partition.xml | 21 ---- tests/storagevolschemadata/vol-qcow2.xml | 31 ----- tests/storagevolschematest | 33 +----- tests/storagevolxml2xmlin/vol-file-backing.xml | 24 ++++ tests/storagevolxml2xmlin/vol-file.xml | 15 +++ tests/storagevolxml2xmlin/vol-logical-backing.xml | 29 +++++ tests/storagevolxml2xmlin/vol-logical.xml | 20 +++ tests/storagevolxml2xmlin/vol-partition.xml | 21 ++++ tests/storagevolxml2xmlin/vol-qcow2.xml | 31 +++++ tests/storagevolxml2xmlout/vol-file-backing.xml | 27 +++++ tests/storagevolxml2xmlout/vol-file.xml | 18 +++ tests/storagevolxml2xmlout/vol-logical-backing.xml | 26 ++++ tests/storagevolxml2xmlout/vol-logical.xml | 17 +++ tests/storagevolxml2xmlout/vol-partition.xml | 18 +++ tests/storagevolxml2xmlout/vol-qcow2.xml | 31 +++++ tests/storagevolxml2xmltest.c | 123 ++++++++++++++++++++ 72 files changed, 1291 insertions(+), 413 deletions(-) delete mode 100644 tests/networkschemadata/netboot-network.xml create mode 100644 tests/networkxml2xmlin/forward-delay-network.xml create mode 100644 tests/networkxml2xmlin/isolated-network.xml create mode 100644 tests/networkxml2xmlin/nat-network.xml create mode 100644 tests/networkxml2xmlin/netboot-network.xml create mode 100644 tests/networkxml2xmlin/routed-network.xml create mode 100644 tests/networkxml2xmlout/forward-delay-network.xml create mode 100644 tests/networkxml2xmlout/isolated-network.xml create mode 100644 tests/networkxml2xmlout/nat-network.xml create mode 100644 tests/networkxml2xmlout/netboot-network.xml create mode 100644 tests/networkxml2xmlout/routed-network.xml create mode 100644 tests/networkxml2xmltest.c create mode 100644 tests/schematestutils.sh delete mode 100644 tests/storagepoolschemadata/pool-dir.xml delete mode 100644 tests/storagepoolschemadata/pool-disk.xml delete mode 100644 tests/storagepoolschemadata/pool-fs.xml delete mode 100644 tests/storagepoolschemadata/pool-iscsi.xml delete mode 100644 tests/storagepoolschemadata/pool-logical.xml delete mode 100644 tests/storagepoolschemadata/pool-netfs.xml create mode 100644 tests/storagepoolxml2xmlin/pool-dir.xml create mode 100644 tests/storagepoolxml2xmlin/pool-disk.xml create mode 100644 tests/storagepoolxml2xmlin/pool-fs.xml create mode 100644 tests/storagepoolxml2xmlin/pool-iscsi-auth.xml create mode 100644 tests/storagepoolxml2xmlin/pool-iscsi.xml create mode 100644 tests/storagepoolxml2xmlin/pool-logical-create.xml create mode 100644 tests/storagepoolxml2xmlin/pool-logical.xml create mode 100644 tests/storagepoolxml2xmlin/pool-mpath.xml create mode 100644 tests/storagepoolxml2xmlin/pool-netfs.xml create mode 100644 tests/storagepoolxml2xmlin/pool-scsi.xml create mode 100644 tests/storagepoolxml2xmlout/pool-dir.xml create mode 100644 tests/storagepoolxml2xmlout/pool-disk.xml create mode 100644 tests/storagepoolxml2xmlout/pool-fs.xml create mode 100644 tests/storagepoolxml2xmlout/pool-iscsi-auth.xml create mode 100644 tests/storagepoolxml2xmlout/pool-iscsi.xml create mode 100644 tests/storagepoolxml2xmlout/pool-logical-create.xml create mode 100644 tests/storagepoolxml2xmlout/pool-logical.xml create mode 100644 tests/storagepoolxml2xmlout/pool-mpath.xml create mode 100644 tests/storagepoolxml2xmlout/pool-netfs.xml create mode 100644 tests/storagepoolxml2xmlout/pool-scsi.xml create mode 100644 tests/storagepoolxml2xmltest.c delete mode 100644 tests/storagevolschemadata/vol-logical.xml delete mode 100644 tests/storagevolschemadata/vol-partition.xml delete mode 100644 tests/storagevolschemadata/vol-qcow2.xml create mode 100644 tests/storagevolxml2xmlin/vol-file-backing.xml create mode 100644 tests/storagevolxml2xmlin/vol-file.xml create mode 100644 tests/storagevolxml2xmlin/vol-logical-backing.xml create mode 100644 tests/storagevolxml2xmlin/vol-logical.xml create mode 100644 tests/storagevolxml2xmlin/vol-partition.xml create mode 100644 tests/storagevolxml2xmlin/vol-qcow2.xml create mode 100644 tests/storagevolxml2xmlout/vol-file-backing.xml create mode 100644 tests/storagevolxml2xmlout/vol-file.xml create mode 100644 tests/storagevolxml2xmlout/vol-logical-backing.xml create mode 100644 tests/storagevolxml2xmlout/vol-logical.xml create mode 100644 tests/storagevolxml2xmlout/vol-partition.xml create mode 100644 tests/storagevolxml2xmlout/vol-qcow2.xml create mode 100644 tests/storagevolxml2xmltest.c

4 40

[libvirt] [PATCH] do not bother rotating short qemu logs
by Dan Kenigsberg 21 Oct '09

21 Oct '09

without this, after few weeks without use, each defined domain grows a tail of empty gzipped logs, instead of keeping just the last log of interest. --- daemon/libvirtd.logrotate.in | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/daemon/libvirtd.logrotate.in b/daemon/libvirtd.logrotate.in index 093651c..0c51fd3 100644 --- a/daemon/libvirtd.logrotate.in +++ b/daemon/libvirtd.logrotate.in @@ -5,4 +5,5 @@ compress delaycompress copytruncate + minsize 100k } -- 1.6.2.5

3 2

[libvirt] [PATCH] LXC fix virCgroupGetValueStr on handling a string terminated with '\n'
by Ryota Ozaki 21 Oct '09

21 Oct '09

Hi there, A cgroup file returns integer value terminated with '\n' and remaining it has sometimes harmful effects, for example it leads virStrToLong_ull failed. The fix just gets rid of '\n' if included. It first has been in virCgroupGetFreezerState, but now merged in virCgroupGetValueStr to cover every functions using virCgroupGetValueStr. I'm not sure that that behavior is from the beginning or changed in a recent kernel, but the fix should work even though anyway. BTW, by the defect I first got the following error and it made me confused. # virsh -c lxc:/// dominfo 4930 Id: 4930 Name: lxc UUID: 084369a0-956a-3010-fc37-ddeb4d627e69 OS Type: exe Autostart: disable error: this function is not supported by the hypervisor: virNodeGetSecurityModel The really error happens in lxcDomainGetInfo, so I guess something is wrong with internal error propagation. Anyone know this unexpected behavior? Thanks, ozaki-r >From 667efde6ad165479817975c6544f6784e1177a32 Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Fri, 16 Oct 2009 00:13:41 +0900 Subject: [PATCH] LXC fix virCgroupGetValueStr on handling a string terminated with '\n' * src/util/cgroup.c: get rid of '\n' from the return value of virCgroupGetValueStr --- src/util/cgroup.c | 12 +++++------- 1 files changed, 5 insertions(+), 7 deletions(-) diff --git a/src/util/cgroup.c b/src/util/cgroup.c index f728a2f..f0e2960 100644 --- a/src/util/cgroup.c+++ b/src/util/cgroup.c @@ -309,6 +309,10 @@ static int virCgroupGetValueStr(virCgroupPtr group, DEBUG("Failed to read %s: %m\n", keypath); rc = -errno; } else { + /* Terminated with '\n' has sometimes harmful effects to the caller */ + char *p = strchr((const char *)value, '\n'); + if (p) *p = '\0'; + rc = 0; } @@ -969,13 +973,7 @@ int virCgroupSetFreezerState(virCgroupPtr group, const char *state) int virCgroupGetFreezerState(virCgroupPtr group, char **state) { - int ret; - ret = virCgroupGetValueStr(group, + return virCgroupGetValueStr(group, VIR_CGROUP_CONTROLLER_CPU, "freezer.state", state); - if (ret == 0) { - char *p = strchr(*state, '\n'); - if (p) *p = '\0'; - } - return ret; } -- 1.6.2.5

3 4

[libvirt] [PATCH] 528575 avoid libvirtd crash on LCX domain autostart
by Daniel Veillard 21 Oct '09

21 Oct '09

https://bugzilla.redhat.com/show_bug.cgi?id=528575 virsh -c lxc:/// autostart vm1 crash the daemon because the driver autostartDir is never initialized and we do a NULL dereference. The enclosed patch is trivial and avoid the crash, but ... [root@paphio ~]# virsh -c lxc:/// dominfo vm1 Id: - Name: vm1 UUID: d320f760-7541-6633-fd42-4f984fca9f51 OS Type: exe State: shut off CPU(s): 1 Max memory: 500000 kB Used memory: 500000 kB Autostart: enable while the domain is properly flagged as autostarting it doesn't seem to actually autostart, and the driver->autostartDir directory is not created on the fly as I would expect, i.e. /etc/libvirt/lxc/autostart/ still doesn't exist and the domain is marked as off and Autostart: enable I think it would be good to ACK the patch to avoid the crash but we also need to have someone look at the autostart of LXC domains and double check what is going on, maybe I missed something :-) Ozaki, any chance you could look at this ? Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

3 3

[libvirt] [PATCH] Fix virsh.c compilation warning
by Jim Fehlig 21 Oct '09

21 Oct '09

A call to vshError() containing 'doexit' parameter sneaked in after I removed said parameter. Regards, Jim

3 2

[libvirt] Re: [virt-tools-list] libvirt 0.7.2 linker issues
by Cole Robinson 21 Oct '09

21 Oct '09

On 10/19/2009 08:54 AM, Laurent Léonard wrote: > Hi, > > I'm trying to package libvirt 0.7.2 for Debian and get some linker issues : > > make[3]: Entering directory `/home/user/debian/libvirt/libvirt-0.7.2/tools' > CC virsh-console.o > CC virsh-virsh.o > CCLD virsh > ../src/.libs/libvirt.so: undefined reference to `security_getenforce' > ../src/.libs/libvirt.so: undefined reference to `freecon' > ../src/.libs/libvirt.so: undefined reference to `setexeccon' > ../src/.libs/libvirt.so: undefined reference to > `selinux_virtual_domain_context_path' > ../src/.libs/libvirt.so: undefined reference to `context_str' > ../src/.libs/libvirt.so: undefined reference to `context_range_set' > ../src/.libs/libvirt.so: undefined reference to `is_selinux_enabled' > ../src/.libs/libvirt.so: undefined reference to `setfilecon' > ../src/.libs/libvirt.so: undefined reference to `context_new' > ../src/.libs/libvirt.so: undefined reference to `getfilecon' > ../src/.libs/libvirt.so: undefined reference to `getpidcon' > ../src/.libs/libvirt.so: undefined reference to `context_free' > ../src/.libs/libvirt.so: undefined reference to `context_range_get' > ../src/.libs/libvirt.so: undefined reference to `matchpathcon' > ../src/.libs/libvirt.so: undefined reference to > `selinux_virtual_image_context_path' > ../src/.libs/libvirt.so: undefined reference to `security_check_context' > collect2: ld returned 1 exit status > make[3]: *** [virsh] Error 1 > make[3]: Leaving directory `/home/user/debian/libvirt/libvirt-0.7.2/tools' > make[2]: *** [all] Error 2 > make[2]: Leaving directory `/home/user/debian/libvirt/libvirt-0.7.2/tools' > make[1]: *** [all-recursive] Error 1 > make[1]: Leaving directory `/home/user/debian/libvirt/libvirt-0.7.2' > make: *** [all] Error 2 > > Any idea what's wrong ? > > Thank you, Libvirt questions should go to libvirt-list(a)redhat.com (cc'd now). You should pass --with-selinux=no to configure, since those are all libselinux functions. Do you have libselinux installed? If not, there's probably an error in configure.in selinux autodetection. - Cole

5 5

[libvirt] LXC implement missing macaddr assignment feature
by Ryota Ozaki 21 Oct '09

21 Oct '09

Hi, Currently MAC address configuration of container veth is just ignored. This patch implements the missing feature. * src/lxc/veth.c, src/lxc/veth.h: add setMacAddr * src/lxc/lxc_driver.c: set macaddr of container veth if specified ozaki-r >From 38ccccf09526bdd125c717beb13769aa2904e751 Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Mon, 19 Oct 2009 22:13:18 +0900 Subject: [PATCH] LXC implement missing macaddr assignment feature Currently MAC address configuration of container veth is just ignored. This patch implements the missing feature. * src/lxc/veth.c, src/lxc/veth.h: add setMacAddr * src/lxc/lxc_driver.c: set macaddr of container veth if specified --- src/lxc/lxc_driver.c | 11 +++++++++++ src/lxc/veth.c | 31 +++++++++++++++++++++++++++++++ src/lxc/veth.h | 1 + 3 files changed, 43 insertions(+), 0 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 783dfcc..ef97364 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -786,6 +786,17 @@ static int lxcSetupInterfaces(virConnectPtr conn, goto error_exit; } + if (def->nets[i]->mac) { + char macaddr[VIR_MAC_STRING_BUFLEN]; + virFormatMacAddr(def->nets[i]->mac, macaddr); + if (0 != (rc = setMacAddr(containerVeth, macaddr))) { + virReportSystemError(conn, rc, + _("failed to set %s to %s"), + macaddr, containerVeth); + goto error_exit; + } + } + if (0 != (rc = brAddInterface(brctl, bridge, parentVeth))) { virReportSystemError(conn, rc, _("failed to add %s device to %s"), diff --git a/src/lxc/veth.c b/src/lxc/veth.c index 71d7de6..b15df8d 100644 --- a/src/lxc/veth.c +++ b/src/lxc/veth.c @@ -216,3 +216,34 @@ error_out: VIR_FREE(pid); return rc; } + +/** + * setMacAddr + * @iface: name of device + * @macaddr: MAC address to be assigned + * + * Changes the MAC address of the given device with the + * given address using this command: + * ip link set @iface address @macaddr + * + * Returns 0 on success or -1 in case of error + */ +int setMacAddr(const char* iface, const char* macaddr) +{ + int rc = -1; + const char *argv[] = { + "ip", "link", "set", iface, "address", macaddr, NULL + }; + int cmdResult; + + if (NULL == iface) { + goto error_out; + } + + rc = virRun(NULL, argv, &cmdResult); + if (0 == rc) + rc = cmdResult; + +error_out: + return rc; +} diff --git a/src/lxc/veth.h b/src/lxc/veth.h index 429eb3d..8f2f514 100644 --- a/src/lxc/veth.h +++ b/src/lxc/veth.h @@ -20,5 +20,6 @@ int vethCreate(char* veth1, int veth1MaxLen, char* veth2, int vethDelete(const char* veth); int vethInterfaceUpOrDown(const char* veth, int upOrDown); int moveInterfaceToNetNs(const char *iface, int pidInNs); +int setMacAddr(const char* iface, const char* macaddr); #endif /* VETH_H */ -- 1.6.2.5

2 1

[libvirt] Using libvirt to obtain mac address of virtual domain
by Carlen, Devin C. (ARC-A)[PEROT SYSTEMS] 21 Oct '09

21 Oct '09

Greetings, I am new to libvirt and am looking for the most efficient way to programmatically obtain the mac address from a virtual domain. Libvirt is providing access to xen and kvm in my case. A nudge in the right direction would be appreciated! Best Regards, Devin Carlen

4 3

[libvirt] [PATCH] tests: Centralize VIR_TEST_DEBUG lookup, and document it
by Cole Robinson 20 Oct '09

20 Oct '09

Provide a simple interface for other tests to lookup the testDebug variable. Also remove a redundant error message in interface tests. If anyone feels inclined to change this env variable to match the existing LIBVIRT_* format, it should now be easier to do so. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- HACKING | 10 ++++++++++ tests/interfacexml2xmltest.c | 2 -- tests/statstest.c | 4 ++-- tests/testutils.c | 38 ++++++++++++++++++++++++++------------ tests/testutils.h | 4 ++-- tests/testutilsqemu.c | 2 +- 6 files changed, 41 insertions(+), 19 deletions(-) diff --git a/HACKING b/HACKING index bcff8c6..fba7778 100644 --- a/HACKING +++ b/HACKING @@ -37,6 +37,16 @@ and run the tests: The latter test checks for memory leaks. +If you encounter any failing tests, the VIR_TEST_DEBUG environment variable +may help: + + VIR_TEST_DEBUG=1 make check (or) + VIR_TEST_DEBUG=2 make check + +Also, individual tests can be run from inside the 'tests/' directory, like: + + ./qemuxml2xmltest + (6) Update tests and/or documentation, particularly if you are adding a new feature or changing the output of a program. diff --git a/tests/interfacexml2xmltest.c b/tests/interfacexml2xmltest.c index 5ffebad..ed3093c 100644 --- a/tests/interfacexml2xmltest.c +++ b/tests/interfacexml2xmltest.c @@ -43,8 +43,6 @@ static int testCompareXMLToXMLFiles(const char *xml) { ret = 0; fail: - if (ret != 0) - fprintf(stderr, "expected: -------\n%s", actual); free(actual); virInterfaceDefFree(dev); return ret; diff --git a/tests/statstest.c b/tests/statstest.c index 82c0daa..4c2ea7f 100644 --- a/tests/statstest.c +++ b/tests/statstest.c @@ -25,7 +25,7 @@ static int testDevice(const char *path, int expect) if (actual == expect) { return 0; } else { - if (getenv("DEBUG_TESTS")) + if (virtTestGetDebug()) fprintf(stderr, "Expect %-6d Actual %-6d\n", expect, actual); return -1; } @@ -55,7 +55,7 @@ mymain(int argc ATTRIBUTE_UNUSED, * register a handler to stop error messages cluttering * up display */ - if (!getenv("VIR_TEST_DEBUG")) + if (!virtTestGetDebug()) virSetErrorFunc(NULL, testQuietError); #define DO_TEST(dev, num) \ diff --git a/tests/testutils.c b/tests/testutils.c index e6f5e61..9daec6d 100644 --- a/tests/testutils.c +++ b/tests/testutils.c @@ -45,7 +45,7 @@ ((((int) ((T)->tv_sec - (U)->tv_sec)) * 1000000.0 + \ ((int) ((T)->tv_usec - (U)->tv_usec))) / 1000.0) -unsigned int testDebug = 0; +unsigned int testDebug = -1; static unsigned int testOOM = 0; static unsigned int testCounter = 0; @@ -255,10 +255,10 @@ int virtTestDifference(FILE *stream, const char *actualStart = actual; const char *actualEnd = actual + (strlen(actual)-1); - if (!testDebug) + if (!virtTestGetDebug()) return 0; - if (testDebug < 2) { + if (virtTestGetDebug() < 2) { /* Skip to first character where they differ */ while (*expectStart && *actualStart && *actualStart == *expectStart) { @@ -322,12 +322,30 @@ virtTestErrorHook(int n, void *data ATTRIBUTE_UNUSED) } #endif +unsigned int +virtTestGetDebug() { + char *debugStr; + unsigned int debug; + + if (testDebug != -1) + return testDebug; + + testDebug = 0; + + if ((debugStr = getenv("VIR_TEST_DEBUG")) == NULL) + return 0; + + if (virStrToLong_ui(debugStr, NULL, 10, &debug) < 0) + return 0; + + testDebug = debug; + return testDebug; +} int virtTestMain(int argc, char **argv, int (*func)(int, char **)) { - char *debugStr; int ret; #if TEST_OOM int approxAlloc = 0; @@ -344,10 +362,6 @@ int virtTestMain(int argc, virRandomInitialize(time(NULL) ^ getpid())) return 1; - if ((debugStr = getenv("VIR_TEST_DEBUG")) != NULL) { - if (virStrToLong_ui(debugStr, NULL, 10, &testDebug) < 0) - testDebug = 0; - } #if TEST_OOM if ((oomStr = getenv("VIR_TEST_OOM")) != NULL) { @@ -375,7 +389,7 @@ int virtTestMain(int argc, goto cleanup; #if TEST_OOM_TRACE - if (testDebug) + if (virtTestGetDebug()) virAllocTestHook(virtTestErrorHook, NULL); #endif @@ -393,7 +407,7 @@ int virtTestMain(int argc, approxAlloc = virAllocTestCount(); testCounter++; - if (testDebug) + if (virtTestGetDebug()) fprintf(stderr, "%d) OOM...\n", testCounter); else fprintf(stderr, "%d) OOM of %d allocs ", testCounter, approxAlloc); @@ -415,7 +429,7 @@ int virtTestMain(int argc, if (mp && (n % mp) != (worker - 1)) continue; - if (!testDebug) { + if (!virtTestGetDebug()) { if (mp) fprintf(stderr, "%d", worker); else @@ -444,7 +458,7 @@ int virtTestMain(int argc, } } - if (testDebug) + if (virtTestGetDebug()) fprintf(stderr, " ... OOM of %d allocs", approxAlloc); if (ret == EXIT_SUCCESS) diff --git a/tests/testutils.h b/tests/testutils.h index f036e0f..aef1179 100644 --- a/tests/testutils.h +++ b/tests/testutils.h @@ -34,6 +34,8 @@ int virtTestDifference(FILE *stream, const char *expect, const char *actual); +unsigned int virtTestGetDebug(void); + int virtTestMain(int argc, char **argv, int (*func)(int, char **)); @@ -43,6 +45,4 @@ int virtTestMain(int argc, return virtTestMain(argc,argv, func); \ } -extern unsigned int testDebug; - #endif /* __VIT_TEST_UTILS_H__ */ diff --git a/tests/testutilsqemu.c b/tests/testutilsqemu.c index 9269f5c..eeeb5ec 100644 --- a/tests/testutilsqemu.c +++ b/tests/testutilsqemu.c @@ -132,7 +132,7 @@ virCapsPtr testQemuCapsInit(void) { NULL) == NULL) goto cleanup; - if (testDebug) { + if (virtTestGetDebug()) { char *caps_str; caps_str = virCapabilitiesFormatXML(caps); -- 1.6.0.6

2 3

[libvirt] [PATCH] tests: Initialize virRandom in for test suite.
by Cole Robinson 20 Oct '09

20 Oct '09

Otherwise any virRandom calls will result in a segfault. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- tests/testutils.c | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/tests/testutils.c b/tests/testutils.c index 536441a..e6f5e61 100644 --- a/tests/testutils.c +++ b/tests/testutils.c @@ -340,7 +340,8 @@ int virtTestMain(int argc, #endif if (virThreadInitialize() < 0 || - virErrorInitialize() < 0) + virErrorInitialize() < 0 || + virRandomInitialize(time(NULL) ^ getpid())) return 1; if ((debugStr = getenv("VIR_TEST_DEBUG")) != NULL) { -- 1.6.0.6

2 2

[libvirt] error: unable to connect to '/usr/local/var/run/libvirt/libvirt-sock': Connection refused
by Liu, Zhentao 20 Oct '09

20 Oct '09

Hello, I updated libvirt-bin to version 0.7.0. I still got a error: ------------------------------------------------------------------------------------------- root@forest:/tmp/libvirt-0.7.0# virsh -c qemu:///system list error: unable to connect to '/usr/local/var/run/libvirt/libvirt-sock': Connection refused error: failed to connect to the hypervisor ------------------------------------------------------------------------------------------- Any helps? Regards Zhentao

3 6

[libvirt] Re: [PATCH] also allow use of XZ for Qemu image compression
by Antonio Diaz Diaz 19 Oct '09

19 Oct '09

Daniel Veillard wrote: > xz package dependancy is IMHO a small price to pay: Size: 443012 > to garantee reliability Do you know lzip (http://www.nongnu.org/lzip/lzip.html)? Lzip package size is only 55156 bytes, it is stable (xz is beta), some people consider it better designed than xz (http://lpar.ath0.com/2009/09/25/documentation-as-an-indicator-of-code-quali…) and may very well be the LZMA file format that succeeds. Best regards, Antonio.

1 0

[libvirt] [PATCH] Better error reporting for failed migration.
by Chris Lalancette 19 Oct '09

19 Oct '09

The comment in the code says most of it, but when the destination hostname resolution is screwed up, print a proper error instead of the very unhelpful "unknown error". Note that I'm not overly fond of the wording in the error message, so I'm open to suggestions. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- src/qemu/qemu_driver.c | 16 ++++++++++++++++ 1 files changed, 16 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index d37b184..02bb5cb 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6288,6 +6288,22 @@ qemudDomainMigratePrepare2 (virConnectPtr dconn, goto cleanup; } + /* Remember that we are running on the destination. The hostname that + * we resolve here will be used on the source machine in the "migrate" + * monitor command. Because of that, localhost is almost always the + * wrong thing. Adding this check explicitly breaks localhost + * migration, but only for those machines that have improperly + * configured hostname resolution. + */ + if (STREQ(hostname, "localhost")) { + VIR_FREE(hostname); + qemudReportError(dconn, NULL, NULL, VIR_ERR_INVALID_ARG, "%s", + _("Could not resolve destination hostname; " + "either fix destination to resolve hostname, " + "or use the optional URI migration parameter")); + goto cleanup; + } + /* XXX this really should have been a properly well-formed * URI, but we can't add in tcp:// now without breaking * compatability with old targets. We at least make the -- 1.6.0.6

2 2

[libvirt] error: failed to connect to the hypervisor
by Liu, Zhentao 19 Oct '09

19 Oct '09

Hello, I have reinstalled libvirt-bin on ubuntu 9.04. Then I got a error: ------------------------------------------------------------------------ root@forest:/usr/local/etc/logrotate.d# virsh -c qemu:///system list error: failed to connect to the hypervisor ----------------------------------------------------------------------- ps: I have installed kvm on ubuntu. Any ideas? Regards Zhentao

1 0

[libvirt] libvirt and Linux KVM migrate_set_speed
by Monchiero, Matteo 19 Oct '09

19 Oct '09

Does Libvirt has an interface for KVM's migrate_set_speed? Thanks --matteo

2 1

[libvirt] Tyrel Datwyler is out of the office.
by Tyrel Datwyler 19 Oct '09

19 Oct '09

I will be out of the office starting 10/14/2009 and will not return until 10/19/2009. I will respond to your message when I return.

1 0

[libvirt] error: failed to connect to the hypervisor
by Liu, Zhentao 18 Oct '09

18 Oct '09

Hello, I have installed libvirt-bin 0.6.1 on the ubuntu 9.04. When I run "virsh -c qemu:///system list", I get this error: ---------------------------------------------------------------------------- unable to connect to '/var/run/libvirt/libvirt-sock': Permission denied failed to connect to the hypervisor ----------------------------------------------------------------------------- Then I search in the Internet, it may be a bug of version 0.6.1. So I update libvirt-bin to version 0.6.3, but I still get a error: --------------------------------------------------------------------------------- root@forest:/home/zli# virsh -c qemu:///system list error: failed to connect to the hypervisor ---------------------------------------------------------------------------------- I have no idea about the reason. Any helps? Regards Zhentao

2 1

[libvirt] [PATCH] util: Make sure random data is initialized when in virRandom
by Cole Robinson 16 Oct '09

16 Oct '09

When writing some tests, I mistakenly attempted to auto-generate a UUID, which caused a segfault: virRandom was being used without calling virRandomInitialize. Make sure this case can't happen. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/util/util.c | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/src/util/util.c b/src/util/util.c index e5135fc..1554097 100644 --- a/src/util/util.c +++ b/src/util/util.c @@ -1910,6 +1910,7 @@ int virKillProcess(pid_t pid, int sig) } +static int random_initialized = 0; static char randomState[128]; static struct random_data randomData; static virMutex randomLock; @@ -1925,6 +1926,7 @@ int virRandomInitialize(unsigned int seed) &randomData) < 0) return -1; + random_initialized = 1; return 0; } @@ -1932,6 +1934,12 @@ int virRandom(int max) { int32_t ret; + if (!random_initialized) { + /* This can error, but what's worse? Unnoticed bogus random data or + * a segfault? */ + virRandomInitialize(time(NULL) ^ getpid()); + } + virMutexLock(&randomLock); random_r(&randomData, &ret); virMutexUnlock(&randomLock); -- 1.6.5.rc2

2 2

[libvirt] node device libudev backend second look
by Dave Allan 16 Oct '09

16 Oct '09

Attached is a patch against the current head containing an implementation of node device enumeration using libudev. It is complete except for the monitor, but I'm submitting it now as I have a few questions about the implementation that I'd like advice on. They are marked XXX in comments in the patch. The other thing that's not clear to me is how the code generates the tree structure for nodedev-list --tree. I'm setting the parent pointer to what I think is correct, but the tree output is broken. I can dig through it until I understand it, but if anyone is familiar with the implementation and would be willing to take a few minutes to walk me through it, it would save me a bunch of time. I think it's also important that people get the code installed on a variety of systems as soon as possible to shake out the inevitable bugs that will arise from differing device models and code versions, and I'll have the final version with the monitor shortly. Dave >From efedc8fee8127b32fb5b761ce5035902d168a207 Mon Sep 17 00:00:00 2001 From: David Allan <dallan(a)redhat.com> Date: Fri, 25 Sep 2009 10:59:28 -0400 Subject: [PATCH 1/1] A second look at the libudev node device backend. This patch is a very much cleaned up version of the libudev based node device backend with all the device types implemented, although I need some guidance on the USB interface code as I am not especially familiar with the terminology and HAL and udev display the details quite differently. This version of the code uses libudev to retrieve all the device information, so there's no manual mucking about in sysfs except for a little bit of existing code for fibre channel adapters. There is a lot of detail work in this code, so we should try to get people running it on a wide variety of hardware so we can shake out the differences in implementation between the HAL and libudev backends. --- configure.in | 47 ++- daemon/libvirtd.c | 3 +- src/Makefile.am | 16 +- src/conf/node_device_conf.c | 15 +- src/conf/node_device_conf.h | 5 + src/node_device/node_device_driver.c | 12 +- src/node_device/node_device_driver.h | 22 + src/node_device/node_device_hal.h | 19 - ...evice_hal_linux.c => node_device_linux_sysfs.c} | 0 src/node_device/node_device_udev.c | 800 ++++++++++++++++++++ src/node_device/node_device_udev.h | 27 + src/util/util.c | 28 + src/util/util.h | 3 + 13 files changed, 967 insertions(+), 30 deletions(-) rename src/node_device/{node_device_hal_linux.c => node_device_linux_sysfs.c} (100%) create mode 100644 src/node_device/node_device_udev.c create mode 100644 src/node_device/node_device_udev.h diff --git a/configure.in b/configure.in index 2f9db72..25ed55f 100644 --- a/configure.in +++ b/configure.in @@ -1640,7 +1640,7 @@ test "$enable_shared" = no && lt_cv_objdir=. LV_LIBTOOL_OBJDIR=${lt_cv_objdir-.} AC_SUBST([LV_LIBTOOL_OBJDIR]) -dnl HAL or DeviceKit library for host device enumeration +dnl HAL, DeviceKit, or libudev library for host device enumeration HAL_REQUIRED=0.0 HAL_CFLAGS= HAL_LIBS= @@ -1734,8 +1734,46 @@ AM_CONDITIONAL([HAVE_DEVKIT], [test "x$with_devkit" = "xyes"]) AC_SUBST([DEVKIT_CFLAGS]) AC_SUBST([DEVKIT_LIBS]) +UDEV_REQUIRED=143 +UDEV_CFLAGS= +UDEV_LIBS= +AC_ARG_WITH([udev], + [ --with-udev use libudev for host device enumeration], + [], + [with_udev=check]) + +if test "$with_libvirtd" = "no" ; then + with_udev=no +fi +if test "x$with_udev" = "xyes" -o "x$with_udev" = "xcheck"; then + PKG_CHECK_MODULES(UDEV, libudev >= $UDEV_REQUIRED, + [with_udev=yes], [ + if test "x$with_udev" = "xcheck" ; then + with_udev=no + else + AC_MSG_ERROR( + [You must install udev-devel >= $UDEV_REQUIRED to compile libvirt]) + fi + ]) + if test "x$with_udev" = "xyes" ; then + AC_DEFINE_UNQUOTED([HAVE_UDEV], 1, + [use UDEV for host device enumeration]) + + old_CFLAGS=$CFLAGS + old_LDFLAGS=$LDFLAGS + CFLAGS="$CFLAGS $UDEV_CFLAGS" + LDFLAGS="$LDFLAGS $UDEV_LIBS" + AC_CHECK_FUNCS([udev_new],,[with_udev=no]) + CFLAGS="$old_CFLAGS" + LDFLAGS="$old_LDFLAGS" + fi +fi +AM_CONDITIONAL([HAVE_UDEV], [test "x$with_udev" = "xyes"]) +AC_SUBST([UDEV_CFLAGS]) +AC_SUBST([UDEV_LIBS]) + with_nodedev=no; -if test "$with_devkit" = "yes" -o "$with_hal" = "yes"; +if test "$with_devkit" = "yes" -o "$with_hal" = "yes" -o "$with_udev" = "yes"; then with_nodedev=yes AC_DEFINE_UNQUOTED([WITH_NODE_DEVICES], 1, [with node device driver]) @@ -1900,6 +1938,11 @@ AC_MSG_NOTICE([ devkit: $DEVKIT_CFLAGS $DEVKIT_LIBS]) else AC_MSG_NOTICE([ devkit: no]) fi +if test "$with_udev" = "yes" ; then +AC_MSG_NOTICE([ udev: $UDEV_CFLAGS $UDEV_LIBS]) +else +AC_MSG_NOTICE([ udev: no]) +fi if test "$with_netcf" = "yes" ; then AC_MSG_NOTICE([ netcf: $NETCF_CFLAGS $NETCF_LIBS]) else diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index 78dfb2d..0c03aae 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -833,8 +833,7 @@ static struct qemud_server *qemudInitialize(int sigread) { #ifdef WITH_STORAGE_DIR storageRegister(); #endif -#if defined(WITH_NODE_DEVICES) && \ - (defined(HAVE_HAL) || defined(HAVE_DEVKIT)) +#if defined(WITH_NODE_DEVICES) nodedevRegister(); #endif secretRegister(); diff --git a/src/Makefile.am b/src/Makefile.am index d0ef7d1..f6f63a2 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -254,16 +254,20 @@ SECURITY_DRIVER_APPARMOR_SOURCES = \ NODE_DEVICE_DRIVER_SOURCES = \ - node_device/node_device_driver.c node_device/node_device_driver.h + node_device/node_device_driver.c \ + node_device/node_device_driver.h \ + node_device/node_device_linux_sysfs.c NODE_DEVICE_DRIVER_HAL_SOURCES = \ node_device/node_device_hal.c \ - node_device/node_device_hal.h \ - node_device/node_device_hal_linux.c + node_device/node_device_hal.h NODE_DEVICE_DRIVER_DEVKIT_SOURCES = \ node_device/node_device_devkit.c +NODE_DEVICE_DRIVER_UDEV_SOURCES = \ + node_device/node_device_udev.c + ######################### # @@ -642,6 +646,11 @@ libvirt_driver_nodedev_la_SOURCES += $(NODE_DEVICE_DRIVER_DEVKIT_SOURCES) libvirt_driver_nodedev_la_CFLAGS += $(DEVKIT_CFLAGS) libvirt_driver_nodedev_la_LDFLAGS += $(DEVKIT_LIBS) endif +if HAVE_UDEV +libvirt_driver_nodedev_la_SOURCES += $(NODE_DEVICE_DRIVER_UDEV_SOURCES) +libvirt_driver_nodedev_la_CFLAGS += $(UDEV_CFLAGS) +libvirt_driver_nodedev_la_LDFLAGS += $(UDEV_LIBS) +endif if WITH_DRIVER_MODULES libvirt_driver_nodedev_la_LDFLAGS += -module -avoid-version @@ -689,6 +698,7 @@ EXTRA_DIST += \ $(NODE_DEVICE_DRIVER_SOURCES) \ $(NODE_DEVICE_DRIVER_HAL_SOURCES) \ $(NODE_DEVICE_DRIVER_DEVKIT_SOURCES) \ + $(NODE_DEVICE_DRIVER_UDEV_SOURCES) \ $(SECURITY_DRIVER_SELINUX_SOURCES) \ $(SECURITY_DRIVER_APPARMOR_SOURCES) \ $(SECRET_DRIVER_SOURCES) \ diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index f09f814..2322819 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -325,6 +325,9 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, data->usb_if.subclass); virBufferVSprintf(&buf, " <protocol>%d</protocol>\n", data->usb_if.protocol); + if (data->usb_if.interface_name) + virBufferVSprintf(&buf, " <interface_name>%s</interface_name>\n", + data->usb_if.interface_name); if (data->usb_if.description) virBufferVSprintf(&buf, " <description>%s</description>\n", data->usb_if.description); @@ -394,10 +397,19 @@ char *virNodeDeviceDefFormat(virConnectPtr conn, "</media_available>\n", avl ? 1 : 0); virBufferVSprintf(&buf, " <media_size>%llu</media_size>\n", data->storage.removable_media_size); - virBufferAddLit(&buf, " </capability>\n"); + virBufferVSprintf(&buf, " <logical_block_size>%llu" + "</logical_block_size>\n", + data->storage.logical_block_size); + virBufferVSprintf(&buf, " <num_blocks>%llu</num_blocks>\n", + data->storage.num_blocks); } else { virBufferVSprintf(&buf, " <size>%llu</size>\n", data->storage.size); + virBufferVSprintf(&buf, " <logical_block_size>%llu" + "</logical_block_size>\n", + data->storage.logical_block_size); + virBufferVSprintf(&buf, " <num_blocks>%llu</num_blocks>\n", + data->storage.num_blocks); } if (data->storage.flags & VIR_NODE_DEV_CAP_STORAGE_HOTPLUGGABLE) virBufferAddLit(&buf, @@ -1239,6 +1251,7 @@ void virNodeDevCapsDefFree(virNodeDevCapsDefPtr caps) VIR_FREE(data->usb_dev.vendor_name); break; case VIR_NODE_DEV_CAP_USB_INTERFACE: + VIR_FREE(data->usb_if.interface_name); VIR_FREE(data->usb_if.description); break; case VIR_NODE_DEV_CAP_NET: diff --git a/src/conf/node_device_conf.h b/src/conf/node_device_conf.h index 29a4d43..95910c5 100644 --- a/src/conf/node_device_conf.h +++ b/src/conf/node_device_conf.h @@ -101,6 +101,7 @@ struct _virNodeDevCapsDef { unsigned function; unsigned product; unsigned vendor; + unsigned class; char *product_name; char *vendor_name; } pci_dev; @@ -117,10 +118,12 @@ struct _virNodeDevCapsDef { unsigned _class; /* "class" is reserved in C */ unsigned subclass; unsigned protocol; + char *interface_name; char *description; } usb_if; struct { char *address; + unsigned address_len; char *ifname; enum virNodeDevNetCapType subtype; /* LAST -> no subtype */ } net; @@ -139,6 +142,8 @@ struct _virNodeDevCapsDef { } scsi; struct { unsigned long long size; + unsigned long long num_blocks; + unsigned long long logical_block_size; unsigned long long removable_media_size; char *block; char *bus; diff --git a/src/node_device/node_device_driver.c b/src/node_device/node_device_driver.c index 14b3098..f3bd45d 100644 --- a/src/node_device/node_device_driver.c +++ b/src/node_device/node_device_driver.c @@ -70,9 +70,12 @@ static int update_caps(virNodeDeviceObjPtr dev) } -#ifdef __linux__ -static int update_driver_name(virConnectPtr conn, - virNodeDeviceObjPtr dev) +#if defined (__linux__) && defined (HAVE_HAL) +/* XXX Why does this function exist? Are there devices that change + * their drivers while running? Under libudev, most devices seem to + * provide their driver name as a property "DRIVER" */ +static int update_driver_name_hal_linux(virConnectPtr conn, + virNodeDeviceObjPtr dev) { char *driver_link = NULL; char devpath[PATH_MAX]; @@ -754,5 +757,8 @@ int nodedevRegister(void) { #ifdef HAVE_DEVKIT return devkitNodeRegister(); #endif +#ifdef HAVE_UDEV + return udevNodeRegister(); +#endif #endif } diff --git a/src/node_device/node_device_driver.h b/src/node_device/node_device_driver.h index db01624..5be0781 100644 --- a/src/node_device/node_device_driver.h +++ b/src/node_device/node_device_driver.h @@ -45,6 +45,9 @@ int halNodeRegister(void); #ifdef HAVE_DEVKIT int devkitNodeRegister(void); #endif +#ifdef HAVE_UDEV +int udevNodeRegister(void); +#endif void nodeDeviceLock(virDeviceMonitorStatePtr driver); void nodeDeviceUnlock(virDeviceMonitorStatePtr driver); @@ -53,4 +56,23 @@ void registerCommonNodeFuncs(virDeviceMonitorPtr mon); int nodedevRegister(void); +#ifdef __linux__ + +#define check_fc_host(d) check_fc_host_linux(d) +int check_fc_host_linux(union _virNodeDevCapData *d); + +#define check_vport_capable(d) check_vport_capable_linux(d) +int check_vport_capable_linux(union _virNodeDevCapData *d); + +#define read_wwn(host, file, wwn) read_wwn_linux(host, file, wwn) +int read_wwn_linux(int host, const char *file, char **wwn); + +#else /* __linux__ */ + +#define check_fc_host(d) +#define check_vport_capable(d) +#define read_wwn(host, file, wwn) + +#endif /* __linux__ */ + #endif /* __VIR_NODE_DEVICE_H__ */ diff --git a/src/node_device/node_device_hal.h b/src/node_device/node_device_hal.h index c859fe3..8ac8a35 100644 --- a/src/node_device/node_device_hal.h +++ b/src/node_device/node_device_hal.h @@ -22,23 +22,4 @@ #ifndef __VIR_NODE_DEVICE_HAL_H__ #define __VIR_NODE_DEVICE_HAL_H__ -#ifdef __linux__ - -#define check_fc_host(d) check_fc_host_linux(d) -int check_fc_host_linux(union _virNodeDevCapData *d); - -#define check_vport_capable(d) check_vport_capable_linux(d) -int check_vport_capable_linux(union _virNodeDevCapData *d); - -#define read_wwn(host, file, wwn) read_wwn_linux(host, file, wwn) -int read_wwn_linux(int host, const char *file, char **wwn); - -#else /* __linux__ */ - -#define check_fc_host(d) -#define check_vport_capable(d) -#define read_wwn(host, file, wwn) - -#endif /* __linux__ */ - #endif /* __VIR_NODE_DEVICE_HAL_H__ */ diff --git a/src/node_device/node_device_hal_linux.c b/src/node_device/node_device_linux_sysfs.c similarity index 100% rename from src/node_device/node_device_hal_linux.c rename to src/node_device/node_device_linux_sysfs.c diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c new file mode 100644 index 0000000..32fde52 --- /dev/null +++ b/src/node_device/node_device_udev.c @@ -0,0 +1,800 @@ +/* + * node_device_udev.c: node device enumeration - libudev implementation + * + * Copyright (C) 2009 Red Hat + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Author: Dave Allan <dallan(a)redhat.com> + */ + +#include <config.h> +#include <libudev.h> + +#include "node_device_udev.h" +#include "virterror_internal.h" +#include "node_device_conf.h" +#include "node_device_driver.h" +#include "driver.h" +#include "datatypes.h" +#include "logging.h" +#include "memory.h" +#include "util.h" +#include "buf.h" + +#define VIR_FROM_THIS VIR_FROM_NODEDEV + +static virDeviceMonitorStatePtr driverState = NULL; + +/* This function allocates memory from the heap for the property + * value. That memory must be later freed by some other code. */ +static int udevGetDeviceProperty(struct udev_device *udev_device, + const char *property_key, + char **property_value) +{ + const char *udev_value = NULL; + int ret = -1; + + udev_value = udev_device_get_property_value(udev_device, property_key); + if (udev_value == NULL) { + VIR_ERROR(_("udev reports property '%s' does not exist"), + property_key); + goto out; + } + + /* If this allocation is changed, the comment at the beginning + * of the function must also be changed. */ + *property_value = strdup(udev_value); + if (*property_value == NULL) { + VIR_ERROR(_("Failed to allocate memory for " + "property '%s' on device '%s'"), + property_key, udev_device_get_sysname(udev_device)); + goto out; + } + + ret = 0; + +out: + if (ret != 0) { + VIR_ERROR(_("Failed to get udev property '%s' for device '%s'"), + property_key, udev_device_get_sysname(udev_device)); + } + + return ret; +} + + +static int udevGetStringProperty(struct udev_device *udev_device, + const char *property_key, + char **value) +{ + return udevGetDeviceProperty(udev_device, property_key, value); +} + + +static int udevGetIntProperty(struct udev_device *udev_device, + const char *property_key, + int *value, + int base) +{ + char *udev_value = NULL; + int ret = 0; + + ret = udevGetDeviceProperty(udev_device, property_key, &udev_value); + + if (ret == 0) { + ret = virStrToLong_i(udev_value, NULL, base, value); + } + + VIR_FREE(udev_value); + return ret; +} + + +static int udevGetUintProperty(struct udev_device *udev_device, + const char *property_key, + unsigned int *value, + int base) +{ + char *udev_value = NULL; + int ret = 0; + + ret = udevGetDeviceProperty(udev_device, property_key, &udev_value); + + if (ret == 0) { + ret = virStrToLong_ui(udev_value, NULL, base, value); + } + + VIR_FREE(udev_value); + return ret; +} + + +static int udevGetStringSysfsAttr(struct udev_device *udev_device, + const char *attr_name, + char **value) +{ + const char *udev_value = NULL; + int ret = -1; + + udev_value = udev_device_get_sysattr_value(udev_device, attr_name); + + if (udev_value != NULL) { + *value = strdup(udev_value); + if (*value != NULL) { + ret = 0; + } + } + + return ret; +} + + +static int udevGetIntSysfsAttr(struct udev_device *udev_device, + const char *attr_name, + int *value, + int base) +{ + const char *udev_value = NULL; + int ret = -1; + + udev_value = udev_device_get_sysattr_value(udev_device, attr_name); + + if (udev_value != NULL) { + ret = virStrToLong_i(udev_value, NULL, base, value); + } + + return ret; +} + + +static int udevGetUintSysfsAttr(struct udev_device *udev_device, + const char *attr_name, + unsigned int *value, + int base) +{ + const char *udev_value = NULL; + int ret = -1; + + udev_value = udev_device_get_sysattr_value(udev_device, attr_name); + + if (udev_value != NULL) { + ret = virStrToLong_ui(udev_value, NULL, base, value); + } + + return ret; +} + + +static int udevGetUint64SysfsAttr(struct udev_device *udev_device, + const char *attr_name, + uint64_t *value) +{ + const char *udev_value = NULL; + int ret = -1; + + udev_value = udev_device_get_sysattr_value(udev_device, attr_name); + + if (udev_value != NULL) { + ret = virStrToLong_ull(udev_value, NULL, 0, value); + } + + return ret; +} + + +static void udevLogFunction(struct udev *udev ATTRIBUTE_UNUSED, + int priority ATTRIBUTE_UNUSED, + const char *file, + int line, + const char *fn, + const char *fmt, + va_list args) +{ + VIR_ERROR_INT(file, fn, line, fmt, args); +} + + +static int udevProcessPCI(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + const char *devpath = NULL; + int ret = -1; + + devpath = udev_device_get_devpath(device); + + char *p = strrchr(devpath, '/'); + if (p) { + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.domain); + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.bus); + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.slot); + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.function); + } + + udevGetUintSysfsAttr(device, "vendor", &def->caps->data.pci_dev.vendor, 0); + udevGetUintSysfsAttr(device, "product", &def->caps->data.pci_dev.product, 0); + + /* FIXME: to do the vendor name and product name, we have to + * parse /usr/share/hwdata/pci.ids. Look in hal/hald/ids.c + */ + + ret = 0; + + return ret; +} + + +static int udevProcessUSBDevice(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int ret = 0; + udevGetUintProperty(device, "BUSNUM", &def->caps->data.usb_dev.bus, 0); + udevGetUintProperty(device, "DEVNUM", &def->caps->data.usb_dev.device, 0); + udevGetUintProperty(device, "ID_VENDOR_ID", &def->caps->data.usb_dev.vendor, 16); + udevGetStringProperty(device, "ID_VENDOR", &def->caps->data.usb_dev.vendor_name); + udevGetUintProperty(device, "ID_MODEL_ID", &def->caps->data.usb_dev.product, 0); + udevGetStringProperty(device, "ID_MODEL", &def->caps->data.usb_dev.product_name); + + return ret; +} + + +/* XXX I don't fully understand the properties here. Advice needed. */ +static int udevProcessUSBInterface(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int ret = 0; + char *tmp = NULL, *p = NULL; + + udevGetStringProperty(device, + "INTERFACE", + &def->caps->data.usb_if.interface_name); + + tmp = strdup(def->caps->data.usb_if.interface_name); + + p = strrchr(tmp, '/'); + if (p) { + *p = '\0'; + virStrToLong_ui(tmp, NULL, 0, &def->caps->data.usb_if.number); + virStrToLong_ui(p+1, &p, 0, &def->caps->data.usb_if._class); + virStrToLong_ui(p+1, &p, 0, &def->caps->data.usb_if.subclass); + } + + VIR_FREE(tmp); + return ret; +} + + +static int udevProcessNetworkInterface(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int ret = 0; + + udevGetStringSysfsAttr(device, "address", &def->caps->data.net.address); + udevGetUintSysfsAttr(device, "addr_len", &def->caps->data.net.address_len, 0); + + return ret; +} + + +static int udevProcessSCSIHost(struct udev_device *device ATTRIBUTE_UNUSED, + virNodeDeviceDefPtr def) +{ + int ret = 0; + char *filename = NULL; + + filename = basename(def->name); + + if (STRPREFIX(filename, "host")) { + /* XXX There's really no better way to get the host #? */ + ret = virStrToLong_ui(filename + strlen("host"), + NULL, + 0, + &def->caps->data.scsi_host.host); + check_fc_host(&def->caps->data); + check_vport_capable(&def->caps->data); + } + + return ret; +} + + +static int udevProcessSCSIDevice(struct udev_device *device ATTRIBUTE_UNUSED, + virNodeDeviceDefPtr def) +{ + int ret = 0; + char *filename = NULL, *p = NULL; + + filename = basename(def->name); + + virStrToLong_ui(filename, &p, 10, &def->caps->data.scsi.host); + virStrToLong_ui(p+1, &p, 10, &def->caps->data.scsi.bus); + virStrToLong_ui(p+1, &p, 10, &def->caps->data.scsi.target); + virStrToLong_ui(p+1, &p, 10, &def->caps->data.scsi.lun); + + return ret; +} + + +static int udevProcessDisk(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int ret = 0; + + def->caps->data.storage.drive_type = strdup("disk"); + if (def->caps->data.storage.drive_type == NULL) { + ret = -1; + goto out; + } + + udevGetUint64SysfsAttr(device, "size", &def->caps->data.storage.num_blocks); + udevGetUint64SysfsAttr(device, "queue/logical_block_size", + &def->caps->data.storage.logical_block_size); + + def->caps->data.storage.size = def->caps->data.storage.num_blocks * + def->caps->data.storage.logical_block_size; + +out: + return ret; +} + + +static int udevProcessCDROM(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int tmp_int = 0, ret = 0; + + if ((udevGetIntSysfsAttr(device, "removable", &tmp_int, 0) == 0) && + (tmp_int == 1)) { + + def->caps->data.storage.flags |= VIR_NODE_DEV_CAP_STORAGE_REMOVABLE; + + if ((udevGetIntProperty(device, "ID_CDROM_MEDIA", &tmp_int, 0) == 0) && + (tmp_int == 1)) { + + def->caps->data.storage.flags |= + VIR_NODE_DEV_CAP_STORAGE_REMOVABLE_MEDIA_AVAILABLE; + + udevGetUint64SysfsAttr(device, "size", &def->caps->data.storage.num_blocks); + udevGetUint64SysfsAttr(device, "queue/logical_block_size", + &def->caps->data.storage.logical_block_size); + + /* XXX This calculation is wrong for the qemu virtual + * cdrom which reports the size in 512 byte blocks, but + * the logical block size as 2048. I don't have a + * physical cdrom on a devel system to see how they + * behave. */ + def->caps->data.storage.removable_media_size = + def->caps->data.storage.num_blocks * + def->caps->data.storage.logical_block_size; + } + } + + return ret; +} + + +/* This function exists to deal with the case in which a driver does + * not provide a device type in the usual place, but udev told us it's + * a storage device, and we can make a good guess at what kind of + * storage device it is from other information that is provided. */ +static int udevKludgeStorageType(virNodeDeviceDefPtr def) +{ + int ret = -1; + + if (STRPREFIX(def->caps->data.storage.block, "/dev/vd")) { + /* virtio disk */ + def->caps->data.storage.drive_type = strdup("disk"); + if (def->caps->data.storage.drive_type != NULL) { + ret = 0; + } + } + + return ret; +} + + +static void udevStripSpaces(char *s) +{ + if (s == NULL) { + return; + } + + while (virFileStripSuffix(s, " ")) { + /* do nothing */ + ; + } + + return; +} + + +static int udevProcessStorage(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int ret = -1; + + def->caps->data.storage.block = strdup(udev_device_get_devnode(device)); + udevGetStringProperty(device, "DEVNAME", &def->caps->data.storage.block); + udevGetStringProperty(device, "ID_BUS", &def->caps->data.storage.bus); + udevGetStringProperty(device, "ID_SERIAL", &def->caps->data.storage.serial); + udevGetStringSysfsAttr(device, "device/vendor", &def->caps->data.storage.vendor); + udevStripSpaces(def->caps->data.storage.vendor); + udevGetStringSysfsAttr(device, "device/model", &def->caps->data.storage.model); + udevStripSpaces(def->caps->data.storage.model); + /* XXX Is there an equivalent of the HAL hotpluggable property? + * Is the hotpluggable property still relevant? */ + + if (udevGetStringProperty(device, + "ID_TYPE", + &def->caps->data.storage.drive_type) != 0) { + /* If udev doesn't have it, perhaps we can guess it. */ + if (udevKludgeStorageType(def) != 0) { + goto out; + } + } + + /* NB: drive_type has changed from HAL; now it's "cd" instead of "cdrom" */ + if (STREQ(def->caps->data.storage.drive_type, "cd")) { + ret = udevProcessCDROM(device, def); + } else if (STREQ(def->caps->data.storage.drive_type, "disk")) { + ret = udevProcessDisk(device, def); + } else { + goto out; + } + +out: + return ret; +} + + +static int udevSetDeviceType(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + const char *devtype = NULL; + int ret = 0; + + if (VIR_ALLOC(def->caps) != 0) { + ret = -1; + goto out; + } + + devtype = udev_device_get_devtype(device); + + if (udevGetUintProperty(device, + "PCI_CLASS", + &def->caps->data.pci_dev.class, + 0) == 0) { + def->caps->type = VIR_NODE_DEV_CAP_PCI_DEV; + goto out; + } + + if (devtype != NULL && STREQ(devtype, "usb_device")) { + def->caps->type = VIR_NODE_DEV_CAP_USB_DEV; + goto out; + } + + if (devtype != NULL && STREQ(devtype, "usb_interface")) { + def->caps->type = VIR_NODE_DEV_CAP_USB_INTERFACE; + goto out; + } + + /* XXX Do some network interfaces set the device type property? */ + if (udevGetStringProperty(device, + "INTERFACE", + &def->caps->data.net.ifname) == 0) { + + def->caps->type = VIR_NODE_DEV_CAP_NET; + goto out; + } + + if (devtype != NULL && STREQ(devtype, "scsi_host")) { + def->caps->type = VIR_NODE_DEV_CAP_SCSI_HOST; + goto out; + } + + if (devtype != NULL && STREQ(devtype, "scsi_device")) { + def->caps->type = VIR_NODE_DEV_CAP_SCSI; + goto out; + } + + if (devtype != NULL && STREQ(devtype, "disk")) { + def->caps->type = VIR_NODE_DEV_CAP_STORAGE; + goto out; + } + + ret = -1; + +out: + return ret; +} + + +static int udevGetDeviceDetails(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + int ret = 0; + + switch (def->caps->type) { + case VIR_NODE_DEV_CAP_SYSTEM: + /* What do we want to do with system--there's no libudev equivalent. */ + break; + case VIR_NODE_DEV_CAP_PCI_DEV: + ret = udevProcessPCI(device, def); + break; + case VIR_NODE_DEV_CAP_USB_DEV: + ret = udevProcessUSBDevice(device, def); + break; + case VIR_NODE_DEV_CAP_USB_INTERFACE: + ret = udevProcessUSBInterface(device, def); + break; + case VIR_NODE_DEV_CAP_NET: + ret = udevProcessNetworkInterface(device, def); + break; + case VIR_NODE_DEV_CAP_SCSI_HOST: + ret = udevProcessSCSIHost(device, def); + break; + case VIR_NODE_DEV_CAP_SCSI: + ret = udevProcessSCSIDevice(device, def); + break; + case VIR_NODE_DEV_CAP_STORAGE: + ret = udevProcessStorage(device, def); + break; + default: + ret = -1; + break; + } + + return ret; +} + + +static int udevProcessOneDevice(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + struct udev_device *parent = NULL; + virNodeDeviceObjPtr dev = NULL; + const char *parent_name = NULL; + int ret = -1; + + def->name = strdup(udev_device_get_syspath(device)); + udevGetStringProperty(device, "DRIVER", &def->driver); + + /* How to handle parent devices? Shall we change the meaning of + * the parent pointer for the udev backend? What does client code + * use the parent relationship for? We really want to preserve + * the behavior of the HAL backend, but that may be difficult in + * practice. */ + parent = udev_device_get_parent(device); + if (parent != NULL) { + parent_name = udev_device_get_syspath(parent); + if (parent_name != NULL) { + def->parent = strdup(parent_name); + } + } + + if (udevSetDeviceType(device, def) == 0) { + ret = udevGetDeviceDetails(device, def); + } + + if (ret == 0) { + dev = virNodeDeviceAssignDef(NULL, &driverState->devs, def); + dev->devicePath = strdup(udev_device_get_devpath(device)); + virNodeDeviceObjUnlock(dev); + } else { + virNodeDeviceDefFree(def); + } + + return ret; +} + + +static int udevProcessDeviceListEntry(struct udev *udev, + struct udev_list_entry *list_entry) +{ + struct udev_device *device; + virNodeDeviceDefPtr def = NULL; + const char *name = NULL; + int ret = -1; + + name = udev_list_entry_get_name(list_entry); + + if (VIR_ALLOC(def) != 0) { + goto out; + } + + device = udev_device_new_from_syspath(udev, name); + if (device != NULL) { + udevProcessOneDevice(device, def); + udev_device_unref(device); + ret = 0; + } + +out: + return ret; +} + + +static int udevEnumerateDevices(struct udev *udev) +{ + struct udev_enumerate *udev_enumerate = NULL; + struct udev_list_entry *list_entry = NULL; + const char *name = NULL; + int ret = 0; + + udev_enumerate = udev_enumerate_new(udev); + + ret = udev_enumerate_scan_devices(udev_enumerate); + if (0 != ret) { + VIR_ERROR("udev scan devices returned %d\n", ret); + goto out; + } + + udev_list_entry_foreach(list_entry, + udev_enumerate_get_list_entry(udev_enumerate)) { + + udevProcessDeviceListEntry(udev, list_entry); + name = udev_list_entry_get_name(list_entry); + } + +out: + udev_enumerate_unref(udev_enumerate); + return ret; +} + + +static int udevDeviceMonitorShutdown(void) +{ + int ret = 0; + + struct udev_monitor *udev_monitor = NULL; + struct udev *udev = NULL; + + if (driverState) { + + nodeDeviceLock(driverState); + udev_monitor = DRV_STATE_UDEV_MONITOR(driverState); + + if (udev_monitor != NULL) { + udev = udev_monitor_get_udev(udev_monitor); + udev_monitor_unref(udev_monitor); + } + + if (udev != NULL) { + udev_unref(udev); + } + + virNodeDeviceObjListFree(&driverState->devs); + nodeDeviceUnlock(driverState); + virMutexDestroy(&driverState->lock); + VIR_FREE(driverState); + + } else { + ret = -1; + } + + return ret; +} + + +static int udevDeviceMonitorStartup(int privileged ATTRIBUTE_UNUSED) +{ + struct udev *udev = NULL; + struct udev_monitor *udev_monitor = NULL; + int ret = 0; + + if (VIR_ALLOC(driverState) < 0) { + ret = -1; + goto out; + } + + if (virMutexInit(&driverState->lock) < 0) { + VIR_FREE(driverState); + ret = -1; + goto out; + } + + nodeDeviceLock(driverState); + + /* + * http://www.kernel.org/pub/linux/utils/kernel/hotplug/libudev/libudev-udev.h… + * + * indicates no return value other than success, so we don't check + * its return value. + */ + udev = udev_new(); + udev_set_log_fn(udev, udevLogFunction); + + udev_monitor = udev_monitor_new_from_netlink(udev, "udev"); + if (udev_monitor == NULL) { + ret = -1; + goto out; + } + + /* udev can be retrieved from udev_monitor */ + driverState->privateData = udev_monitor; + nodeDeviceUnlock(driverState); + + /* Populate with known devices */ + + if (udevEnumerateDevices(udev) != 0) { + ret = -1; + goto out; + } + +out: + if (ret == -1) { + udevDeviceMonitorShutdown(); + } + return ret; +} + + +static int udevDeviceMonitorReload(void) +{ + return 0; +} + + +static int udevDeviceMonitorActive(void) +{ + /* Always ready to deal with a shutdown */ + return 0; +} + + +static virDrvOpenStatus udevNodeDrvOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (driverState == NULL) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->devMonPrivateData = driverState; + + return VIR_DRV_OPEN_SUCCESS; +} + +static int udevNodeDrvClose(virConnectPtr conn) +{ + conn->devMonPrivateData = NULL; + return 0; +} + +static virDeviceMonitor udevDeviceMonitor = { + .name = "udevDeviceMonitor", + .open = udevNodeDrvOpen, + .close = udevNodeDrvClose, +}; + +static virStateDriver udevStateDriver = { + .initialize = udevDeviceMonitorStartup, + .cleanup = udevDeviceMonitorShutdown, + .reload = udevDeviceMonitorReload, + .active = udevDeviceMonitorActive, +}; + +int udevNodeRegister(void) +{ + VIR_ERROR0("Registering udev node device backend\n"); + + registerCommonNodeFuncs(&udevDeviceMonitor); + if (virRegisterDeviceMonitor(&udevDeviceMonitor) < 0) { + return -1; + } + + return virRegisterStateDriver(&udevStateDriver); +} diff --git a/src/node_device/node_device_udev.h b/src/node_device/node_device_udev.h new file mode 100644 index 0000000..bac628d --- /dev/null +++ b/src/node_device/node_device_udev.h @@ -0,0 +1,27 @@ +/* + * node_device_udev.h: node device enumeration - libudev implementation + * + * Copyright (C) 2009 Red Hat + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Author: Dave Allan <dallan(a)redhat.com> + */ + +#include <libudev.h> +#include <stdint.h> + +#define SYSFS_DATA_SIZE 4096 +#define DRV_STATE_UDEV_MONITOR(ds) ((struct udev_monitor *)((ds)->privateData)) diff --git a/src/util/util.c b/src/util/util.c index 98f8a14..f63bd39 100644 --- a/src/util/util.c +++ b/src/util/util.c @@ -2114,3 +2114,31 @@ void virFileWaitForDevices(virConnectPtr conn) void virFileWaitForDevices(virConnectPtr conn ATTRIBUTE_UNUSED) {} #endif #endif + +int virBuildPathInternal(char **path, ...) +{ + char *path_component = NULL; + virBuffer buf = VIR_BUFFER_INITIALIZER; + va_list ap; + int ret = 0; + + va_start(ap, *path); + + path_component = va_arg(ap, char *); + virBufferAdd(&buf, path_component, -1); + + while ((path_component = va_arg(ap, char *)) != NULL) + { + virBufferAddChar(&buf, '/'); + virBufferAdd(&buf, path_component, -1); + } + + va_end(ap); + + *path = virBufferContentAndReset(&buf); + if (*path == NULL) { + ret = -1; + } + + return ret; +} diff --git a/src/util/util.h b/src/util/util.h index 8679636..da410e1 100644 --- a/src/util/util.h +++ b/src/util/util.h @@ -248,4 +248,7 @@ char *virFileFindMountPoint(const char *type); void virFileWaitForDevices(virConnectPtr conn); +#define virBuildPath(path, ...) virBuildPathInternal(path, __VA_ARGS__, NULL) +int virBuildPathInternal(char **path, ...) __attribute__ ((sentinel)); + #endif /* __VIR_UTIL_H__ */ -- 1.6.4.4

3 3

[libvirt] [RFC PATCH 0/3] Add MAC filtering to qemu
by Gerhard Stenzel 16 Oct '09

16 Oct '09

The following series of patches is a prototype implementation of a ebtables based MAC address filter. I hope to have addressed all the previous comments and I am still interested to get more feedback/comments. At the moment, some defaults are set when libvirtd starts and when a domain is started or destroyed. The plan is to extend this filter capability to the API level and virsh command level. The plan is also to add more filtering features like VLAN filtering and QoS filtering. Thanks, Gerhard --- Gerhard Stenzel (3): add MAC address based port filtering to qemu add build support for ebtables wrapper add ebtables wrapper configure.in | 3 src/Makefile.am | 1 src/libvirt_private.syms | 27 + src/qemu/qemu.conf | 3 src/qemu/qemu_conf.c | 85 ++++ src/qemu/qemu_conf.h | 8 src/qemu/qemu_driver.c | 44 ++ src/util/ebtables.c | 982 ++++++++++++++++++++++++++++++++++++++++++++++ src/util/ebtables.h | 134 ++++++ 9 files changed, 1287 insertions(+), 0 deletions(-) create mode 100644 src/util/ebtables.c create mode 100644 src/util/ebtables.h --

2 6

[libvirt] [PATCH 0/5] Misc changes to domain driver stuff
by Daniel P. Berrange 16 Oct '09

16 Oct '09

The patches that follow make some changes to the domain driver infrastructure, and the QEMU driver. The important goal is to improve the concurrency of the QEMU driver, by allowing multiple readers on the driver (change mutex to a rwlock) and removing the O(n) locks on the virDomainObjPtr instances These have not been heavily tested yet, beyond checking the TCK passes with QEMU driver. I just send them now for early review, while I continue to work on the rest of the QEMU monitor related patches which this series supports. The CIL object locking test also needs to be extended to understand the idea of RWlocks instead of just mutexes so it can properly validate this code Daniel

3 14

[libvirt] [PATCH] LXC complement PATH environment variable of a controller
by Ryota Ozaki 16 Oct '09

16 Oct '09

Hi, lxc controller usually executes 'ip' command to control veths, however now PATH environment variable of the controller is not set (logging code resets environment variables all) and the execution will definitely fail because the controller cannot find the executable. This patch complements PATH with driver's one. The code is just copied from qemu_conf.c Thanks, ozaki-r >From c98136628c1ea0964191be976b23ac99d0f7bb7b Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Wed, 14 Oct 2009 05:09:04 +0900 Subject: [PATCH] LXC complement PATH environment variable of a controller * src/lxc/lxc_driver.c: without PATH, the controller will definitely fail to call ip command --- src/lxc/lxc_driver.c | 13 +++++++++++++ 1 files changed, 13 insertions(+), 0 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 0b614e3..783dfcc 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -1005,6 +1005,19 @@ static int lxcControllerStart(virConnectPtr conn, lenv[lenvc++] = envval; ¥ } while (0) +#define ADD_ENV_COPY(envname) ¥ + do { ¥ + char *val = getenv(envname); ¥ + if (val != NULL) { ¥ + ADD_ENV_PAIR(envname, val); ¥ + } ¥ + } while (0) + + /* + * The controller may call ip command, so we have to remain PATH. + */ + ADD_ENV_COPY("PATH"); + log_level = virLogGetDefaultPriority(); if (virAsprintf(&tmp, "LIBVIRT_DEBUG=%d", log_level) < 0) goto no_memory; -- 1.6.2.5

4 3

[libvirt] [RFC] Support for CPUID masking
by Jiri Denemark 16 Oct '09

16 Oct '09

Hi, We need to provide support for CPU ID masking. Xen and VMware ESX are examples of current hypervisors which support such masking. My proposal is to define new 'cpuid' feature advertised in guest capabilities: <guest> ... <features> <cpuid/> </feature> </guest> When a driver supports cpuid feature, one can use it to mask/check for specific bits returned by CPU ID as follows: <domain> ... <features> <cpuid> <mask level='hex' register='eax|ebx|ecx|edx'>MASK</mask> </cpuid> </features> ... </domain> Where - level is a hexadecimal number used as an input to CPU ID (i.e. eax register), - register is one of eax, ebx, ecx or edx, - and MASK is a string with the following format: mmmm:mmmm:mmmm:mmmm:mmmm:mmmm:mmmm:mmmm with m being 1-bit mask for the corresponding bit in the register. There are three possibilities of specifying what values can be used for 'm': - let it be driver-specific, - define all possible values, - define a common set of values and allow drivers to specify their own additional values. I think the third is the way to go as it lowers the confusion of different values used by different drivers for the same purpose while maintaining the flexibility to support driver-specific masks. The following could be a good set of predefined common values: - 1 force the bit to be 1 - 0 force the bit to be 0 - x don't care, i.e., use driver's default value - T require the bit to be 1 - F require the bit to be 0 Some examples of what it could look like follow: <capabilities> ... <guest> <os_type>xen</os_type> ... <features> <pae/> <cpuid/> </features> </guest> ... </capabilities> <domain type='xen' id='42'> ... <features> <pae/> <acpi/> <apic/> <cpuid> <mask level='1' register='ebx'> xxxx:xxxx:0000:1010:xxxx:xxxx:xxxx:xxxx </mask> <mask level='1' register='ecx'> xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xx1x:xxxx </mask> <mask level='1' register='edx'> xxx1:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx </mask> <mask level='80000001' register='ecx'> xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xx1x </mask> <mask level='80000008' register='ecx'> xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xx00:1001 </mask> </cpuid> </features> ... </domain> What are your opinions about this? Thank for all comments. Jirka

8 12

[libvirt] [PATCH] Avoid crash in virBufferEscapeString
by Laine Stump 15 Oct '09

15 Oct '09

If virBufferEscapeString is called on a buffer that has 0 bytes of space, a size of -1 will be passed to snprintf, resulting in a segmentation fault. This patch checks for 0 space, and grows the buffer if needed prior to determining size. I discovered this when I accidentally made virBufferEscapeString the first function to add something to a newly minted buffer. --- src/util/buf.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/src/util/buf.c b/src/util/buf.c index c802aa2..9681635 100644 --- a/src/util/buf.c +++ b/src/util/buf.c @@ -318,6 +318,12 @@ virBufferEscapeString(const virBufferPtr buf, const char *format, const char *st } *out = 0; + if ((buf->use >= buf->size) && + virBufferGrow(buf, 100) < 0) { + VIR_FREE(escaped); + return; + } + size = buf->size - buf->use - 1; while (((count = snprintf(&buf->content[buf->use], size, format, (char *)escaped)) < 0) || (count >= size - 1)) { -- 1.6.2.5

3 3

[libvirt] [PATCH] Don't copy old machines from a domain which has none
by Mark McLoughlin 15 Oct '09

15 Oct '09

If the the qemu and kvm binaries are the same, we don't include machine types in the kvm domain info. However, the code which refreshes the machine types info from the previous capabilities structure first looks at the kvm domain's info, finds it matches and then copies the empty machine types list over for the top-level qemu domain. That doesn't make sense, we shouldn't copy an empty machin types list. * src/qemu/qemu_conf.c: qemudGetOldMachinesFromInfo(): don't copy an empty machine types list. --- src/qemu/qemu_conf.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index ac63570..b881f1e 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -505,6 +505,9 @@ qemudGetOldMachinesFromInfo(virCapsGuestDomainInfoPtr info, virCapsGuestMachinePtr *list; int i; + if (!info->nmachines) + return 0; + if (!info->emulator || !STREQ(emulator, info->emulator)) return 0; -- 1.6.2.5

2 1

[libvirt] [PATCH] Add ocfs2 to list of fs pool types
by Jim Fehlig 15 Oct '09

15 Oct '09

Ensures mount.ocfs2 is called when <format type="ocfs2"/> is specified in storage pool XML. Regards, Jim

3 3

[libvirt] Re: Libvir Xen and domain events
by Christopher Pyper 14 Oct '09

14 Oct '09

I have been looking into registering some domain events though the python Libvir API. However, I am getting errors and I believe it may be due to lack of event support in Xen. This is the error I am getting: libvir: Xen error : this function is not supported by the hypervisor: xenUnifiedDomainEventRegister Traceback (most recent call last): File "./testcallback.py", line 19, in ? conn.domainEventRegister(callback,None) File "/usr/lib64/python2.4/site-packages/libvirt.py", line 1578, in domainEventRegister if ret == -1: raise libvirtError ('virConnectDomainEventRegister () failed', conn=self) libvirt.libvirtError: this function is not supported by the hypervisor: xenUnifiedDomainEventRegister If this lack of support is the case, what would you recommend is the best route for catching domain events? Currently, I am thinking of just polling regularly and monitoring for changes in domain state. Cheers. Chris Pyper. P.S. Libvir is awesome.

2 2

[libvirt] Release of libvirt-0.7.2
by Daniel Veillard 14 Oct '09

14 Oct '09

This is an intermediate release to flush out one month worth of development and patches, as we have another release scheduled for the 30th October. It is available on ftp://libvirt.org/libvirt as usual and tagged in GIT The main new features are AppArmor security driver and the possibility to define new domain in the ESX driver. The code had a lot of cleanup, including a big tree refactoring and a fairly long list of bug fixes: Features: - sVirt AppArmor security driver (Jamie Strandboge) - Add public API definition for data stream handling (Daniel P. Berrange) - ESX add esxDomainDefineXML() (Matthias Bolte) - LXC: suspend/resume support (Ryota Ozaki) - Big code tree cleanup (Daniel P. Berrange) Documentation: - Documentation and examples for SVirt Apparmor driver (Jamie Strandboge) - Fix documentation and comment typos (Paolo Bonzini) - Fix up a few typos in the tree. (Chris Lalancette) - Fix a typo in virNetHasValidPciAddr() too (Mark McLoughlin) - Fix a typo in virDiskHasValidPciAddr() (Jiri Denemark) - Fix a number of small typos (Dan Kenigsberg) - add doc for graphic and video elements (Florian Vichot) - Fix up 'neccessary -> necessary' in a comment. (Chris Lalancette) - Fix up comments for domainXML{To,From}Native. (Chris Lalancette) - Simple fix of a comment in qemuStringToArgvEnv. (Chris Lalancette) - Add a README file to src/ explaining the directory structure (Daniel P. Berrange) - don't emit trailing blanks into generated and VC'd NEWS file (Jim Meyering) Portability: - Misc win32 build fixes (Daniel P. Berrange) - Don't require full daemon install for libvirt python bindings (Daniel P. Berrange) - Tweak specfile to fix RHEL6 rules & ESX/PHYP enablement (Daniel P. Berrange) - Bug Fixes: - network: Fix printing XML 'delay' attribute (Cole Robinson) - Fix virFileReadLimFD/virFileReadAll to handle EINTR (Daniel P. Berrange) - storage: Fix generating iscsi 'auth' xml (Cole Robinson) - Fix QEMU restore from file in raw format (Daniel P. Berrange) - Take domain type into account when looking up default machine (Mark McLoughlin) - Fix schema to allow missing machine type (Mark McLoughlin) - Fix stream abort upon I/O failure during migration (Daniel P. Berrange) - Create /var/log/libvirt/{lxc,uml} dirs (Mark McLoughlin) - nodedev: Add locking in nodeNumOfDevices (Cole Robinson) - test: Throw a proper error in GetBridgeName (Cole Robinson) - 526769 change logrotate config default to weekly (Daniel Veillard) - Fix emission of domain events messages (Daniel P. Berrange) - unbreak `make rpcgen' (Paolo Bonzini) - unbreak migration (Paolo Bonzini) - Fix USB device re-labelling (Mark McLoughlin) - Avoid a libvirtd crash on broken input 523418 (Daniel Veillard) - Re-label image file backing stores (Mark McLoughlin) - Fix memory leaks in libvirtd's message processing (Matthias Bolte) - Fix QEMU test suite with new VNC env variable (Daniel P. Berrange) - VBox vboxDomainDestroy forgot to wait for completion (Pritesh Kothari) - Vbox call OpenHardDisk with "" instead of NULL (Pritesh Kothari) - Avoid double free in errors in virsh (Jim Fehlig) - Fix crash in device hotplug cleanup code (Daniel P. Berrange) - Maintain value of ctxt->node in virInterfaceDefParseDhcp (Laine Stump) - Fix some XPath relative node resets (Daniel Veillard) - Fix unitialized variable in qemudDomainDetachHostPciDevice() (Charles Duffy) - ESX: Check if a datastore is accessible first (Matthias Bolte) - Fix handling of Xen(ner) detection (Daniel P. Berrange) - Fix xen driver refcounting. (Matthias Bolte) - prevent attempt to call cat -c during virDomainSave to raw (Charles Duffy) - Don't do virSetConnError when virDrvSupportsFeature is successful. (Chris Lalancette) - Fix a double-free in qemudRunLoop() (Chris Lalancette) - Fix leak in PCI hostdev hot-unplug (Mark McLoughlin) - Fix net/disk hot-unplug segfault (Mark McLoughlin) Improvements: - schema: Update storage pool schema. (Cole Robinson) - test: Activate interfaces specified through driver config file. (Cole Robinson) - Rewrite example domain events programm for python (Daniel P. Berrange) - Support a new peer-to-peer migration mode & public API (Daniel P. Berrange) - LXC add augeas support for config file (Amy Griffis) - LXC add driver config file lxc.conf (Amy Griffis) - LXC do not truncate container log files on restart (Amy Griffis) - LXC initialize logging configuration (Amy Griffis) - Add debug for envp[] in virExecWithHook() (Amy Griffis) - Add accessors for logging filters and outputs (Amy Griffis) - Add virFileAbsPath() utility (Amy Griffis) - LXC implement memory control APIs (Ryota Ozaki) - Add a domain argument to SVirt *RestoreImageLabel (Jamie Strandboge) - test: Support loading node device info from file/XML (Cole Robinson) - test: Implement node device driver. (Cole Robinson) - configure: Add explict --with-python option. (Cole Robinson) - Tunnelled migration. (Chris Lalancette) - Various monitor improvements for migration. (Chris Lalancette) - 523639 Allows a <description> tag for domains (Daniel Veillard) - Add src/util/storage_file.c to the POTFILES.in. (Chris Lalancette) - Add a qemu feature flag for unix socket migration. (Chris Lalancette) - Let remoteClientStream only do RX if requested. (Chris Lalancette) - Introduce virStorageFileMetadata structure (Mark McLoughlin) - Allow control over QEMU audio backend (Daniel P. Berrange) - Handle data streams in remote client (Daniel P. Berrange) - Handle outgoing data streams in libvirtd (Daniel P. Berrange) - Handle incoming data streams in libvirtd (Daniel P. Berrange) - Lots of cleanups and improvement on QEmu monitor code (Daniel P. Berrange) - ESX add esxVI_Occurence enum to for occurences (Matthias Bolte) - ESX add x86_64 detection based on the CPUID (Matthias Bolte) - ESX add tests for the VMX to/from domain XML mapping (Matthias Bolte) - ESX Add esxDomainXMLToNative() (Matthias Bolte) - ESX Set challenge for auth callback to hostname (Matthias Bolte) - ESX Add esxNodeGetFreeMemory() (Matthias Bolte) - network: add 'bootp' and 'tftp' config (Paolo Bonzini) - OpenVZ Fix a restriction about domain names (Yuji NISHIDA) - Make pki_check.sh into an installed & supported tool (Daniel P. Berrange) - ESX add support for vmxnet3 virtual device (Shahar Klein) Cleanups: - remote: Don't print a warning every time a remote call fails (Cole Robinson) - storage: Report errors in FindPoolSources (Cole Robinson) - LXC fix return code handling in lxcVmStart (Ryota Ozaki) - Add a target for libvirt.devhelp (Daniel Veillard) - Remove some auto-generated files (Daniel P. Berrange) - Re-arrange doTunnelMigrate to simplify cleanup code (Daniel P. Berrange) - Separate out code for sending tunnelled data (Daniel P. Berrange) - Pull connection handling code out of doTunnelMigrate (Daniel P. Berrange) - Refactor native QEMU migration code (Daniel P. Berrange) - Don't force dconn to be NULL in virDomainMigrate (Daniel P. Berrange) - Remove unneccessary uri_in parameter from virMigratePrepareTunnel (Daniel P. Berrange) - Move the VIR_DRV_FEATURE* constants (Daniel P. Berrange) - Fix configure.ac message vertical alignment (Daniel P. Berrange) - cgroup: Fix -Werror breakage (Cole Robinson) - Fix handling return value of qemuMonitorSetBalloon (Ryota Ozaki) - Fix up "make check" (Chris Lalancette) - Fix rebuilding of devhelp files (Daniel P. Berrange) - Fix ordering of <exports> in API description file (Daniel P. Berrange) - node conf: Make parsing routines consistent with other drivers (Cole Robinson) - nodedev: Break out virNodeDeviceHasCap to node_conf (Cole Robinson) - python: Add a newline after custom classes (Cole Robinson) - python: Fix generated virInterface method names (Cole Robinson) - python: Use a pure python implementation of 'vir*GetConnect' (Cole Robinson) - python: Don't generate bindings for vir*Ref (Cole Robinson) - python: Don't generate conflicting conn.createXML functions. (Cole Robinson) - python: Remove use of xmllib in generator.py (Cole Robinson) - python: Remove FastParser from generator. (Cole Robinson) - Fix typo in Makefile.am breaking NEWS file generation (Daniel P. Berrange) - Fix build in separate build directory (Jiri Denemark) - Incorrect error message in virDomainNetDefParseXML (Florian Vichot) - Fix a few 'make rpm' breakages (Daniel Veillard) - Pass remote_message_header to the dispatch functions. (Chris Lalancette) - Fix up some warnings from stream DEBUG statements. (Chris Lalancette) - Fix apibuild.py warnings (Matthias Bolte) - Change signature of remoteSendStreamData() to fix compile warning (Matthias Bolte) - Add virStorageFileGetMetadata() helper (Mark McLoughlin) - Move virStorageGetMetadataFromFD() to libvirt_util (Mark McLoughlin) - Split virStorageGetMetadataFromFD() from virStorageBackendProbeTarget() (Mark McLoughlin) - Move file format enum to libvirt_util (Mark McLoughlin) - Remove hand-crafted UUID parsers (Daniel P. Berrange) - Helper functions for processing data streams in libvirtd (Daniel P. Berrange) - Standardize debugging messages in QEMU monitor code (Daniel P. Berrange) - Remove low level monitor APIs from header file (Daniel P. Berrange) - Rename qemudMonitorSendCont to qemuMonitorStartCPUs (Daniel P. Berrange) - Pull QEMU monitor interaction out to separate file (Daniel P. Berrange) - util.h needs libvirt.h for virConnectPtr (Mark McLoughlin) - Fix API doc extractor to stop munging comment formatting (Daniel P. Berrange) - Fix secret_driver compile warning, bug. (Charles Duffy) - ESX remove phantom mode (Matthias Bolte) - ESX replace esxUtil_EqualSuffix() with virFileHasSuffix() (Matthias Bolte) - ESX Whitespace cleanup (Matthias Bolte) - Fix up "make syntax-check" after the tree restructuring. (Chris Lalancette) - Introduce virStrncpy. (Chris Lalancette) - Ignore auto-generated header file (Daniel P. Berrange) - Remove an unnecessary variable from remoteIOReadMessage(). (Chris Lalancette) - Remove auto-generated header file from repo (Daniel P. Berrange) - Move example XML files into examples/xml (Daniel P. Berrange) - Remove all generated docs from source control (Daniel P. Berrange) - Fix missing data file in qemuhelpdata (Daniel P. Berrange) - Misc syntax-check fixes (Daniel P. Berrange) - Move remote protocol definition into src/remote/ (Daniel P. Berrange) - Move all shared utility files to src/util/ (Daniel P. Berrange) - Move all XML configuration handling to src/conf/ (Daniel P. Berrange) - Re-arrange python generator to make it clear what's auto-generated (Daniel P. Berrange) - Remove obsolete files (Daniel P. Berrange) - Move docs/examples into examples/ (Daniel P. Berrange) - Remove unused images from docs/ directory (Daniel P. Berrange) - Rename daemon main code (Daniel P. Berrange) - Move config files to align with driver sources (Daniel P. Berrange) - Move virsh into tools/ directory (Daniel P. Berrange) - Move security drivers to src/security/ (Daniel P. Berrange) - Move secret driver into src/secret/ (Daniel P. Berrange) - Move netcf interface driver into src/interface/ (Daniel P. Berrange) - Move network driver into src/network (Daniel P. Berrange) - Move remote driver to src/remote/ (Daniel P. Berrange) - Move test driver into src/test/ (Daniel P. Berrange) - Move node device drivers to src/node_device/ (Daniel P. Berrange) - Move storage drivers into src/storage/ (Daniel P. Berrange) - Move OpenVZ driver to src/openvz/ (Daniel P. Berrange) - Move UML driver to src/uml/ (Daniel P. Berrange) - Move QEMU driver to src/qemu/ (Daniel P. Berrange) - Move LXC driver into src/lxc/ (Daniel P. Berrange) - Move xen driver code into src/xen/ directory (Daniel P. Berrange) - Rename qemud/ directory to daemon/ (Daniel P. Berrange) - Refactor libvirt.spec to allow client-only builds (Daniel P. Berrange) Please give it a try, and thanks everybody for reports, patches or documentation ! Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt]: Not able to login to container when cgroup is enabled
by Rishikesh 14 Oct '09

14 Oct '09

Hi All, On F12 i was playing with libvirt support for container. I am seeing two issue over here: 1> Not able to create a container when start cgconfig services. 2> When i stop cgconfig, then able to create a container, but seeing few error messages to libvirtd daemon. 3> After creation of container through libvirt, i am not able to touch any file or directory. Attached: vm1.xml , through which i created container. I referred following link: http://libvirt.org/drvlxc.html http://openvz.org/pipermail/devel/2008-September/014314.html Detailed Bug Report: When i create a container without starting "cgconfig" service, i am able to create but getting some error message in libvirtd daemon: [root@mx3950 ~]# libvirtd 14:47:24.556: warning : qemudStartup:565 : Unable to create cgroup for driver: No such device or address 14:47:24.647: warning : lxcStartup:1462 : Unable to create cgroup for driver: No such device or address libvir: Linux Container error : failed to connect to client socket: No such file or directory libvir: Linux Container error : failed to connect to client socket: No such file or directory libvir: Linux Container error : failed to connect to client socket: No such file or directory But when i start cgconfig service, then i am not able to open console for my container. Below is the detailed steps which i executed. [root@mx3950 ~]# /etc/init.d/cgconfig stop Stopping cgconfig service: [ OK ] [root@mx3950 ~]# virsh -c lxc:/// define vm1.xml error: Failed to define domain from vm1.xml error: operation failed: domain 'vm1' is already defined with uuid ef558ab3-54ea-5b95-205d-8c0096ae9a4f [root@mx3950 ~]# virsh -c lxc:/// start vm1 Domain vm1 started [root@mx3950 ~]# virsh -c lxc:/// console vm1 Connected to domain vm1 Escape character is ^] sh-4.0# ps -ef UID PID PPID C STIME TTY TIME CMD root 1 0 0 15:23 pts/4 00:00:00 /bin/sh root 4 1 0 15:23 pts/4 00:00:00 ps -ef sh-4.0# touch rishi touch: cannot touch `rishi': Permission denied sh-4.0# exit exit [root@mx3950 ~]# /etc/init.d/cgconfig start Starting cgconfig service: [ OK ] [root@mx3950 ~]# virsh -c lxc:/// start vm1 Domain vm1 started [root@mx3950 ~]# virsh -c lxc:/// console vm1 Connected to domain vm1 Escape character is ^] 15:24:12.869: error : vshRunConsole:77 : unable to open tty /dev/pts/3: No such file or directory [root@mx3950 ~]# Package List: [root@mx3950 ~]# rpm -qa | grep -e libcgroup -e libvirt libcgroup-0.34-2.fc12.x86_64 libvirt-0.7.1-1.fc12.x86_64 libvirt-client-0.7.1-1.fc12.x86_64 libvirt-python-0.7.1-1.fc12.x86_64 [root@mx3950 ~]# uname -a Linux mx3950.in.ibm.com 2.6.31-14.fc12.x86_64 #1 SMP Tue Sep 15 03:48:57 EDT 2009 x86_64 x86_64 x86_64 GNU/Linux

5 17

[libvirt] Execute script on migration
by Radek Hladík 13 Oct '09

13 Oct '09

Short question: What is the best way to run some script on VM migration? Detailed description: We use own system for volume management. Its based on md and we found it very flexible. Basic idea is that we have md mirror everywhere we can. This suits great for libvirt: We have a disk on iSCSI server, its "partioned" using device mapper, the resulting "partition" is member of md raid1, that raid is exported via iSCSI to the libvirt node, it is on the node included in another md raid1 and that "final" md raid is used as storage by the VM. This provides high level of abstraction that allows us to migrate the storage to another disk, another iSCSI server, etc... The only glitch is on live migration of the VM. It is possible to assemble and run the array on two nodes. Also all superblock modifications are done on first write to the array. However it would be more secure to stop the array after the source vm stops and before destination vm starts. What is the best way to run some script in this particular moment? I know we could create a new storage driver but I am not sure that it is the best option. -- Open WebMail Project (http://openwebmail.org)

3 2

[libvirt] [PATCH] test: Activate interfaces specified through driver config file.
by Cole Robinson 13 Oct '09

13 Oct '09

This matches the behavior of domains, storage, etc. Also, fix up some whitespace issues in the area. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/test/test_driver.c | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 0b866d2..0541a73 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -779,13 +779,16 @@ static int testOpenFromFile(virConnectPtr conn, if (!def) goto error; } else { - if ((def = virInterfaceDefParseNode(conn, xml, ifaces[i])) == NULL) + if ((def = virInterfaceDefParseNode(conn, xml, ifaces[i])) == NULL) goto error; } + if (!(iface = virInterfaceAssignDef(conn, &privconn->ifaces, def))) { virInterfaceDefFree(def); goto error; } + + iface->active = 1; virInterfaceObjUnlock(iface); } VIR_FREE(ifaces); -- 1.6.5.rc2

3 3

[libvirt] [PATCH] storage: Report errors in FindPoolSources
by Cole Robinson 13 Oct '09

13 Oct '09

Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/storage/storage_driver.c | 15 ++++++++++++--- 1 files changed, 12 insertions(+), 3 deletions(-) diff --git a/src/storage/storage_driver.c b/src/storage/storage_driver.c index 9ab53e1..4f8949b 100644 --- a/src/storage/storage_driver.c +++ b/src/storage/storage_driver.c @@ -440,15 +440,24 @@ storageFindPoolSources(virConnectPtr conn, char *ret = NULL; backend_type = virStoragePoolTypeFromString(type); - if (backend_type < 0) + if (backend_type < 0) { + virStorageReportError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown storage pool type %s"), type); goto cleanup; + } backend = virStorageBackendForType(backend_type); if (backend == NULL) goto cleanup; - if (backend->findPoolSources) - ret = backend->findPoolSources(conn, srcSpec, flags); + if (!backend->findPoolSources) { + virStorageReportError(conn, VIR_ERR_NO_SUPPORT, + _("pool type '%s' does not support source " + "discovery"), type); + goto cleanup; + } + + ret = backend->findPoolSources(conn, srcSpec, flags); cleanup: return ret; -- 1.6.5.rc2

3 3

[libvirt] [PATCH] remote: Don't print a warning every time a remote call fails
by Cole Robinson 13 Oct '09

13 Oct '09

$ sudo virsh pool-start idontexist 10:58:18.716: warning : processCallDispatchReply:7612 : Method call error error: failed to get pool 'idontexist' error: Storage pool not found: no pool with matching name 'idontexist' That warning doesn't server much purpose being printed via a virsh call. Change the warning to a debug message. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/remote/remote_driver.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index 284593a..162d682 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -7606,7 +7606,7 @@ processCallDispatchReply(virConnectPtr conn, struct private_data *priv, return 0; case REMOTE_ERROR: - VIR_WARN0("Method call error"); + VIR_DEBUG0("Method call error"); memset (&thecall->err, 0, sizeof thecall->err); if (!xdr_remote_error (xdr, &thecall->err)) { error (in_open ? NULL : conn, -- 1.6.5.rc2

2 2

[libvirt] [PATCH] LXC fix rc handling in lxcVmStart
by Ryota Ozaki 13 Oct '09

13 Oct '09

Hi, In lxcVmStart we have to set rc = -1 before jumping to the cleanup code when a local procedure call fails and the rc is overwritten, however, some codes don't observe the rule and so invalid cleanups are likely to happen. A simple solution for the defect is to add rc = -1 before every jumping if needed, however, IMHO, we first should not reuse rc for another purpose rather than the return value of lxcVmStart. Instead the patch introduces yet another variable for storing only the return value of the local procedure calls. By doing so, we don't need to care about resetting rc anymore. Thanks, ozaki-r >From c3eb2cfa1280b220d5b92ad2ad8bbf6cf6ca0a04 Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Tue, 13 Oct 2009 00:36:41 +0900 Subject: [PATCH] LXC fix rc handling in lxcVmStart * src/lxc/lxc_driver.c: don't reuse rc for local procedure calls --- src/lxc/lxc_driver.c | 15 ++++++--------- 1 files changed, 6 insertions(+), 9 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index ef0d368..0b614e3 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -1117,7 +1117,7 @@ static int lxcVmStart(virConnectPtr conn, lxc_driver_t * driver, virDomainObjPtr vm) { - int rc = -1; + int rc = -1, r; unsigned int i; int parentTty; char *parentTtyPath = NULL; @@ -1126,8 +1126,8 @@ static int lxcVmStart(virConnectPtr conn, unsigned int nveths = 0; char **veths = NULL; - if ((rc = virFileMakePath(driver->logDir)) < 0) { - virReportSystemError(conn, rc, + if ((r = virFileMakePath(driver->logDir)) < 0) { + virReportSystemError(conn, r, _("cannot create log directory '%s'"), driver->logDir); return -1; @@ -1157,10 +1157,8 @@ static int lxcVmStart(virConnectPtr conn, goto cleanup; /* Persist the live configuration now we have veth & tty info */ - if (virDomainSaveConfig(conn, driver->stateDir, vm->def) < 0) { - rc = -1; + if (virDomainSaveConfig(conn, driver->stateDir, vm->def) < 0) goto cleanup; - } if ((logfd = open(logfile, O_WRONLY | O_APPEND | O_CREAT, S_IRUSR|S_IWUSR)) < 0) { @@ -1183,11 +1181,10 @@ static int lxcVmStart(virConnectPtr conn, goto cleanup; /* And get its pid */ - if ((rc = virFileReadPid(driver->stateDir, vm->def->name, &vm->pid)) != 0) { - virReportSystemError(conn, rc, + if ((r = virFileReadPid(driver->stateDir, vm->def->name, &vm->pid)) != 0) { + virReportSystemError(conn, r, _("Failed to read pid file %s/%s.pid"), driver->stateDir, vm->def->name); - rc = -1; goto cleanup; } -- 1.6.2.5

4 3

[libvirt] unable to start guest under kvm (revisited)
by Anton Protopopov 13 Oct '09

13 Oct '09

The problem, that I post here http://www.mail-archive.com/libvir-list@redhat.com/msg16154.html and that was solved, comes back in a different way. Consider a working kvm machine with the following description: ... <os> <type arch='x86_64' machine='pc'>hvm</type> <boot dev='hd'/> </os> ... If I delete "machine='pc'" attribute from <type>, so description becomes ... <os> <type arch='x86_64'>hvm</type> <boot dev='hd'/> </os> ... then libvirt chooses a machine='pc-0.11' value (again: qemu API version instead of kvm version) and I get virsh # start DD error: Failed to start domain DD error: internal error unable to start guest: Supported machines are: pc Standard PC (default) isapc ISA-only PC --- This problem actually occurs when using virt-manager, see https://www.redhat.com/archives/virt-tools-list/2009-October/msg00037.html

4 6

[libvirt] feature request: virDomainSendkeys API
by Dan Kenigsberg 13 Oct '09

13 Oct '09

Our QA team makes use of qemu's sendkeys monitor command to test interesting things such as Windows guest installation-from-scratch. When we move to libvirt, we loose this ability. On irc, dpb suggested using vnc to send the required keys. Are you aware of a command-line tool for that, or should we build our own? If we go down this path, we would have to build a similar tool for testing VMs with SPICE. (and have nothing to do for VMs with no graphics support, but these are not used in the product anyway) Alternatively, libvirt can supply a new API for injecting key strokes into a domain. Surely it's useful only for debug and qa, but so is virDomainMemoryPeek. Is that reasonable? Dan.

3 2

[libvirt] cannot set ownership on /nfs/mounted/image.iso: Operation not permitted
by Dan Kenigsberg 13 Oct '09

13 Oct '09

I having libvirt run qemu as non-root, but I don't think libvirt should try chowning image files. Cerntainly not if they sit on a root_squashing nfs server. Also, the behavior of libvirt in this regard is a bit surprising for local image files and even for host devices. It should be enough that an image is accessible to qemu; it does not have to be owned by it. The fact that qemuDomainSetAllDeviceOwnership() with restore=1 does not really restore ownership of device/file but force-chowns it to root:root is anouther surprise. Dan

2 1

[libvirt] [PATCH] Fix the spec file
by Daniel Veillard 13 Oct '09

13 Oct '09

I think I erased a % when trying to apply the last patch from Amy, trivial fix, will commit, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] [PATCH] Fix QEMU restore from file in raw format
by Daniel P. Berrange 13 Oct '09

13 Oct '09

The logic for running the decompression programs was broken in commit f238709304f9f6c57204cdd943e542cbae38fa5f, so that for non-raw formats the decompression program was never run, and for raw formats, it tried to exec an argv[] with initial NULL in the program name. * src/qemu/qemu_driver.c: Fix logic in runing decompression program --- src/qemu/qemu_driver.c | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 1e23a9b..556984a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3781,9 +3781,8 @@ static int qemudDomainRestore(virConnectPtr conn, goto cleanup; } - if (header.compressed != QEMUD_SAVE_FORMAT_RAW) + if (header.compressed != QEMUD_SAVE_FORMAT_RAW) { intermediate_argv[0] = prog; - else { intermediatefd = fd; fd = -1; if (virExec(conn, intermediate_argv, NULL, NULL, -- 1.6.2.5

3 2

[libvirt] [PATCH 0/9] Support disk-hotremove and controller hotplugging
by Wolfgang Mauerer 13 Oct '09

13 Oct '09

Hi, this patch reworks libvirt's disk-hotadd support and introduces support for disk controller hotplugging and disk-hotremove (see http://thread.gmane.org/gmane.comp.emulators.libvirt/15860 for more details). Currently, it targets only qemu and also requires some additions to qemu that have only recently been submitted, which is why this is cross-posted to qemu-devel. Hopefully the lack of support in qemu does not prevent the community from reviewing the code, though ;-) Notice that no documentation and testcases are included yet; I'll supply those once the interface is fixed and possible issues are settled. Thanks for your comments, Wolfgang docs/schemas/domain.rng | 145 ++++++++++---- src/domain_conf.c | 208 ++++++++++++++++++++- src/domain_conf.h | 39 ++++- src/libvirt_private.syms | 4 +- src/qemu_driver.c | 479 +++++++++++++++++++++++++++++++++++++++++++--- 5 files changed, 802 insertions(+), 73 deletions(-) -- Siemens AG, Corporate Technology, CT SE 2 Corporate Competence Center Embedded Linux

5 26

[libvirt] [PATCH] Fix virFileReadLimFD/virFileReadAll to handle EINTR
by Daniel P. Berrange 13 Oct '09

13 Oct '09

The fread_file_lim() function uses fread() but never handles EINTR results, causing unexpected failures when reading QEMU help arg info. It was unneccessarily using FILE * instead of plain UNIX file handles, which prevented use of saferead() * src/util/util.c: Switch fread_file_lim over to use saferead instead of fread, remove FILE * use, and rename --- src/util/util.c | 45 ++++++++++++--------------------------------- 1 files changed, 12 insertions(+), 33 deletions(-) diff --git a/src/util/util.c b/src/util/util.c index e5135fc..98f8a14 100644 --- a/src/util/util.c +++ b/src/util/util.c @@ -898,7 +898,7 @@ virExec(virConnectPtr conn, number of bytes. If the length of the input is <= max_len, and upon error while reading that data, it works just like fread_file. */ static char * -fread_file_lim (FILE *stream, size_t max_len, size_t *length) +saferead_lim (int fd, size_t max_len, size_t *length) { char *buf = NULL; size_t alloc = 0; @@ -906,8 +906,8 @@ fread_file_lim (FILE *stream, size_t max_len, size_t *length) int save_errno; for (;;) { - size_t count; - size_t requested; + int count; + int requested; if (size + BUFSIZ + 1 > alloc) { alloc += alloc / 2; @@ -923,12 +923,12 @@ fread_file_lim (FILE *stream, size_t max_len, size_t *length) /* Ensure that (size + requested <= max_len); */ requested = MIN (size < max_len ? max_len - size : 0, alloc - size - 1); - count = fread (buf + size, 1, requested, stream); + count = saferead (fd, buf + size, requested); size += count; if (count != requested || requested == 0) { save_errno = errno; - if (ferror (stream)) + if (count < 0) break; buf[size] = '\0'; *length = size; @@ -941,12 +941,12 @@ fread_file_lim (FILE *stream, size_t max_len, size_t *length) return NULL; } -/* A wrapper around fread_file_lim that maps a failure due to +/* A wrapper around saferead_lim that maps a failure due to exceeding the maximum size limitation to EOVERFLOW. */ -static int virFileReadLimFP(FILE *fp, int maxlen, char **buf) +int virFileReadLimFD(int fd, int maxlen, char **buf) { size_t len; - char *s = fread_file_lim (fp, maxlen+1, &len); + char *s = saferead_lim (fd, maxlen+1, &len); if (s == NULL) return -1; if (len > maxlen || (int)len != len) { @@ -960,37 +960,16 @@ static int virFileReadLimFP(FILE *fp, int maxlen, char **buf) return len; } -/* Like virFileReadLimFP, but use a file descriptor rather than a FILE*. */ -int virFileReadLimFD(int fd_arg, int maxlen, char **buf) -{ - int fd = dup (fd_arg); - if (fd >= 0) { - FILE *fp = fdopen (fd, "r"); - if (fp) { - int len = virFileReadLimFP (fp, maxlen, buf); - int saved_errno = errno; - fclose (fp); - errno = saved_errno; - return len; - } else { - int saved_errno = errno; - close (fd); - errno = saved_errno; - } - } - return -1; -} - int virFileReadAll(const char *path, int maxlen, char **buf) { - FILE *fh = fopen(path, "r"); - if (fh == NULL) { + int fd = open(path, O_RDONLY); + if (fd < 0) { virReportSystemError(NULL, errno, _("Failed to open file '%s'"), path); return -1; } - int len = virFileReadLimFP (fh, maxlen, buf); - fclose(fh); + int len = virFileReadLimFD(fd, maxlen, buf); + close(fd); if (len < 0) { virReportSystemError(NULL, errno, _("Failed to read file '%s'"), path); return -1; -- 1.6.2.5

2 2

[libvirt] Tentative schedule for next release
by Daniel Veillard 09 Oct '09

09 Oct '09

I'm suggesting to try to push 0.7.2 around Fri 16 Oct, we already have a number of cleanup patches in and serious improvements for ESX driver. If we go with that plan I think patches and big changes should be pushed in the tree by Fri 9, this includes among other things the AppArmor security driver. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 3

[libvirt] dynamic scsi disk attach seems to be broken in qemu(-kvm)-0.11, libvirt-0.7.1
by Dustin Kirkland 09 Oct '09

09 Oct '09

We're experiencing what we believe to be a regression in qemu-kvm-0.11 and libvirt-0.7.0, from previous versions of kvm-84 and libvirt-0.6.1: Dynamically attaching sdb (a second scsi disk) fails. (Note that I have tested this against libvirt-0.7.1 built from source as well--same results.) Here are some instructions to reproduce this issue: Create some small auxiliary storage to dynamically add to a running VM. (host) $ dd if=/dev/zero of=/tmp/foo bs=1M count=64 Boot a vm in under qemu-kvm-0.11.0 like this: (host) $ kvm -drive file=server.img,if=scsi,boot=on -boot c To reproduce the problem, you must use if=scsi. This problem manifests itself when subsequent scsi storage devices are added. If the device is the first scsi device on the system, then it would succeed. It's the 2nd scsi device that causes the problem. Once the system is booted, login and in the vm, load this module: (vm) $ sudo modprobe acpiphp Check that the acpiphp slots are loaded in dmesg. And note that there is only one /dev/sd[a-z] device. Now, drop to the qemu console with ctrl-alt-2, and add the storage: (qemu) pci_add auto storage file=/tmp/foo,if=scsi OK domain 0, bus 0, slot 6, function 0 Switch back to the vm linux shell with ctrl-alt-1, and look at the dmesg output. (vm) $ dmesg | tail -n 12 [ 44.033397] pci 0000:00:06.0: reg 10 io port: [0x00-0xff] [ 44.033443] pci 0000:00:06.0: reg 14 32bit mmio: [0x000000-0x0003ff] [ 44.033486] pci 0000:00:06.0: reg 18 32bit mmio: [0x000000-0x001fff] [ 44.033899] pci 0000:00:02.0: BAR 6: bogus alignment [0x0-0x0] flags 0x2 [ 44.033975] decode_hpp: Could not get hotplug parameters. Use defaults [ 44.042277] sym53c8xx 0000:00:06.0: enabling device (0000 -> 0003) [ 44.043230] ACPI: PCI Interrupt Link [LNKB] enabled at IRQ 11 [ 44.043247] sym53c8xx 0000:00:06.0: PCI INT A -> Link[LNKB] -> GSI 11 (level, high) -> IRQ 11 [ 44.045237] sym1: <895a> rev 0x0 at pci 0000:00:06.0 irq 11 [ 44.047586] sym1: No NVRAM, ID 7, Fast-40, LVD, parity checking [ 44.055399] sym1: SCSI BUS has been reset. [ 44.063329] scsi3 : sym-2.2.3 More importantly, note that no /dev/sd[b-z] device shows up. If you now drop to the qemu console with ctrl-alt-2, and do this: (qemu) pci_add auto storage file=/tmp/foo,if=virtio OK domain 0, bus 0, slot 7, function 0 Going back to the vm with ctrl-alt-1, you can now see a new /dev/vda device registered. Now, all of that said, it is actually possible to hot-add a second scsi device. However, as far as I can tell, this method is not yet supported by libvirt. It looks to me that with modern qemu, you have to do it this way: Drop to a qemu console with ctrl-alt-2. Get the address of the current scsi bus: (qemu) info pci Look for "SCSI Controller". In my case, it's on Bus 0, device 4, function 0 Now instead of pci_add, use drive_add (qemu) drive_add 0:4 file=/tmp/foo,if=scsi OK bus 0, unit 1 This is not ideal, however, as I tried re-scan the scsi bus with rescan-scsi-bus.sh (from scsitools) without luck. It did not pick up the new sdb device. However, I did reboot the vm, and voila, I now have /dev/sdb. I would like some advice on how to proceed with this bug, and where the solution lies...in qemu or in libvirt. Ultimately, I would like the behavior we had in our previous release with kvm-84 and libvirt-0.6.1, where we could dynamically add scsi devices without a problem, using: pci_add 1 storage file=/tmp/foo,if=scsi Can anyone else reproduce this? Is this considered a regression by anyone else? Where should I look to solve this, in libvirt, or in qemu? -- :-Dustin

3 4

[libvirt] Help needed: simple python api to obtain events
by Dan Kenigsberg 09 Oct '09

09 Oct '09

Would someone help me have a shrink-wrapped solution for obtaining libvirt events in python? I'm looking for a python module with no API calls (or maybe just a few). I'd like to register a callback functions with virConnect.domainEventRegister(), and want them to be called in the context of a specialized (automatically-created?) thread. I don't mind if the module has two API calls instead: register() and startPollThread(). I learned that it takes me far too long to do it on my own, and that just modifying events-python.py (attached) reaches miserable outcomes ("libvir: Domain error : invalid domain pointer in virDomainFree", and deadlocked python: #0 0x00000037fe40d174 in __lll_lock_wait () from /lib64/libpthread.so.0 #1 0x00000037fe408aca in _L_lock_1034 () from /lib64/libpthread.so.0 #2 0x00000037fe40898c in pthread_mutex_lock () from /lib64/libpthread.so.0 #3 0x0000003cce04db84 in remoteDomainEventFired () #4 0x00002b4cc96e1612 in libvirt_virEventInvokeHandleCallback () ) Help most welcome, Dan.

4 10

[libvirt] [PATCH 0/9] Support new peer-2-peer migration mode & public API
by Daniel P. Berrange 09 Oct '09

09 Oct '09

This series is an update of http://www.redhat.com/archives/libvir-list/2009-September/msg00540.html There isn't as much functional change here as you might presume from the number of patches. Since Chris' tunnelled migration code is added, I thought it better to do alot of small refactoring steps rather than munge it all in one patch. One particular notable change since last time is that the new virDomainMigrateToURI method does *not* mandate use of the new "VIR_MIGRATE_PEER2PEER" flag anymore. I will now explain why... Traditionally migration has been a 3 step process 1. client -> dest (prepare) 2. client -> source (perform) 3. client -> dest (finish) This is why virDomainMigrate requires the extra 'dconn' parameter from the client app With the tunnelled migration / peer-to-peer migration, the extra 'dconn' parameter is no longer required from the client, because the source libvirt driver directly opens a connection to the destination libvirtd daemon on the remote host. ie libvirtd is talking to another libvirtd peer-2-peer. In considering the implementation of the PEER_TO_PEER flag for the Xen and VMWare drivers though I realized that even this is not technically neccessary, and in fact detrimental to use of libvirt for migrating with these drivers. Both Xen and VMWare have their own management daemon which is perfectly capable of handling the entire process without any need for libvirtd to be involved on the destination host. libvirt need only initiate the migration op on the source host. If you look at the impl of Prepare/Finish ops in the Xen driver this should be obvious, since they are pretty much no-ops Thus we in fact have 3 possible migration processes * Traditional libvirt client managed client -> dest (prepare) client -> source (perform) client -> dest (finish) * New peer-to-peer migration (on which tunnelled mig is built) client -> source (perform) source -> dest (open) source -> dest (prepareTunnel) source -> dest (streamSend) source -> dest (finish) source -> dest (close) * New direct migration (hypervisor managed) client -> source (perform) source -> dest (whatever the HV's native migration does) In terms of libvirt public APIs this works out as follows * Traditional libvirt client managed A hypervisor specific URI style, passed to virDomainMigrate() with no flags set * New peer-to-peer migration (on which tunnelled mig is built) The standard libvirt URI style, passed to virDomainMigrate() or virDomainMigrateToURI() with VIR_DOMAIN_MIGRATE_PEER2PEER flag set. The virDomainMigrateToURI() method is better since it avoids the unneccessary burden on the client of connecting to the remote libvirtd. * New direct migration (hypervisor managed) A hypervisor specific URI style, passed to virDomainMigrateToURI() with no flags set The QEMU driver can't support the latter, since it has no native management daemon - it requires libvirtd's help. Xen and VMWare, and probably VirtualBox can support this just fine. This is good for flexibility of usage in terms of authentication too, since each of these 3 modes has different characteristics * Traditional libvirt client managed client -> source libvirtd auth client -> dest libvirtd auth possible hypervisor -> hypervisor auth * New peer-to-peer migration, without tunnelling client -> source libvirtd auth source -> dest libvirtd auth possible hypervisor -> hypervisor auth * New peer-to-peer migration, without tunnelling client -> source libvirtd auth source -> dest libvirtd auth * New direct migration (hypervisor managed) client -> source libvirtd auth possible hypervisor -> hypervisor auth NB, 'client -> source libvirtd auth' is essentially no-op if initiating migration from the source host directly in all these cases. With with the combination of the two virDomainMigrate and virDomainMigrateToURI methods, and the VIR_MIGRATE_PEER2PEER flag, I believe we're well placed to cover all the main deployment/auth scenarios for migration in all hypervisors without imposing extra auth burden ourselves as we did in the past Further things that need to be done though: - Xen could easily support PEER2PEER and TUNNELLED flags - Add a VIR_MIGRATE_SECURE flag, to indicate that the app only wants the migration to be performed if the HV can guarentee the data channel is secure. - Document the scenarios I just outlined and give some mre guidance to app developers/administrators about the tradeoff between each scenario Regards, Daniel

3 15

[libvirt] [PATCH] Remove some auto-generated files
by Daniel P. Berrange 09 Oct '09

09 Oct '09

The original mail i tried to send was far too large, so here's it edited to remove the boring bits. Daniel > From: "Daniel P. Berrange" <berrange(a)redhat.com> > To: libvir-list(a)redhat.com > Cc: "Daniel P. Berrange" <berrange(a)redhat.com> > Subject: [PATCH] Remove some auto-generated files > Date: Thu, 8 Oct 2009 17:03:07 +0100 > > Removes some auto-generated files still under version control. > It also moves the rule for generating NEWS into the Makefile.am > that's in the same directory as the output file to avoid confusion > > * docs/libvirt-api.xml, docs/libvirt-refs.xml, NEWS: Remove > auto-generated files from source control > * Makefile.am: Add rule for generating NEWS file > * docs/Makefile.am: Remove rule for generating NEWS file > --- > Makefile.am | 9 + > NEWS | 1620 ---------- > docs/Makefile.am | 10 +- > docs/libvirt-api.xml | 2986 ------------------ > docs/libvirt-refs.xml | 8313 ------------------------------------------------- > 5 files changed, 10 insertions(+), 12928 deletions(-) > delete mode 100644 NEWS > delete mode 100644 docs/libvirt-api.xml > delete mode 100644 docs/libvirt-refs.xml > > diff --git a/Makefile.am b/Makefile.am > index add7e2a..977ad0c 100644 > --- a/Makefile.am > +++ b/Makefile.am > @@ -31,6 +31,15 @@ EXTRA_DIST = \ > pkgconfigdir = $(libdir)/pkgconfig > pkgconfig_DATA = libvirt.pc > > +NEWS: $(top_srcdir)/docs/news.xsl $(top_srcdir)/docs/news.html.in > + -@(if [ -x $(XSLTPROC) ] ; then \ > + $(XSLTPROC) --nonet $(top_srcdir)/docs/news.xsl \ > + $(top_srcdir)/docs/news.html.in \ > + | perl -0777 -pe 's/\n\n+$$/\n/' \ > + | perl -pe 's/[ \t]+$$//' \ > + > $@-t && mv $@-t $@ ; fi ); > + > + > rpm: clean > @(unset CDPATH ; $(MAKE) dist && rpmbuild -ta $(distdir).tar.gz) > > diff --git a/docs/Makefile.am b/docs/Makefile.am > index 8832b72..0c206c3 100644 > --- a/docs/Makefile.am > +++ b/docs/Makefile.am > @@ -80,7 +80,7 @@ EXTRA_DIST= \ > > CLEANFILES = $(dot_html) $(apihtml) $(devhelphtml) > > -all: web $(top_builddir)/NEWS > +all: web > > api: libvirt-api.xml libvirt-refs.xml > > @@ -133,14 +133,6 @@ libvirt-api.xml libvirt-refs.xml: apibuild.py \ > $(srcdir)/../src/libvirt.c $(srcdir)/../src/util/virterror.c > -(./apibuild.py) > > -$(top_builddir)/NEWS: $(top_srcdir)/docs/news.xsl $(top_srcdir)/docs/news.html.in > - -@(if [ -x $(XSLTPROC) ] ; then \ > - $(XSLTPROC) --nonet $(top_srcdir)/docs/news.xsl \ > - $(top_srcdir)/docs/news.html.in \ > - | perl -0777 -pe 's/\n\n+$$/\n/' \ > - | perl -pe 's/[ \t]+$$//' \ > - > $@-t && mv $@-t $@ ; fi ); > - > clean-local: > rm -f *~ *.bak *.hierarchy *.signals *-unused.txt > ----- End forwarded message ----- -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] Xen: Domain.migrate() causes invalid listDomains() output
by Thomas Treutner 09 Oct '09

09 Oct '09

Hi, I'm using Xen 3.2 (Debian Lenny) and libvirt 0.7.1. When I use Domain.migrate() from libvirt-java 0.3.0, a following Connect.listDomains() gives wrong DomainIDs (they seem to be the ones before the migration) and therefore DomainLookupByID() fails because of a nonexistent ID. Restarting libvirtd and/or xend doesn't help. xm list etc. output is still correct, but anything that uses libvirt (like virt-top) is having troubles. The migration itself is working without glitches, the domains are up&running and responsive. But at this point, libvirt is completely mistaken about the running domains and their IDs. There are some quite similar bug reports and mails here, but they are rather old. Shall I provide more debug output? tia&kr, thomas

2 4

[libvirt] libvirt/kvm config for routed networking
by William Wagner 08 Oct '09

08 Oct '09

Hello, I hope this is the right list I should post to as my question is not development related. I am trying to setup a kvm/libvirt VM on my host (Ubuntu jaunty). My host has a public static IP and my VM also has a public static IP. Unfortunately I can not use bridged networking as the hosting provider has configured their switch to only accept packets from the MAC address of the host. I want to be able to setup my VM so it has the public static IP and it appears to be directly connected to the net. I believe I can do this with routed networking. I have created a new routed network: <network> <name>routed-net</name> <bridge name="routed%d" /> <forward mode="route" dev="eth0"/> <ip address="10.255.255.2" netmask="255.255.255.255"> </ip> </network> and in my vm's config I have: <interface type='user'> <source network='routed-net'/> <mac address='54:52:00:47:a8:38'/> <model type='virtio'/> </interface> Will this mean that the VM is placed on the routed network? Then I just need to add suitable routing rules on the host and everything should work? Is there a way to get libvirt to add the rules automatically when the VM starts. I have previously used Xen where you are able to specify what the IP address of the VM is and entries are automatically added to iptables. Is there similar syntax for libvirt and if so what is it? If not how do you recommend adding the routing rules? Thanks. Will. -- ------------------------------------------------------------------------ Will Wagner will_wagner(a)carallon.com Development Manager Office Tel: +44 (0)20 7371 2032 Carallon Ltd, Studio G20, Shepherds Building, Rockley Rd, London W14 0DA ------------------------------------------------------------------------

2 1

[libvirt] [PATCH 0/2] Take 2 - live interface info + IPv6 interface config
by Laine Stump 08 Oct '09

08 Oct '09

This is similar to the (now deprecated) patches I sent on 9/28 and 10/04, but without the unpopular "source" attribute in the ip address. You can successfully build libvirt if you have the latest release of netcf installed, but you'll need to get the netcf source, apply the patches in the thread listed below, and build/install netcf to get correct/proper output. https://fedorahosted.org/pipermail/netcf-devel/2009-October/000289.html

3 10

[libvirt] [PATCH 0/7] Series short description
by Amy Griffis 08 Oct '09

08 Oct '09

The following patches fix logging for the lxc controller. The lxc controller makes use of the libvirt logging module, but because it doesn't re-initialize the logging configuration the messages are currently going nowhere. There are a couple of options in fixing this. The lxc controller could use libvirtd's logging configuration exactly, or it adopt a more independent mode of logging to the container log file. After talking with Daniel Veillard a while back, it seemed clear that both behaviors are desirable. I resolved this by adding a lxc.conf file for the lxc driver, with a boolean variable to toggle between modes. I hope this looks good to both libvirt and container folks. libvirt.spec.in | 16 ++++++ src/Makefile.am | 25 +++++++-- src/libvirt_private.syms | 3 + src/lxc/libvirtd_lxc.aug | 30 +++++++++++ src/lxc/lxc.conf | 13 +++++ src/lxc/lxc_conf.c | 24 +++++++++ src/lxc/lxc_conf.h | 1 src/lxc/lxc_controller.c | 3 + src/lxc/lxc_driver.c | 86 ++++++++++++++++++++++++++++---- src/lxc/test_libvirtd_lxc.aug | 31 ++++++++++++ src/util/logging.c | 109 ++++++++++++++++++++++++++++++++++++++--- src/util/logging.h | 14 +++++ src/util/util.c | 49 ++++++++++++++++++ src/util/util.h | 3 + 14 files changed, 379 insertions(+), 28 deletions(-) create mode 100644 src/lxc/libvirtd_lxc.aug create mode 100644 src/lxc/lxc.conf create mode 100644 src/lxc/test_libvirtd_lxc.aug Regards, Amy

6 35

[libvirt] Resubmission #2: [PATCH 0/3] sVirt AppArmor security driver
by Jamie Strandboge 08 Oct '09

08 Oct '09

Resubmitting again based on feedback from this list. Notably, the driver has been reworked to use: xml = virDomainDefFormat(conn, vm->def, ... Then the helper program uses: ctl->def = virDomainDefParseString(... xmlStr ...); This simplifies adding new files. Part of this process involved adding support for kernel, initrd, loader, serial, and console. I have commented out code for hostdev (and pci would be easy enough), but it requires a patch to to move the _usbDevice struct fully into hostusb.h. I figure once the AppArmor driver is in, I can submit a patch for that. This iteration also properly works with attach/detach. While I use virDomainDefFormat() primarily, this does not work with attach because the definition is not updated with the information yet (neither def nor newDef), so I pass the disk file on the command line. For now this is ok since SetSecurityImageLabel currently only deals with a single file, but this will need to be adjusted depending on how the storage backend patch works out. I thought about creating a new def and inserting the disk into it and sending it off to virt-aa-helper, but I couldn't find an easy way to get an accurate virCapsPtr in security_apparmor.c. I had a similar problem with virt-aa-helper, but there I just created a fake virCapsPtr (which defaults to 'hvm', which was a compromise I was willing to make at this time since sVirt only supports qemu right now). The issues with virCapsPtr aside, the code is overall cleaner and much more easily extendable, so thanks for the excellent feedback. :) The patches are against up to date trunk, compiles with no warnings, 'make check' passes for the tests I added, and the code passes syntax-check. With the exception of the three calls to strncpy I had to convert to virStr*, this is the code that is currently in Ubuntu 9.10 (0.7.0-1ubuntu8). Among other things, that means 9.10 now has your preferred licensing. [PATCH 1] patch_1_reenable-nonfile-labels.patch (Updated based on prior feedback): When James Morris originally submitted his sVirt patches (as seen in libvirt 0.6.1), he did not require on disk labelling for virSecurityDomainRestoreImageLabel. A later commit[2] changed this behavior to assume on disk labelling, which halts implementations for path-based MAC systems such as AppArmor and TOMOYO where vm->def->seclabel is required to obtain the label. This patch simply adds the 'virDomainObjPtr vm' argument back to *RestoreImageLabel. [PATCH 2] patch_2_apparmor_driver.patch (updated and merged into one patch based on prior feedback): - Updates src/security.c for AppArmor - Adds security_apparmor.c, security_apparmor.h, virt-aa-helper.c and updates po/POTFILES.in. virt-aa-helper.c is a new binary which is used exclusively by the AppArmor security driver to manipulate AppArmor. - Adds tests for virt-aa-helper and the security driver. secaatest.c is identical to seclabeltest.c except it initializes the 'apparmor' driver instead of 'selinux'. These tests are integrated into 'make check' and pass. - Updates configure.in, Makefile.am, src/Makefile.am, tests/Makefile.am, and tools/Makefile.am for AppArmor. It is based on and should operate the same as the SELinux configuration. [PATCH 3] patch_3_docs.patch (Updated based on prior feedback): Updates docs/drvqemu.html.in for AppArmor and adds profile examples to examples/apparmor. Updated based on prior feedback. Jamie On Fri, 04 Sep 2009, Jamie Strandboge wrote: > This patch series implements the AppArmor security driver for sVirt. > This implementation was developed for the Ubuntu AppArmorLibvirtProfile > specification[1], but is general enough for any AppArmor deployment > (such as Ubuntu, *SUSE and Mandriva). > > This patch has seen quite a bit of real world testing in Ubuntu 9.10 > (our development release) in our 0.7.0-1ubuntu3 package. I did make a > few small changes after going through HACKING, but mostly I got the > tests going and added documentation. > > DESIGN > ------ > When a virtual machine is started, determine if a profile is currently > defined for the machine, and use it if available. If not, generate a new > profile for the machine based on a template, which is by default a very > restrictive profile allowing access to disk files, and anything else > needed to run, such as the pid, monitor and log files. > > Virtual machines should have a unique profile specific to that machine. > To ensure uniqueness, the profile name will be derived from the UUID of > the virtual machine. These profiles should be configurable, either by > adjusting the profile template for new machines, creating/modifying the > VM profile directly or through the use of AppArmor abstractions. This > will allow for administrators to fine-tune confinement for individual > machines if desired. > > If enabled at compile time, the sVirt security model will be activated > if AppArmor is available on the host OS and a profile for the libvirtd > daemon is loaded when libvirtd is started. > > libvirtd should not be allowed to create arbitrary profiles or modify > profiles directly, so as to not allow libvirtd to potentially (ie via a > security bug in libvirtd itself) bootstrap out of AppArmor confinement. > > Because root privileges are needed to manipulate AppArmor profiles, > qemu:///session will not be supported at this time, but the > implementation must allow for a confined libvirtd with qemu:///session > guests running unconfined. This can be revisited when AppArmor supports > per-user profiles. > > Please see the specification[1] for more details. -- Jamie Strandboge | http://www.canonical.com

3 12

[libvirt] [PATCH] cgroup -Werror fix
by Cole Robinson 08 Oct '09

08 Oct '09

Building with -Werror, I'm getting this error: make[3]: *** [libvirt_lxc-cgroup.o] Error 1 make[3]: *** Waiting for unfinished jobs.... cc1: warnings being treated as errors util/cgroup.c: In function 'virCgroupGetMemoryUsage': util/cgroup.c:743: error: passing argument 4 of 'virCgroupGetValueU64' from incompatible pointer type util/cgroup.c:381: note: expected 'long long unsigned int *' but argument is of type 'uint64_t *' Patch below fixes it, though I'm not familiar with the code. - Cole diff --git a/src/util/cgroup.c b/src/util/cgroup.c index 77f19a5..f728a2f 100644 --- a/src/util/cgroup.c +++ b/src/util/cgroup.c @@ -736,7 +736,7 @@ */ int virCgroupGetMemoryUsage(virCgroupPtr group, unsigned long *kb) { - uint64_t usage_in_bytes; + long long unsigned int usage_in_bytes; int ret; ret = virCgroupGetValueU64(group, VIR_CGROUP_CONTROLLER_MEMORY,

2 1

[libvirt] [PATCH] in vbox driver, interface type bridge should really be type ethernet
by Florian Vichot 08 Oct '09

08 Oct '09

If I understand correctly, the "bridge" type for an interface means libvirt is in charge of creating a tun device and adding it to the specified bridge in the <source bridge=".."> attribute. This is not what the (badly named IHMO) "Bridged Networking" mode in vbox does: all it does is read and write its packets on the specified interface. Which is what a type "ethernet" interface does in my opinion. I joined a quick patch for that, Pritesh could you check it ? Also, I was about to patch to add support for the proper "bridge" mode in the vbox driver (because it's what I need in the end), but I see no way of storing the fact that it's not just an "ethernet" interface using the vbox API, so that the config would persist over libvirtd restarts. There is always the possibility of naming the interface a special way like "br1_myTap", and assume any type "ethernet" interface with a device containing a "br1_" prefix would actually be a type "bridge" interface with an associated "vbox_br1" bridge. Would that be acceptable, or is it too hackish ? BTW, just to make sure, I can use pretty much anything as a <target> ie. ethN, panN etc in a type "ethernet" interface, right ? If <target> is specified, it's not going to create a tun and use that ? Thanks, Florian

4 6

[libvirt] [PATCH 0/6] correctly migrate paused VMs (and a bunch of bugfixes)
by Paolo Bonzini 07 Oct '09

07 Oct '09

Currently, the QEMU driver tries to migrate paused VMs, but the effect is that the migrated virtual machine will always run on the destination. Furthermore, the state is erroneously stored as paused, so that the driver is left confused and it is not possible to re-pause the VM without doing a (useless except to make the libvirtd state consistent) "virsh resume" first. While pausing is not very used, this feature is still nice-to-have in case for example you're migrating all VMs in emergency, or (when QEMU will have asynchronous notifications) if you want to run QEMU with the werror=stop option (*). This set of patches is structured as follows: 1) patches 1 and 2 fix two bugs that were introduced by the recent reorganization. 2) patch 3 fixes a bug that happens to be in the paths touched later on; 3) patch 4 adds the possibility to migrate a VM and leave it suspended on the destination VM. I added the feature for debugging and thought I might as well submit it for inclusion. Especially if the last two patches turn out to be a no-go (at least in their current form), this patch at least provides the infrastructure and a way to test it. 4) patches 5 and 6 actually fix the bug. Unfortunately, this requires a change to the RPC protocol. I'm not sure whether it is required the libvirtd's on both sides of the migration to have the same version, or rather if this is a deal breaker. The change to the protocol is in patch 5; the bug fix is in patch 6. Please review and ack (or, for the last two, nack). Paolo (*) While the state of the migrated VM is currently fetched at the beginning of migration, the communication happens at the end of it (from Perform to Finish), so this is ready for being improved in the future once QEMU provides asynchronous notifications.

5 26

[libvirt] [PATCH] Misc win32 build fixes
by Daniel P. Berrange 07 Oct '09

07 Oct '09

* daemon/Makefile.am: Fix missing sasl rule * src/datatypes.c: Add unistd.h to avoid gnulib bug * src/util/cgroup.c: Disable mntent if not available --- daemon/Makefile.am | 3 ++ src/datatypes.c | 1 + src/util/cgroup.c | 66 ++++++++++++++++++++++++++++++++++++++++++--------- 3 files changed, 58 insertions(+), 12 deletions(-) diff --git a/daemon/Makefile.am b/daemon/Makefile.am index 26347d6..84aab04 100644 --- a/daemon/Makefile.am +++ b/daemon/Makefile.am @@ -243,6 +243,9 @@ install-data-sasl: uninstall-data-sasl: rm -f $(DESTDIR)$(sysconfdir)/sasl2/libvirt.conf rmdir $(DESTDIR)$(sysconfdir)/sasl2/ +else +install-data-sasl: +uninstall-data-sasl: endif diff --git a/src/datatypes.c b/src/datatypes.c index 6741b9e..89ad309 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -20,6 +20,7 @@ */ #include <config.h> +#include <unistd.h> #include "datatypes.h" #include "virterror_internal.h" diff --git a/src/util/cgroup.c b/src/util/cgroup.c index 2e646fd..66297b9 100644 --- a/src/util/cgroup.c +++ b/src/util/cgroup.c @@ -13,7 +13,9 @@ #include <stdio.h> #include <stdint.h> #include <inttypes.h> +#ifdef HAVE_MNTENT_H #include <mntent.h> +#endif #include <fcntl.h> #include <string.h> #include <errno.h> @@ -68,7 +70,7 @@ void virCgroupFree(virCgroupPtr *group) VIR_FREE(*group); } - +#ifdef HAVE_MNTENT_H /* * Process /proc/mounts figuring out what controllers are * mounted and where @@ -233,6 +235,7 @@ static int virCgroupDetect(virCgroupPtr group) return rc; } +#endif static int virCgroupPathOfController(virCgroupPtr group, @@ -317,12 +320,12 @@ static int virCgroupGetValueStr(virCgroupPtr group, static int virCgroupSetValueU64(virCgroupPtr group, int controller, const char *key, - uint64_t value) + unsigned long long int value) { char *strval = NULL; int rc; - if (virAsprintf(&strval, "%" PRIu64, value) == -1) + if (virAsprintf(&strval, "%llu", value) == -1) return -ENOMEM; rc = virCgroupSetValueStr(group, controller, key, strval); @@ -339,12 +342,12 @@ static int virCgroupSetValueU64(virCgroupPtr group, static int virCgroupSetValueI64(virCgroupPtr group, int controller, const char *key, - int64_t value) + long long int value) { char *strval = NULL; int rc; - if (virAsprintf(&strval, "%" PRIi64, value) == -1) + if (virAsprintf(&strval, "%lld", value) == -1) return -ENOMEM; rc = virCgroupSetValueStr(group, controller, key, strval); @@ -357,7 +360,7 @@ static int virCgroupSetValueI64(virCgroupPtr group, static int virCgroupGetValueI64(virCgroupPtr group, int controller, const char *key, - int64_t *value) + long long int *value) { char *strval = NULL; int rc = 0; @@ -378,7 +381,7 @@ out: static int virCgroupGetValueU64(virCgroupPtr group, int controller, const char *key, - uint64_t *value) + unsigned long long int *value) { char *strval = NULL; int rc = 0; @@ -387,7 +390,7 @@ static int virCgroupGetValueU64(virCgroupPtr group, if (rc != 0) goto out; - if (sscanf(strval, "%" SCNu64, value) != 1) + if (virStrToLong_ull(strval, NULL, 10, value) < 0) rc = -EINVAL; out: VIR_FREE(strval); @@ -396,6 +399,7 @@ out: } +#ifdef HAVE_MNTENT_H static int virCgroupCpuSetInherit(virCgroupPtr parent, virCgroupPtr group) { int i; @@ -546,7 +550,7 @@ cleanup: virCgroupFree(&rootgrp); return rc; } - +#endif /** * virCgroupRemove: @@ -617,6 +621,7 @@ int virCgroupAddTask(virCgroupPtr group, pid_t pid) * * Returns 0 on success */ +#ifdef HAVE_MNTENT_H int virCgroupForDriver(const char *name, virCgroupPtr *group, int privileged, @@ -650,6 +655,16 @@ out: return rc; } +#else +int virCgroupForDriver(const char *name ATTRIBUTE_UNUSED, + virCgroupPtr *group ATTRIBUTE_UNUSED, + int privileged ATTRIBUTE_UNUSED, + int create ATTRIBUTE_UNUSED) +{ + /* Claim no support */ + return -ENXIO; +} +#endif /** @@ -661,6 +676,7 @@ out: * * Returns 0 on success */ +#ifdef HAVE_MNTENT_H int virCgroupForDomain(virCgroupPtr driver, const char *name, virCgroupPtr *group, @@ -684,6 +700,15 @@ int virCgroupForDomain(virCgroupPtr driver, return rc; } +#else +int virCgroupForDomain(virCgroupPtr driver ATTRIBUTE_UNUSED, + const char *name ATTRIBUTE_UNUSED, + virCgroupPtr *group ATTRIBUTE_UNUSED, + int create ATTRIBUTE_UNUSED) +{ + return -ENXIO; +} +#endif /** * virCgroupSetMemory: @@ -786,6 +811,7 @@ int virCgroupAllowDeviceMajor(virCgroupPtr group, char type, int major) * * Returns: 0 on success */ +#if defined(major) && defined(minor) int virCgroupAllowDevicePath(virCgroupPtr group, const char *path) { struct stat sb; @@ -801,6 +827,14 @@ int virCgroupAllowDevicePath(virCgroupPtr group, const char *path) major(sb.st_rdev), minor(sb.st_rdev)); } +#else +int virCgroupAllowDevicePath(virCgroupPtr group ATTRIBUTE_UNUSED, + const char *path ATTRIBUTE_UNUSED) +{ + return -ENOSYS; +} +#endif + /** * virCgroupDenyDevice: @@ -861,6 +895,7 @@ int virCgroupDenyDeviceMajor(virCgroupPtr group, char type, int major) return rc; } +#if defined(major) && defined(minor) int virCgroupDenyDevicePath(virCgroupPtr group, const char *path) { struct stat sb; @@ -876,26 +911,33 @@ int virCgroupDenyDevicePath(virCgroupPtr group, const char *path) major(sb.st_rdev), minor(sb.st_rdev)); } +#else +int virCgroupDenyDevicePath(virCgroupPtr group ATTRIBUTE_UNUSED, + const char *path ATTRIBUTE_UNUSED) +{ + return -ENOSYS; +} +#endif int virCgroupSetCpuShares(virCgroupPtr group, unsigned long long shares) { return virCgroupSetValueU64(group, VIR_CGROUP_CONTROLLER_CPU, - "cpu.shares", (uint64_t)shares); + "cpu.shares", shares); } int virCgroupGetCpuShares(virCgroupPtr group, unsigned long long *shares) { return virCgroupGetValueU64(group, VIR_CGROUP_CONTROLLER_CPU, - "cpu.shares", (uint64_t *)shares); + "cpu.shares", shares); } int virCgroupGetCpuacctUsage(virCgroupPtr group, unsigned long long *usage) { return virCgroupGetValueU64(group, VIR_CGROUP_CONTROLLER_CPUACCT, - "cpuacct.usage", (uint64_t *)usage); + "cpuacct.usage", usage); } int virCgroupSetFreezerState(virCgroupPtr group, const char *state) -- 1.6.2.5

2 3

[libvirt] [PATCH] Fix handling the return value of qemuMonitorSetBalloon
by Ryota Ozaki 07 Oct '09

07 Oct '09

Hi, The current code doesn't correctly handle the positive return value of qemuMonitorSetBalloon, i.e., the success case. The return value is directly passed to the return of qemudDomainSetMemory which treats a positive value as a failure. This patch fixes the defect. Thanks, ozaki-r >From f94b4c920ea30ae4a7f55fda343b4fff3bba040f Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Mon, 5 Oct 2009 03:14:58 +0900 Subject: [PATCH] Fix handling the return value of qemuMonitorSetBalloon * src/qemu/qemu_driver.c: The positive return value of qemuMonitorSetBalloon should be handled as a success --- src/qemu/qemu_driver.c | 13 ++++++++----- 1 files changed, 8 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 95e672b..8cf9002 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3020,17 +3020,20 @@ static int qemudDomainSetMemory(virDomainPtr dom, unsigned long newmem) { } if (virDomainIsActive(vm)) { - ret = qemuMonitorSetBalloon(vm, newmem); - /* Turn lack of balloon support into a fatal error */ - if (ret == 0) { + int r = qemuMonitorSetBalloon(vm, newmem); + if (r < 0) + goto cleanup; + + /* Lack of balloon support is a fatal error */ + if (r == 0) { qemudReportError(dom->conn, dom, NULL, VIR_ERR_NO_SUPPORT, "%s", _("cannot set memory of an active domain")); - ret = -1; + goto cleanup; } } else { vm->def->memory = newmem; - ret = 0; } + ret = 0; cleanup: if (vm) -- 1.6.5.rc2

3 2

[libvirt] [PATCH] lxc: implement missing APIs to control the amount of memory
by Ryota Ozaki 07 Oct '09

07 Oct '09

Hi, The patch implements the missing memory control APIs for lxc, i.e., domainGetMaxMemory, domainSetMaxMemory, domainSetMemory, and improves domainGetInfo to return proper amount of used memory via cgroup. And also it includes trivial two fixes; add missing virCgroupSetMemory in libvirt_private.syms and fix invalid lxcError argument in lxcDomainGetInfo. Thanks, ozaki-r >From 530e56db9367fcb325aeaa7fb0d52a008db2f19f Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Mon, 5 Oct 2009 02:07:04 +0900 Subject: [PATCH] lxc: implement missing APIs to control the amount of memory * src/libvirt_private.syms: Export virCgroupGetMemoryUsage and add missing virCgroupSetMemory * src/lxc/lxc_driver.c: Implement missing memory functions * src/util/cgroup.c, src/util/cgroup.h: Add the function to get used memory --- src/libvirt_private.syms | 2 + src/lxc/lxc_driver.c | 122 ++++++++++++++++++++++++++++++++++++++++++++-- src/util/cgroup.c | 20 ++++++++ src/util/cgroup.h | 1 + 4 files changed, 140 insertions(+), 5 deletions(-) diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 49bbf96..3be3724 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -53,6 +53,8 @@ virCgroupForDriver; virCgroupRemove; virCgroupFree; virCgroupAddTask; +virCgroupSetMemory; +virCgroupGetMemoryUsage; virCgroupSetCpuShares; virCgroupGetCpuShares; virCgroupDenyDevicePath; diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 5fb4105..8ec4ae4 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -450,6 +450,7 @@ static int lxcDomainGetInfo(virDomainPtr dom, if (!virDomainIsActive(vm) || driver->cgroup == NULL) { info->cpuTime = 0; + info->memory = vm->def->memory; } else { if (virCgroupForDomain(driver->cgroup, vm->def->name, &cgroup, 0) != 0) { lxcError(dom->conn, dom, VIR_ERR_INTERNAL_ERROR, @@ -458,13 +459,18 @@ static int lxcDomainGetInfo(virDomainPtr dom, } if (virCgroupGetCpuacctUsage(cgroup, &(info->cpuTime)) < 0) { - lxcError(dom->conn, dom, VIR_ERR_OPERATION_FAILED, ("cannot read cputime for domain")); + lxcError(dom->conn, dom, VIR_ERR_OPERATION_FAILED, + "%s", _("cannot read cputime for domain")); + goto cleanup; + } + if (virCgroupGetMemoryUsage(cgroup, &(info->memory)) < 0) { + lxcError(dom->conn, dom, VIR_ERR_OPERATION_FAILED, + "%s", _("cannot read memory usage for domain")); goto cleanup; } } info->maxMem = vm->def->maxmem; - info->memory = vm->def->memory; info->nrVirtCpu = 1; ret = 0; @@ -501,6 +507,112 @@ cleanup: return ret; } +/* Returns max memory in kb, 0 if error */ +static unsigned long lxcDomainGetMaxMemory(virDomainPtr dom) { + lxc_driver_t *driver = dom->conn->privateData; + virDomainObjPtr vm; + unsigned long ret = 0; + + lxcDriverLock(driver); + vm = virDomainFindByUUID(&driver->domains, dom->uuid); + lxcDriverUnlock(driver); + + if (!vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(dom->uuid, uuidstr); + lxcError(dom->conn, dom, VIR_ERR_NO_DOMAIN, + _("no domain with matching uuid '%s'"), uuidstr); + goto cleanup; + } + + ret = vm->def->maxmem; + +cleanup: + if (vm) + virDomainObjUnlock(vm); + return ret; +} + +static int lxcDomainSetMaxMemory(virDomainPtr dom, unsigned long newmax) { + lxc_driver_t *driver = dom->conn->privateData; + virDomainObjPtr vm; + int ret = -1; + + lxcDriverLock(driver); + vm = virDomainFindByUUID(&driver->domains, dom->uuid); + lxcDriverUnlock(driver); + + if (!vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(dom->uuid, uuidstr); + lxcError(dom->conn, dom, VIR_ERR_NO_DOMAIN, + _("no domain with matching uuid '%s'"), uuidstr); + goto cleanup; + } + + if (newmax < vm->def->memory) { + lxcError(dom->conn, dom, VIR_ERR_INVALID_ARG, + "%s", _("cannot set max memory lower than current memory")); + goto cleanup; + } + + vm->def->maxmem = newmax; + ret = 0; + +cleanup: + if (vm) + virDomainObjUnlock(vm); + return ret; +} + +static int lxcDomainSetMemory(virDomainPtr dom, unsigned long newmem) { + lxc_driver_t *driver = dom->conn->privateData; + virDomainObjPtr vm; + virCgroupPtr cgroup = NULL; + int ret = -1; + + lxcDriverLock(driver); + vm = virDomainFindByUUID(&driver->domains, dom->uuid); + lxcDriverUnlock(driver); + if (!vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(dom->uuid, uuidstr); + lxcError(dom->conn, dom, VIR_ERR_NO_DOMAIN, + _("no domain with matching uuid '%s'"), uuidstr); + goto cleanup; + } + + if (newmem > vm->def->maxmem) { + lxcError(dom->conn, dom, VIR_ERR_INVALID_ARG, + "%s", _("cannot set memory higher than max memory")); + goto cleanup; + } + + if (virDomainIsActive(vm)) { + if (virCgroupForDomain(driver->cgroup, vm->def->name, &cgroup, 0) != 0) { + lxcError(dom->conn, dom, VIR_ERR_INTERNAL_ERROR, + _("Unable to get cgroup for %s¥n"), vm->def->name); + goto cleanup; + } + + if (virCgroupSetMemory(cgroup, newmem) < 0) { + lxcError(dom->conn, dom, VIR_ERR_OPERATION_FAILED, + "%s", _("cannot set memory for domain")); + goto cleanup; + } + } else { + vm->def->memory = newmem; + } + ret = 0; + +cleanup: + if (vm) + virDomainObjUnlock(vm); + if (cgroup) + virCgroupFree(&cgroup); + return ret; +} + static char *lxcDomainDumpXML(virDomainPtr dom, int flags) { @@ -2103,9 +2215,9 @@ static virDriver lxcDriver = { NULL, /* domainReboot */ lxcDomainDestroy, /* domainDestroy */ lxcGetOSType, /* domainGetOSType */ - NULL, /* domainGetMaxMemory */ - NULL, /* domainSetMaxMemory */ - NULL, /* domainSetMemory */ + lxcDomainGetMaxMemory, /* domainGetMaxMemory */ + lxcDomainSetMaxMemory, /* domainSetMaxMemory */ + lxcDomainSetMemory, /* domainSetMemory */ lxcDomainGetInfo, /* domainGetInfo */ NULL, /* domainSave */ NULL, /* domainRestore */ diff --git a/src/util/cgroup.c b/src/util/cgroup.c index 2e646fd..e56d293 100644 --- a/src/util/cgroup.c +++ b/src/util/cgroup.c @@ -702,6 +702,26 @@ int virCgroupSetMemory(virCgroupPtr group, unsigned long kb) } /** + * virCgroupGetMemoryUsage: + * + * @group: The cgroup to change memory for + * @kb: Pointer to returned used memory in kilobytes + * + * Returns: 0 on success + */ +int virCgroupGetMemoryUsage(virCgroupPtr group, unsigned long *kb) +{ + uint64_t usage_in_bytes; + int ret; + ret = virCgroupGetValueU64(group, + VIR_CGROUP_CONTROLLER_MEMORY, + "memory.usage_in_bytes", &usage_in_bytes); + if (ret == 0) + *kb = (unsigned long) usage_in_bytes >> 10; + return ret; +} + +/** * virCgroupDenyAllDevices: * * @group: The cgroup to deny devices for diff --git a/src/util/cgroup.h b/src/util/cgroup.h index aba56c6..aa36632 100644 --- a/src/util/cgroup.h +++ b/src/util/cgroup.h @@ -41,6 +41,7 @@ int virCgroupForDomain(virCgroupPtr driver, int virCgroupAddTask(virCgroupPtr group, pid_t pid); int virCgroupSetMemory(virCgroupPtr group, unsigned long kb); +int virCgroupGetMemoryUsage(virCgroupPtr group, unsigned long *kb); int virCgroupDenyAllDevices(virCgroupPtr group); -- 1.6.5.rc2

3 2

[libvirt] [PATCH] Make dnsmasq run from libvirtd reading /etc/ethers
by Satoru SATOH 07 Oct '09

07 Oct '09

The following small patch makes dnsmasq run from libvirtd reading /etc/ethers and enables static ip address assignment with it. Libvirtd already supports static ip address assignment by //ip/dhcp/host element in network xml definition file, however the number of the assignments looks limited by ARG_MAX or similar parameters when dnsmasq is exec-ed. This patch implements a workaround for this issue and the number of static ip assignments becomes only limited by dnsmasq (I'm not sure the exact number but it should large enough for most cases, I guess). Signed-off-by: Satoru SATOH <satoru.satoh(a)gmail.com> --- src/network/bridge_driver.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 95bc810..725d340 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -391,6 +391,7 @@ networkBuildDnsmasqArgv(virConnectPtr conn, 1 + /* dnsmasq */ 1 + /* --strict-order */ 1 + /* --bind-interfaces */ + 1 + /* --read-ethers */ (network->def->domain?2:0) + /* --domain name */ 2 + /* --pid-file /var/run/libvirt/network/$NAME.pid */ 2 + /* --conf-file "" */ @@ -427,6 +428,7 @@ networkBuildDnsmasqArgv(virConnectPtr conn, */ APPEND_ARG(*argv, i++, "--strict-order"); APPEND_ARG(*argv, i++, "--bind-interfaces"); + APPEND_ARG(*argv, i++, "--read-ethers"); if (network->def->domain) { APPEND_ARG(*argv, i++, "--domain"); -- 1.6.2.5

2 2

[libvirt] [PATCH] Fix up "make check"
by Chris Lalancette 07 Oct '09

07 Oct '09

While running make check, I noticed that it was actually using the virsh binary from my system, in /usr/bin/virsh, and not the one that was just compiled. This is actually caused by a bug in Makefile.am, where we didn't update the PATH to include tools. While here, I also updated all of the scripts to properly define the srcdir, abs_top_srcdir, and abs_top_builddir environment variables. This is required if you want to be able to run the tests standalone (i.e. ./test instead of from make check). I've tested this on both RHEL-5 and Fedora-10 machines, and make check works on both, as does running the individual tests by hand. Signed-off-by: Chris Lalancette <clalance(a)redhat.com> --- tests/Makefile.am | 2 +- tests/cpuset | 12 ++++++++---- tests/daemon-conf | 18 +++++++++++------- tests/define-dev-segfault | 10 ++++++---- tests/int-overflow | 20 ++++++++++---------- tests/libvirtd-fail | 10 ++++++---- tests/libvirtd-pool | 10 ++++++---- tests/read-bufsiz | 15 ++++++++------- tests/read-non-seekable | 12 ++++++++---- tests/start | 10 ++++++---- tests/test_conf.sh | 14 ++++++++------ tests/undefine | 14 +++++++++----- tests/vcpupin | 12 ++++++++---- tests/virsh-all | 12 +++++++----- tests/virsh-synopsis | 11 +++++++---- 15 files changed, 109 insertions(+), 73 deletions(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index abda4ad..4caca8f 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -150,7 +150,7 @@ TESTS += nodedevxml2xmltest TESTS += interfacexml2xmltest -path_add = $$abs_top_builddir/src$(PATH_SEPARATOR)$$abs_top_builddir/daemon +path_add = $$abs_top_builddir/src$(PATH_SEPARATOR)$$abs_top_builddir/daemon$(PATH_SEPARATOR)$$abs_top_builddir/tools # NB, automake < 1.10 does not provide the real # abs_top_{src/build}dir variables, so don't rely diff --git a/tests/cpuset b/tests/cpuset index eeb0ee2..9f43269 100755 --- a/tests/cpuset +++ b/tests/cpuset @@ -16,17 +16,21 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" fail=0 # generate input -virsh --connect test:///default dumpxml 1 > xml || fail=1 +$abs_top_builddir/tools/virsh --connect test:///default dumpxml 1 > xml || fail=1 # require the presence of the string we'll transform grep '<vcpu>' xml > /dev/null || fail=1 @@ -34,7 +38,7 @@ grep '<vcpu>' xml > /dev/null || fail=1 sed "s/vcpu>/vcpu cpuset='aaa'>/" xml > xml-invalid || fail=1 # Require failure and a diagnostic. -virsh --connect test:///default define xml-invalid > out 2>&1 && fail=1 +$abs_top_builddir/tools/virsh --connect test:///default define xml-invalid > out 2>&1 && fail=1 cat <<\EOF > exp || fail=1 error: Failed to define domain from xml-invalid error: failed Xen syscall topology cpuset syntax error diff --git a/tests/daemon-conf b/tests/daemon-conf index 071497e..722fe4e 100755 --- a/tests/daemon-conf +++ b/tests/daemon-conf @@ -1,16 +1,20 @@ #!/bin/sh # Get coverage of libvirtd's config-parsing code. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - libvirtd --version + $abs_top_builddir/daemon/libvirtd --version fi -test -z "$srcdir" && srcdir=$(pwd) -test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" + +test -z "$CONFIG_HEADER" && CONFIG_HEADER="$abs_top_builddir/config.h" -grep '^#define WITH_QEMU 1' $CONFIG_HEADER > /dev/null || +grep '^#define WITH_QEMU 1' "$CONFIG_HEADER" > /dev/null || skip_test_ "configured without QEMU support" conf="$abs_top_srcdir/daemon/libvirtd.conf" @@ -40,7 +44,7 @@ while :; do esac # Run libvirtd, expecting it to fail. - libvirtd --config=$f 2> err && fail=1 + $abs_top_builddir/daemon/libvirtd --config=$f 2> err && fail=1 case $rhs in # '"'*) msg='should be a string';; @@ -73,7 +77,7 @@ sed 's,^log_outputs.*,log_outputs="3:file:'"$(pwd)/log"'",' tmp.conf > k \ mv k tmp.conf || fail=1 printf "running libvirtd with a valid config file ($sleep_secs seconds)\n" 1>&2 -libvirtd --pid-file=pid-file --config=tmp.conf > log 2>&1 & pid=$! +$abs_top_builddir/daemon/libvirtd --pid-file=pid-file --config=tmp.conf > log 2>&1 & pid=$! sleep $sleep_secs kill $pid diff --git a/tests/define-dev-segfault b/tests/define-dev-segfault index 4ae286f..e0b953a 100755 --- a/tests/define-dev-segfault +++ b/tests/define-dev-segfault @@ -2,13 +2,15 @@ # Exercise a bug whereby defining a valid domain could kill libvirtd. # The bug can also be exercised with a simple define/dumpxml pair to virsh. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -test -z "$srcdir" && srcdir=$(pwd) -test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. . "$srcdir/test-lib.sh" fail=0 @@ -62,7 +64,7 @@ cat <<\EOF > D.xml || fail=1 EOF url=test:///default -virsh --connect "$url" 'define D.xml; dumpxml D' > out 2>&1 || fail=1 +$abs_top_builddir/tools/virsh --connect "$url" 'define D.xml; dumpxml D' > out 2>&1 || fail=1 cat > exp <<EOF || fail=1 Domain D defined from D.xml diff --git a/tests/int-overflow b/tests/int-overflow index ac3119b..c9f5de9 100755 --- a/tests/int-overflow +++ b/tests/int-overflow @@ -2,16 +2,16 @@ # Ensure that an invalid domain ID isn't interpreted as a valid one. # Before, an ID of 2^32+2 would be treated just like an ID of 2. -# Boilerplate code to set up a test directory, cd into it, -# and to ensure we remove it upon completion. -this_test_() { echo "./$0" | sed 's,.*/,,'; } -t_=$(this_test_)-$$ -init_cwd_=$(pwd) -trap 'st=$?; d='"$t_"'; - cd '"$init_cwd_"' && chmod -R u+rwx "$d" && rm -rf "$d" && exit $st' 0 -trap '(exit $?); exit $?' 1 2 13 15 -mkdir "$t_" || fail=1 -cd "$t_" || fail=1 +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + +if test "$VERBOSE" = yes; then + set -x + $abs_top_builddir/tools/virsh --version +fi + +. "$srcdir/test-lib.sh" echo "error: failed to get domain '4294967298'" > exp || fail=1 echo domname 4294967298 | $abs_top_builddir/tools/virsh --quiet \ diff --git a/tests/libvirtd-fail b/tests/libvirtd-fail index eeb1ee6..c6b6876 100755 --- a/tests/libvirtd-fail +++ b/tests/libvirtd-fail @@ -1,18 +1,20 @@ #!/bin/sh # Ensure that libvirt fails when given nonexistent --config=FILE +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - libvirtd --version + $abs_top_builddir/daemon/libvirtd --version fi -test -z "$srcdir" && srcdir=$(pwd) -test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. . "$srcdir/test-lib.sh" fail=0 -libvirtd --config=no-such-conf --timeout=5 2> log +$abs_top_builddir/daemon/libvirtd --config=no-such-conf --timeout=5 2> log RET=$? test "$RET" != "0" && exit 0 || exit 1 diff --git a/tests/libvirtd-pool b/tests/libvirtd-pool index 59da5a6..ca1db94 100755 --- a/tests/libvirtd-pool +++ b/tests/libvirtd-pool @@ -1,20 +1,22 @@ #!/bin/sh # Get coverage of virsh pool-define-as XML formatting +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -test -z "$srcdir" && srcdir=$(pwd) -test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. . "$srcdir/test-lib.sh" fail=0 pwd=$(pwd) || fail=1 -virsh --connect test:///default \ +$abs_top_builddir/tools/virsh --connect test:///default \ pool-define-as --print-xml \ P dir src-host /src/path /src/dev S /target-path \ 1>out 2>&1 diff --git a/tests/read-bufsiz b/tests/read-bufsiz index 7d53735..5baa7c5 100755 --- a/tests/read-bufsiz +++ b/tests/read-bufsiz @@ -16,20 +16,21 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=`pwd` +test -z "$abs_top_srcdir" && abs_top_srcdir=`pwd`/.. +test -z "$abs_top_builddir" && abs_top_builddir=`pwd`/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -test -z "$srcdir" && srcdir=`pwd` -test -z "$abs_top_srcdir" && abs_top_srcdir=`pwd`/.. - -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" fail=0 # Output a valid definition, to be used as input. -virsh -c test:///default dumpxml 1 > xml || fail=1 +$abs_top_builddir/tools/virsh -c test:///default dumpxml 1 > xml || fail=1 for i in before after; do # The largest BUFSIZ I've seen is 128K. This is slightly larger. @@ -38,7 +39,7 @@ for i in before after; do # Append or prepend enough spaces to push the size over the limit: ( test $i = before && cat sp xml || cat xml sp ) > $in || fail=1 - virsh --connect test:///default define $in > out || fail=1 + $abs_top_builddir/tools/virsh --connect test:///default define $in > out || fail=1 printf "Domain test defined from $in\n\n" > exp || fail=1 compare exp out || fail=1 done diff --git a/tests/read-non-seekable b/tests/read-non-seekable index 8a7bdcd..59c2389 100755 --- a/tests/read-non-seekable +++ b/tests/read-non-seekable @@ -16,12 +16,16 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=`pwd` +test -z "$abs_top_srcdir" && abs_top_srcdir=`pwd`/.. +test -z "$abs_top_builddir" && abs_top_builddir=`pwd`/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" fail=0 @@ -40,11 +44,11 @@ cat <<\EOF > dom </domain> EOF -virsh -c test:///default define dom > /dev/null || fail=1 +$abs_top_builddir/tools/virsh -c test:///default define dom > /dev/null || fail=1 mkfifo_or_skip_ fifo cat dom > fifo & -virsh -c test:///default define fifo > /dev/null || fail=1 +$abs_top_builddir/tools/virsh -c test:///default define fifo > /dev/null || fail=1 (exit $fail); exit $fail diff --git a/tests/start b/tests/start index f457d59..930a6d9 100755 --- a/tests/start +++ b/tests/start @@ -16,13 +16,15 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -test -z "$srcdir" && srcdir=$(pwd) -test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. . "$srcdir/test-lib.sh" fail=0 @@ -30,7 +32,7 @@ fail=0 test_url=test:///default # expect this to fail -virsh -c $test_url start test > out 2> err && fail=1 +$abs_top_builddir/tools/virsh -c $test_url start test > out 2> err && fail=1 # stdout gets a newline echo > exp || fail=1 diff --git a/tests/test_conf.sh b/tests/test_conf.sh index 62c2324..682f1f5 100755 --- a/tests/test_conf.sh +++ b/tests/test_conf.sh @@ -1,17 +1,19 @@ #!/bin/sh +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. +test -z "$abs_srcdir" && abs_srcdir=$(pwd) +test -z "$abs_builddir" && abs_builddir=$(pwd) + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" set -e -if test "x$abs_srcdir" = x; then - abs_srcdir=`pwd` - abs_builddir=`pwd` -fi fail=0 i=1 diff --git a/tests/undefine b/tests/undefine index fafdae7..48b0ad9 100755 --- a/tests/undefine +++ b/tests/undefine @@ -16,17 +16,21 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" fail=0 # Attempt to undefine a running domain, by domain name. -virsh -q -c test:///default undefine test > out 2>&1 +$abs_top_builddir/tools/virsh -q -c test:///default undefine test > out 2>&1 test $? = 1 || fail=1 cat <<\EOF > exp || fail=1 error: Failed to undefine domain test @@ -35,7 +39,7 @@ EOF compare exp out || fail=1 # A different diagnostic when specifying a domain ID -virsh -q -c test:///default undefine 1 > out 2>&1 +$abs_top_builddir/tools/virsh -q -c test:///default undefine 1 > out 2>&1 test $? = 1 || fail=1 cat <<\EOF > exp || fail=1 error: a running domain like 1 cannot be undefined; @@ -44,7 +48,7 @@ EOF compare exp out || fail=1 # Succeed, now: first shut down, then undefine, both via name. -virsh -q -c test:///default 'shutdown test; undefine test' > out 2>&1 +$abs_top_builddir/tools/virsh -q -c test:///default 'shutdown test; undefine test' > out 2>&1 test $? = 0 || fail=1 cat <<\EOF > exp || fail=1 Domain test is being shutdown diff --git a/tests/vcpupin b/tests/vcpupin index 79b02dc..a72ad4c 100755 --- a/tests/vcpupin +++ b/tests/vcpupin @@ -16,17 +16,21 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -. $srcdir/test-lib.sh +. "$srcdir/test-lib.sh" fail=0 # Invalid syntax. -virsh --connect test:///default vcpupin test a 0,1 > out 2>&1 +$abs_top_builddir/tools/virsh --connect test:///default vcpupin test a 0,1 > out 2>&1 test $? = 1 || fail=1 cat <<\EOF > exp || fail=1 error: vcpupin: Invalid or missing vCPU number. @@ -35,7 +39,7 @@ EOF compare exp out || fail=1 # An out-of-range vCPU number deserves a diagnostic, too. -virsh --connect test:///default vcpupin test 100 0,1 > out 2>&1 +$abs_top_builddir/tools/virsh --connect test:///default vcpupin test 100 0,1 > out 2>&1 test $? = 1 || fail=1 cat <<\EOF > exp || fail=1 error: vcpupin: Invalid vCPU number. diff --git a/tests/virsh-all b/tests/virsh-all index 03ea466..81b3e57 100755 --- a/tests/virsh-all +++ b/tests/virsh-all @@ -16,27 +16,29 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -test -z "$srcdir" && srcdir=$(pwd) -test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. . "$srcdir/test-lib.sh" fail=0 test_url=test:///default -virsh -c $test_url help > cmds || framework_failure +$abs_top_builddir/tools/virsh -c $test_url help > cmds || framework_failure cmds=$(sed -n 's/^ $[^ ][^ ]*$ .*/\1/p' cmds) || framework_failure test -n "$cmds" || framework_failure for i in $cmds; do echo testing $i... 1>&2 # For now, just run the command and ignore output and exit status. - virsh -c $test_url $i < /dev/null > /dev/null 2>&1 + $abs_top_builddir/tools/virsh -c $test_url $i < /dev/null > /dev/null 2>&1 done (exit $fail); exit $fail diff --git a/tests/virsh-synopsis b/tests/virsh-synopsis index 24038dd..d72e887 100755 --- a/tests/virsh-synopsis +++ b/tests/virsh-synopsis @@ -16,24 +16,27 @@ # You should have received a copy of the GNU General Public License # along with this program. If not, see <http://www.gnu.org/licenses/>. +test -z "$srcdir" && srcdir=$(pwd) +test -z "$abs_top_srcdir" && abs_top_srcdir=$(pwd)/.. +test -z "$abs_top_builddir" && abs_top_builddir=$(pwd)/.. + if test "$VERBOSE" = yes; then set -x - virsh --version + $abs_top_builddir/tools/virsh --version fi -test -z "$srcdir" && srcdir=$(pwd) . "$srcdir/test-lib.sh" fail=0 test_url=test:///default -virsh -c $test_url help > cmds || framework_failure +$abs_top_builddir/tools/virsh -c $test_url help > cmds || framework_failure cmds=$(sed -n 's/^ $[^ ][^ ]*$ .*/\1/p' cmds) || framework_failure test -n "$cmds" || framework_failure for i in $cmds; do - virsh -c $test_url help $i > help || fail=1 + $abs_top_builddir/tools/virsh -c $test_url help $i > help || fail=1 grep -A1 '^ SYNOPSIS$' help > synopsis \ || { echo 1>&2 missing or invalid help SYNOPSIS for $i; fail=1; } sed -n 2p synopsis > s2 || framework_failure -- 1.6.0.6

2 2

[libvirt] boot.kernel.org support in libvirt
by Itamar Reis Peixoto 07 Oct '09

07 Oct '09

Hello guy's what you think to add bko support in libvirt/virt-manager for example boot from hard disk boot from network / pxe boot from kernel.org please look --> http://boot.kernel.org/ -- ------------ Itamar Reis Peixoto e-mail/msn: itamar(a)ispbrasil.com.br sip: itamar(a)ispbrasil.com.br skype: itamarjp icq: 81053601 +55 11 4063 5033 +55 34 3221 8599

2 1

Re: [libvirt] Xen: Domain.migrate() causes invalid listDomains() output
by Thomas Treutner 07 Oct '09

07 Oct '09

On Wednesday 07 October 2009 10:25:11 Thomas Treutner wrote: > Shall I provide more debug output? Here the debug output of libvirtd: http://tt.scripty.at/tmp/node01.txt.gz http://tt.scripty.at/tmp/node02.txt.gz At start, domain wp01 is @ node01, and wp02 @ node02. Then their placement is exchanged. After migration: node01:~# xm list Name ID Mem VCPUs State Time(s) Domain-0 0 256 2 r----- 37.9 wp02 2 512 2 -b---- 0.9 node02:~# xm list Name ID Mem VCPUs State Time(s) Domain-0 0 192 2 r----- 52.4 wp01 3 512 2 -b---- 1.0 virt-top etc. don't see wp01@node02 anymore. libvirt obviously returns ID 1 for it, which can't be resolved. As there is no alternative to int[] listDomains() (f.e., return UUIDs), I'm stuck here. NB: Doing migration with xm migrate works fine, libvirt isn't confused by that. -t-

1 0

[libvirt] cann't connect kvm hypervisor
by Liu, Zhentao 07 Oct '09

07 Oct '09

Hello, I have used virsh to connect my hypervisor, what is the reason? ------------------------------------------------------------------------------- root@forest:/# virsh -c qemu:///system list error: unable to connect to '/var/run/libvirt/libvirt-sock': Connection refused error: failed to connect to the hypervisor --------------------------------------------------------------------------------- Thanks Zhentao Liu

2 1

[libvirt] [PATCH] Support for IPv6 / multiple ifaces in interface_conf.[ch]
by Laine Stump 06 Oct '09

06 Oct '09

This patch updates the xml parsing and formatting, and the associated virInterfaceDef data structure to support IPv6, along the way adding support for multiple protocols per interface, and multiple IP addresses per protocol. Note that netcf appears to not accept defining both ipv4 and ipv6 on the same interface, and still can't report live config of IPv6 (or multiple IPv4 addresses), so the usefulness of this patch is limited until those items are fixed in netcf. This patch applies on top of commit 1ceabc5a152efa8892954cb4f45b85553edfee9f, which I submitted on Sept 28. --- src/conf/interface_conf.c | 308 ++++++++++++++++++++++++++++++++++----------- src/conf/interface_conf.h | 21 +++- 2 files changed, 247 insertions(+), 82 deletions(-) diff --git a/src/conf/interface_conf.c b/src/conf/interface_conf.c index b422464..24f111c 100644 --- a/src/conf/interface_conf.c +++ b/src/conf/interface_conf.c @@ -53,9 +53,32 @@ void virInterfaceBareDefFree(virInterfaceBareDefPtr def) { VIR_FREE(def); } +static +void virInterfaceIpDefFree(virInterfaceIpDefPtr def) { + if (def == NULL) + return; + VIR_FREE(def->address); + VIR_FREE(def->source); +} + +static +void virInterfaceProtocolDefFree(virInterfaceProtocolDefPtr def) { + int ii; + + if (def == NULL) + return; + for (ii = 0; ii < def->nips; ii++) { + virInterfaceIpDefFree(def->ips[ii]); + } + VIR_FREE(def->ips); + VIR_FREE(def->family); + VIR_FREE(def->gateway); + VIR_FREE(def); +} + void virInterfaceDefFree(virInterfaceDefPtr def) { - int i; + int i, pp; if (def == NULL) return; @@ -89,11 +112,11 @@ void virInterfaceDefFree(virInterfaceDefPtr def) break; } - VIR_FREE(def->proto.family); - VIR_FREE(def->proto.address); - VIR_FREE(def->proto.gateway); - VIR_FREE(def->proto.source); - + /* free all protos */ + for (pp = 0; pp < def->nprotos; pp++) { + virInterfaceProtocolDefFree(def->protos[pp]); + } + VIR_FREE(def->protos); VIR_FREE(def); } @@ -226,22 +249,22 @@ virInterfaceDefParseBondArpValid(virConnectPtr conn, xmlXPathContextPtr ctxt) { } static int -virInterfaceDefParseDhcp(virConnectPtr conn, virInterfaceDefPtr def, +virInterfaceDefParseDhcp(virConnectPtr conn, virInterfaceProtocolDefPtr def, xmlNodePtr dhcp, xmlXPathContextPtr ctxt) { xmlNodePtr save; char *tmp; int ret = 0; - def->proto.dhcp = 1; + def->dhcp = 1; save = ctxt->node; ctxt->node = dhcp; /* Not much to do in the current version */ tmp = virXPathString(conn, "string(./@peerdns)", ctxt); if (tmp) { if (STREQ(tmp, "yes")) - def->proto.peerdns = 1; + def->peerdns = 1; else if (STREQ(tmp, "no")) - def->proto.peerdns = 0; + def->peerdns = 0; else { virInterfaceReportError(conn, VIR_ERR_XML_ERROR, _("unknown dhcp peerdns value %s"), tmp); @@ -249,88 +272,209 @@ virInterfaceDefParseDhcp(virConnectPtr conn, virInterfaceDefPtr def, } VIR_FREE(tmp); } else - def->proto.peerdns = -1; + def->peerdns = -1; ctxt->node = save; return(ret); } static int -virInterfaceDefParseIp(virConnectPtr conn, virInterfaceDefPtr def, - xmlNodePtr ip ATTRIBUTE_UNUSED, xmlXPathContextPtr ctxt) { +virInterfaceDefParseIp(virConnectPtr conn, virInterfaceIpDefPtr def, + xmlXPathContextPtr ctxt) { int ret = 0; char *tmp; long l; - tmp = virXPathString(conn, "string(./ip[1]/@address)", ctxt); - def->proto.address = tmp; + tmp = virXPathString(conn, "string(./@address)", ctxt); + def->address = tmp; if (tmp != NULL) { - ret = virXPathLong(conn, "string(./ip[1]/@prefix)", ctxt, &l); + ret = virXPathLong(conn, "string(./@prefix)", ctxt, &l); if (ret == 0) - def->proto.prefix = (int) l; + def->prefix = (int) l; else if (ret == -2) { virInterfaceReportError(conn, VIR_ERR_XML_ERROR, "%s", _("Invalid ip address prefix value")); return(-1); } - tmp = virXPathString(conn, "string(./ip[1]/@source)", ctxt); - def->proto.source = tmp; + tmp = virXPathString(conn, "string(./@source)", ctxt); + def->source = tmp; } - tmp = virXPathString(conn, "string(./route[1]/@gateway)", ctxt); - def->proto.gateway = tmp; return(0); } static int -virInterfaceDefParseProtoIPv4(virConnectPtr conn, virInterfaceDefPtr def, +virInterfaceDefParseProtoIPv4(virConnectPtr conn, virInterfaceProtocolDefPtr def, xmlXPathContextPtr ctxt) { - xmlNodePtr dhcp, ip; - int ret = 0; + xmlNodePtr dhcp; + xmlNodePtr *ipNodes = NULL; + int nIpNodes, ii, ret = 0; + char *tmp; + + tmp = virXPathString(conn, "string(./route[1]/@gateway)", ctxt); + def->gateway = tmp; dhcp = virXPathNode(conn, "./dhcp", ctxt); if (dhcp != NULL) ret = virInterfaceDefParseDhcp(conn, def, dhcp, ctxt); + if (ret != 0) + return(ret); + + nIpNodes = virXPathNodeSet(conn, "./ip", ctxt, &ipNodes); + if (ipNodes == NULL) + return 0; + + if (VIR_ALLOC_N(def->ips, nIpNodes) < 0) { + virReportOOMError(conn); + ret = -1; + goto error; + } + + def->nips = 0; + for (ii = 0; ii < nIpNodes; ii++) { + virInterfaceIpDefPtr ip; + + if (VIR_ALLOC(ip) < 0) { + virReportOOMError(conn); + break; + } + + ctxt->node = ipNodes[ii]; + ret = virInterfaceDefParseIp(conn, ip, ctxt); + if (ret != 0) { + virInterfaceIpDefFree(ip); + break; + } + def->ips[def->nips++] = ip; + } + +error: + VIR_FREE(ipNodes); + return(ret); +} + +static int +virInterfaceDefParseProtoIPv6(virConnectPtr conn, virInterfaceProtocolDefPtr def, + xmlXPathContextPtr ctxt) { + xmlNodePtr dhcp, autoconf; + xmlNodePtr *ipNodes = NULL; + int nIpNodes, ii, ret = 0; + char *tmp; + + tmp = virXPathString(conn, "string(./route[1]/@gateway)", ctxt); + def->gateway = tmp; + + autoconf = virXPathNode(conn, "./autoconf", ctxt); + if (autoconf != NULL) + def->autoconf = 1; + + dhcp = virXPathNode(conn, "./dhcp", ctxt); + if (dhcp != NULL) + ret = virInterfaceDefParseDhcp(conn, def, dhcp, ctxt); if (ret != 0) return(ret); - ip = virXPathNode(conn, "./ip", ctxt); - if (ip != NULL) - ret = virInterfaceDefParseIp(conn, def, ip, ctxt); + nIpNodes = virXPathNodeSet(conn, "./ip", ctxt, &ipNodes); + if (ipNodes == NULL) + return 0; + + if (VIR_ALLOC_N(def->ips, nIpNodes) < 0) { + virReportOOMError(conn); + ret = -1; + goto error; + } + + def->nips = 0; + for (ii = 0; ii < nIpNodes; ii++) { + + virInterfaceIpDefPtr ip; + + if (VIR_ALLOC(ip) < 0) { + virReportOOMError(conn); + break; + } + + ctxt->node = ipNodes[ii]; + ret = virInterfaceDefParseIp(conn, ip, ctxt); + if (ret != 0) { + virInterfaceIpDefFree(ip); + break; + } + def->ips[def->nips++] = ip; + } + +error: + VIR_FREE(ipNodes); return(ret); } static int virInterfaceDefParseIfAdressing(virConnectPtr conn, virInterfaceDefPtr def, xmlXPathContextPtr ctxt) { - xmlNodePtr cur, save; - int ret; + xmlNodePtr save; + xmlNodePtr *protoNodes = NULL; + int nProtoNodes, pp, ret = 0; char *tmp; - cur = virXPathNode(conn, "./protocol[1]", ctxt); - if (cur == NULL) - return(0); save = ctxt->node; - ctxt->node = cur; - tmp = virXPathString(conn, "string(./@family)", ctxt); - if (tmp == NULL) { - virInterfaceReportError(conn, VIR_ERR_XML_ERROR, - "%s", _("protocol misses the family attribute")); - ret = -1; - goto done; + + nProtoNodes = virXPathNodeSet(conn, "./protocol", ctxt, &protoNodes); + if (nProtoNodes <= 0) { + /* no protocols is an acceptable outcome */ + return 0; } - if (STREQ(tmp, "ipv4")) { - def->proto.family = tmp; - ret = virInterfaceDefParseProtoIPv4(conn, def, ctxt); - } else { - virInterfaceReportError(conn, VIR_ERR_XML_ERROR, - _("unsupported protocol family '%s'"), tmp); + + if (VIR_ALLOC_N(def->protos, nProtoNodes) < 0) { + virReportOOMError(conn); ret = -1; - VIR_FREE(tmp); + goto error; } -done: + def->nprotos = 0; + for (pp = 0; pp < nProtoNodes; pp++) { + + virInterfaceProtocolDefPtr proto; + + if (VIR_ALLOC(proto) < 0) { + virReportOOMError(conn); + break; + } + + ctxt->node = protoNodes[pp]; + tmp = virXPathString(conn, "string(./@family)", ctxt); + if (tmp == NULL) { + virInterfaceReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("protocol misses the family attribute")); + virInterfaceProtocolDefFree(proto); + ret = -1; + break; + } + proto->family = tmp; + if (STREQ(tmp, "ipv4")) { + ret = virInterfaceDefParseProtoIPv4(conn, proto, ctxt); + if (ret != 0) { + virInterfaceProtocolDefFree(proto); + break; + } + } else if (STREQ(tmp, "ipv6")) { + ret = virInterfaceDefParseProtoIPv6(conn, proto, ctxt); + if (ret != 0) { + virInterfaceProtocolDefFree(proto); + break; + } + } else { + virInterfaceReportError(conn, VIR_ERR_XML_ERROR, + _("unsupported protocol family '%s'"), tmp); + ret = -1; + virInterfaceProtocolDefFree(proto); + break; + } + def->protos[def->nprotos++] = proto; + } + +error: + VIR_FREE(protoNodes); ctxt->node = save; return(ret); @@ -1005,36 +1149,48 @@ virInterfaceVlanDefFormat(virConnectPtr conn, virBufferPtr buf, static int virInterfaceProtocolDefFormat(virConnectPtr conn ATTRIBUTE_UNUSED, virBufferPtr buf, const virInterfaceDefPtr def) { - char prefixStr[16]; - if (def->proto.family == NULL) - return(0); - virBufferVSprintf(buf, " <protocol family='%s'>\n", def->proto.family); - if (def->proto.dhcp) { - if (def->proto.peerdns == 0) - virBufferAddLit(buf, " <dhcp peerdns='no'/>\n"); - else if (def->proto.peerdns == 1) - virBufferAddLit(buf, " <dhcp peerdns='yes'/>\n"); - else - virBufferAddLit(buf, " <dhcp/>\n"); - } - if (def->proto.address != NULL) { - if (def->proto.prefix != 0) - snprintf(prefixStr, sizeof(prefixStr), "%d", def->proto.prefix); - - virBufferVSprintf(buf, " <ip address='%s'%s%s%s%s%s%s/>\n", - def->proto.address, - def->proto.prefix ? " prefix='" : "", - def->proto.prefix ? prefixStr : "", - def->proto.prefix ? "'" : "", - def->proto.source ? " source='" : "", - def->proto.source ? def->proto.source : "", - def->proto.source ? "'" : ""); - } - if (def->proto.gateway != NULL) { - virBufferVSprintf(buf, " <route gateway='%s'/>\n", - def->proto.gateway); + int pp, ii; + + for (pp = 0; pp < def->nprotos; pp++) { + + virBufferVSprintf(buf, " <protocol family='%s'>\n", def->protos[pp]->family); + + if (def->protos[pp]->autoconf) { + virBufferAddLit(buf, " <autoconf/>\n"); + } + + if (def->protos[pp]->dhcp) { + if (def->protos[pp]->peerdns == 0) + virBufferAddLit(buf, " <dhcp peerdns='no'/>\n"); + else if (def->protos[pp]->peerdns == 1) + virBufferAddLit(buf, " <dhcp peerdns='yes'/>\n"); + else + virBufferAddLit(buf, " <dhcp/>\n"); + } + + for (ii = 0; ii < def->protos[pp]->nips; ii++) { + if (def->protos[pp]->ips[ii]->address != NULL) { + + virBufferVSprintf(buf, " <ip address='%s'", + def->protos[pp]->ips[ii]->address); + if (def->protos[pp]->ips[ii]->prefix != 0) { + virBufferVSprintf(buf, " prefix='%d'", + def->protos[pp]->ips[ii]->prefix); + } + if (def->protos[pp]->ips[ii]->source != NULL) { + virBufferVSprintf(buf, " source='%s'", + def->protos[pp]->ips[ii]->source); + } + virBufferAddLit(buf, "/>\n"); + } + } + if (def->protos[pp]->gateway != NULL) { + virBufferVSprintf(buf, " <route gateway='%s'/>\n", + def->protos[pp]->gateway); + } + + virBufferAddLit(buf, " </protocol>\n"); } - virBufferAddLit(buf, " </protocol>\n"); return(0); } diff --git a/src/conf/interface_conf.h b/src/conf/interface_conf.h index df871aa..49acdd6 100644 --- a/src/conf/interface_conf.h +++ b/src/conf/interface_conf.h @@ -123,16 +123,25 @@ struct _virInterfaceVlanDef { char *devname; /* device name for vlan */ }; +typedef struct _virInterfaceIpDef virInterfaceIpDef; +typedef virInterfaceIpDef *virInterfaceIpDefPtr; +struct _virInterfaceIpDef { + char *address; /* ip address */ + int prefix; /* ip prefix */ + char *source; /* source of address - "device" or "config" (default) */ +}; + + typedef struct _virInterfaceProtocolDef virInterfaceProtocolDef; typedef virInterfaceProtocolDef *virInterfaceProtocolDefPtr; struct _virInterfaceProtocolDef { - char *family; /* ipv4 only right now */ + char *family; /* ipv4 or ipv6 */ int dhcp; /* use dhcp */ int peerdns; /* dhcp peerdns ? */ - char *address; /* ip address */ - int prefix; /* ip prefix */ + int autoconf; /* only useful if family is ipv6 */ + int nips; + virInterfaceIpDefPtr *ips; /* ptr to array of ips[nips] */ char *gateway; /* route gateway */ - char *source; /* source of address - "device" or "config" (default) */ }; @@ -152,8 +161,8 @@ struct _virInterfaceDef { virInterfaceBondDef bond; } data; - /* separated as we may allow multiple of those in the future */ - virInterfaceProtocolDef proto; + int nprotos; + virInterfaceProtocolDefPtr *protos; /* ptr to array of protos[nprotos] */ }; typedef struct _virInterfaceObj virInterfaceObj; -- 1.6.2.5

3 5

[libvirt] [PATCH] Support reporting live interface IP/netmask.
by Laine Stump 06 Oct '09

06 Oct '09

From: root <root(a)vlap.laine.org> This patch adds the flag VIR_INTERFACE_XML_INACTIVE to virInterfaceGetXMLDesc's flags. When it is *not* set (the default), the live interface info will be returned in the XML. in particular, the IP address(es) and netmask(s) will be retrieved by querying the device directly, rather than just reporting what's in the config file. The backend of this is in netcf's new ncf_if_xml_state() function. Any live interface ip address info in the xml will have the property "source" set to "device", eg: <ip address='10.24.0.1' prefix='24' source='device'/> Also, if an interface is currently inactive, no ip addresses will be returned, since an inactive interface device can't be queried for IP addresses (effectively it has none). A difference in the XML from previously - it is now acceptable to have both a dhcp *and* an ip node (or neither) within the protocol node. Before you had to have one or the other, but couldn't have both. Note that you need at least netcf 0.1.2 for this to build and work, and an upcoming release (patches submitted today) for it to work exactly as described above. --- include/libvirt/libvirt.h.in | 4 +++ src/conf/interface_conf.c | 62 ++++++++++++++++++++++------------------- src/conf/interface_conf.h | 1 + src/interface/netcf_driver.c | 8 ++++- src/libvirt.c | 15 +++++++--- tools/virsh.c | 10 +++++- 6 files changed, 63 insertions(+), 37 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 4e63e48..fd0c90b 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -922,6 +922,10 @@ virInterfacePtr virInterfaceLookupByMACString (virConnectPtr conn, const char* virInterfaceGetName (virInterfacePtr iface); const char* virInterfaceGetMACString (virInterfacePtr iface); +typedef enum { + VIR_INTERFACE_XML_INACTIVE = 1 /* dump inactive interface information */ +} virInterfaceXMLFlags; + char * virInterfaceGetXMLDesc (virInterfacePtr iface, unsigned int flags); virInterfacePtr virInterfaceDefineXML (virConnectPtr conn, diff --git a/src/conf/interface_conf.c b/src/conf/interface_conf.c index e646351..b422464 100644 --- a/src/conf/interface_conf.c +++ b/src/conf/interface_conf.c @@ -92,6 +92,7 @@ void virInterfaceDefFree(virInterfaceDefPtr def) VIR_FREE(def->proto.family); VIR_FREE(def->proto.address); VIR_FREE(def->proto.gateway); + VIR_FREE(def->proto.source); VIR_FREE(def); } @@ -272,6 +273,8 @@ virInterfaceDefParseIp(virConnectPtr conn, virInterfaceDefPtr def, "%s", _("Invalid ip address prefix value")); return(-1); } + tmp = virXPathString(conn, "string(./ip[1]/@source)", ctxt); + def->proto.source = tmp; } tmp = virXPathString(conn, "string(./route[1]/@gateway)", ctxt); def->proto.gateway = tmp; @@ -282,22 +285,19 @@ virInterfaceDefParseIp(virConnectPtr conn, virInterfaceDefPtr def, static int virInterfaceDefParseProtoIPv4(virConnectPtr conn, virInterfaceDefPtr def, xmlXPathContextPtr ctxt) { - xmlNodePtr cur; - int ret; + xmlNodePtr dhcp, ip; + int ret = 0; - cur = virXPathNode(conn, "./dhcp", ctxt); - if (cur != NULL) - ret = virInterfaceDefParseDhcp(conn, def, cur, ctxt); - else { - cur = virXPathNode(conn, "./ip", ctxt); - if (cur != NULL) - ret = virInterfaceDefParseIp(conn, def, cur, ctxt); - else { - virInterfaceReportError(conn, VIR_ERR_XML_ERROR, - "%s", _("interface miss dhcp or ip adressing")); - ret = -1; - } - } + dhcp = virXPathNode(conn, "./dhcp", ctxt); + if (dhcp != NULL) + ret = virInterfaceDefParseDhcp(conn, def, dhcp, ctxt); + + if (ret != 0) + return(ret); + + ip = virXPathNode(conn, "./ip", ctxt); + if (ip != NULL) + ret = virInterfaceDefParseIp(conn, def, ip, ctxt); return(ret); } @@ -1005,6 +1005,7 @@ virInterfaceVlanDefFormat(virConnectPtr conn, virBufferPtr buf, static int virInterfaceProtocolDefFormat(virConnectPtr conn ATTRIBUTE_UNUSED, virBufferPtr buf, const virInterfaceDefPtr def) { + char prefixStr[16]; if (def->proto.family == NULL) return(0); virBufferVSprintf(buf, " <protocol family='%s'>\n", def->proto.family); @@ -1015,20 +1016,23 @@ virInterfaceProtocolDefFormat(virConnectPtr conn ATTRIBUTE_UNUSED, virBufferAddLit(buf, " <dhcp peerdns='yes'/>\n"); else virBufferAddLit(buf, " <dhcp/>\n"); - } else { - /* theorically if we don't have dhcp we should have an address */ - if (def->proto.address != NULL) { - if (def->proto.prefix != 0) - virBufferVSprintf(buf, " <ip address='%s' prefix='%d'/>\n", - def->proto.address, def->proto.prefix); - else - virBufferVSprintf(buf, " <ip address='%s'/>\n", - def->proto.address); - } - if (def->proto.gateway != NULL) { - virBufferVSprintf(buf, " <route gateway='%s'/>\n", - def->proto.gateway); - } + } + if (def->proto.address != NULL) { + if (def->proto.prefix != 0) + snprintf(prefixStr, sizeof(prefixStr), "%d", def->proto.prefix); + + virBufferVSprintf(buf, " <ip address='%s'%s%s%s%s%s%s/>\n", + def->proto.address, + def->proto.prefix ? " prefix='" : "", + def->proto.prefix ? prefixStr : "", + def->proto.prefix ? "'" : "", + def->proto.source ? " source='" : "", + def->proto.source ? def->proto.source : "", + def->proto.source ? "'" : ""); + } + if (def->proto.gateway != NULL) { + virBufferVSprintf(buf, " <route gateway='%s'/>\n", + def->proto.gateway); } virBufferAddLit(buf, " </protocol>\n"); return(0); diff --git a/src/conf/interface_conf.h b/src/conf/interface_conf.h index bb9dce4..df871aa 100644 --- a/src/conf/interface_conf.h +++ b/src/conf/interface_conf.h @@ -132,6 +132,7 @@ struct _virInterfaceProtocolDef { char *address; /* ip address */ int prefix; /* ip prefix */ char *gateway; /* route gateway */ + char *source; /* source of address - "device" or "config" (default) */ }; diff --git a/src/interface/netcf_driver.c b/src/interface/netcf_driver.c index ca14fb0..b5c3664 100644 --- a/src/interface/netcf_driver.c +++ b/src/interface/netcf_driver.c @@ -326,7 +326,7 @@ cleanup: } static char *interfaceGetXMLDesc(virInterfacePtr ifinfo, - unsigned int flags ATTRIBUTE_UNUSED) + unsigned int flags) { struct interface_driver *driver = ifinfo->conn->interfacePrivateData; struct netcf_if *iface = NULL; @@ -342,7 +342,11 @@ static char *interfaceGetXMLDesc(virInterfacePtr ifinfo, goto cleanup; } - xmlstr = ncf_if_xml_desc(iface); + if ((flags & VIR_INTERFACE_XML_INACTIVE)) { + xmlstr = ncf_if_xml_desc(iface); + } else { + xmlstr = ncf_if_xml_state(iface); + } if (!xmlstr) { const char *errmsg, *details; int errcode = ncf_error(driver->netcf, &errmsg, &details); diff --git a/src/libvirt.c b/src/libvirt.c index bcb89e1..bd712dc 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -5915,10 +5915,17 @@ virInterfaceGetMACString(virInterfacePtr iface) /** * virInterfaceGetXMLDesc: * @iface: an interface object - * @flags: an OR'ed set of extraction flags, not used yet + * @flags: an OR'ed set of extraction flags. Current valid bits: + * + * VIR_INTERFACE_XML_INACTIVE - return the static configuration, + * suitable for use redefining the + * interface via virInterfaceDefineXML() * - * Provide an XML description of the interface. The description may be reused - * later to redefine the interface with virInterfaceDefineXML(). + * Provide an XML description of the interface. If + * VIR_INTERFACE_XML_INACTIVE is set, the description may be reused + * later to redefine the interface with virInterfaceDefineXML(). If it + * is not set, the ip address and netmask will be the current live + * setting of the interface, not the settings from the config files. * * Returns a 0 terminated UTF-8 encoded XML instance, or NULL in case of error. * the caller must free() the returned value. @@ -5935,7 +5942,7 @@ virInterfaceGetXMLDesc(virInterfacePtr iface, unsigned int flags) virLibInterfaceError(NULL, VIR_ERR_INVALID_INTERFACE, __FUNCTION__); return (NULL); } - if (flags != 0) { + if ((flags & ~VIR_INTERFACE_XML_INACTIVE) != 0) { virLibInterfaceError(iface, VIR_ERR_INVALID_ARG, __FUNCTION__); goto error; } diff --git a/tools/virsh.c b/tools/virsh.c index 3482389..77f1d70 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -2763,7 +2763,7 @@ cmdInterfaceEdit (vshControl *ctl, const vshCmd *cmd) char *doc = NULL; char *doc_edited = NULL; char *doc_reread = NULL; - int flags = 0; + int flags = VIR_INTERFACE_XML_INACTIVE; if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) goto cleanup; @@ -3297,6 +3297,7 @@ static const vshCmdInfo info_interface_dumpxml[] = { static const vshCmdOptDef opts_interface_dumpxml[] = { {"interface", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("interface name or MAC address")}, + {"inactive", VSH_OT_BOOL, 0, gettext_noop("show inactive defined XML")}, {NULL, 0, 0, NULL} }; @@ -3306,6 +3307,11 @@ cmdInterfaceDumpXML(vshControl *ctl, const vshCmd *cmd) virInterfacePtr iface; int ret = TRUE; char *dump; + int flags = 0; + int inactive = vshCommandOptBool(cmd, "inactive"); + + if (inactive) + flags |= VIR_INTERFACE_XML_INACTIVE; if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -3313,7 +3319,7 @@ cmdInterfaceDumpXML(vshControl *ctl, const vshCmd *cmd) if (!(iface = vshCommandOptInterface(ctl, cmd, NULL))) return FALSE; - dump = virInterfaceGetXMLDesc(iface, 0); + dump = virInterfaceGetXMLDesc(iface, flags); if (dump != NULL) { printf("%s", dump); free(dump); -- 1.6.2.5

5 16

[libvirt] [RFC] Multi-IQN proposal
by Shyam_Iyer＠Dell.com 06 Oct '09

06 Oct '09

Would this proposal be acceptable ? Example XML schema for an iSCSI storage pool created -- <pool type="iscsi"> <name>dell</name> <uuid>cf354733-b01b-8870-2040-94888640e24d</uuid> <capacity>0</capacity> <allocation>0</allocation> <available>0</available> - <source> <initiator iqnname = "<initiator IQN1>"> <initiator iqnname = "<initiator IQN2>"> ........................................ ........................................ <host name="<iSCSI target hostname or Target IP address>" /> <device path="<iSCSI Target IQN name>" /> </source> - <target> <path>/dev/disk/by-path</path> - <permissions> <mode>0700</mode> <owner>0</owner> <group>0</group> </permissions> </target> </pool> Each initiator iqn name can be parsed to create the unique sessions. This should solve the following possibilities -- * possibility of multiple IQNs for a single Guest * option for Guest's own BIOS & initiator to use these IQNs (iSCSI in Guest) * option for hypervisor's initiator to use these IQNs on behalf of the guest (Multi-IQN) Compile tested only. Needs beatification. diff --git a/src/storage_conf.c b/src/storage_conf.c index cb063cc..915e897 100644 --- a/src/storage_conf.c +++ b/src/storage_conf.c @@ -116,6 +116,7 @@ enum { VIR_STORAGE_POOL_SOURCE_DIR = (1<<2), VIR_STORAGE_POOL_SOURCE_ADAPTER = (1<<3), VIR_STORAGE_POOL_SOURCE_NAME = (1<<4), + VIR_STORAGE_POOL_SOURCE_INITIATOR_IQN = (1<<5), }; @@ -537,6 +538,33 @@ virStoragePoolDefParseXML(virConnectPtr conn, goto cleanup; } } + +/* Read the conf here */ + if (options->flags & VIR_STORAGE_POOL_SOURCE_INITATOR_IQN) { + int niqn, i; + + if ((niqn = virXPathNodeSet(conn, "./source/host/initiator", ctxt, &nodeset)) < 0) { + virStorageReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("cannot extract storage pool source devices")); + goto cleanup; + } + if (VIR_ALLOC_N(ret->source.niqn, niqn) < 0) { + VIR_FREE(nodeset); + virReportOOMError(conn); + goto cleanup; + } + for (i = 0 ; i < niqn ; i++ ) { + xmlChar *name = xmlGetProp(nodeset[i], BAD_CAST "iqnname"); + if (path == NULL) { + VIR_FREE(nodeset); + virStorageReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("missing storage pool source device path")); + goto cleanup; + } + ret->source.initiator[i].iqnname = (char *)name; + } + } if (options->flags & VIR_STORAGE_POOL_SOURCE_DEVICE) { xmlNodePtr *nodeset = NULL; int nsource, i; diff --git a/src/storage_conf.h b/src/storage_conf.h index 421d305..c2504be 100644 --- a/src/storage_conf.h +++ b/src/storage_conf.h @@ -202,7 +202,12 @@ struct _virStoragePoolSourceDevice { } geometry; }; - +/* Initiator Attributes */ +typedef virStoragePoolSourceInitiatorAttr *virStoragePoolSourceInitiatorAttrPtr; +struct virStoragePoolSourceInitiatorAttr { + char *iqnname; + /* We could put other initiator specific things here */ +} typedef struct _virStoragePoolSource virStoragePoolSource; typedef virStoragePoolSource *virStoragePoolSourcePtr; @@ -214,6 +219,9 @@ struct _virStoragePoolSource { int ndevice; virStoragePoolSourceDevicePtr devices; + /*And either one or more initiators*/ + virStoragePoolSourceInitiatorAttrPtr initiator; + /* Or a directory */ char *dir;

4 9

[libvirt] [RFC PATCH 0/6] Series short description
by Gerhard Stenzel 06 Oct '09

06 Oct '09

The following series of patches is a prototype implementation of a ebtables based MAC address filter. This is currently not intended for inclusion, but rather to get some feedback/comments about the approach and the implementation. At the moment, some defaults are set when libvirtd starts and when a domain is started or destroyed. The plan is to extend this filter capability to the API level and virsh command level. The plan is also to add more filtering features like VLAN filtering and QoS filtering. Thanks, Gerhard --- Gerhard Stenzel (6): remove currently unused parameter to pass make check test add MAC address based port filtering to qemu add MAC address based port filtering to libvirt add the ebtables wrapper to network driver add build support for ebtables wrapper add ebtables wrapper configure.in | 3 include/libvirt/libvirt.h.in | 14 + src/Makefile.am | 1 src/conf/network_conf.h | 10 src/driver.h | 14 + src/libvirt.c | 119 +++++ src/libvirt_private.syms | 26 + src/libvirt_public.syms | 3 src/network/bridge_driver.c | 175 ++++++++ src/qemu/qemu.conf | 3 src/qemu/qemu_conf.c | 14 + src/qemu/qemu_conf.h | 2 src/qemu/qemu_driver.c | 23 + src/test/test_driver.c | 3 src/util/ebtables.c | 961 ++++++++++++++++++++++++++++++++++++++++++ src/util/ebtables.h | 134 ++++++ 16 files changed, 1505 insertions(+), 0 deletions(-) create mode 100644 src/util/ebtables.c create mode 100644 src/util/ebtables.h --

2 11

[libvirt] [PATCH] 526769 change logrotate config default to weekly
by Daniel Veillard 06 Oct '09

06 Oct '09

https://bugzilla.redhat.com/show_bug.cgi?id=526769 Current setup is daily with 7 backups so one week worth of logs per domain in /var/log/libvirt/qemu/*.log First a week worth of backup is a bit short, second for most domains the logs are really small, and with 7 backup files this directory gets crowded quickly. This patches changes to weekly, with 4 backups, so always a month worth of logs, and will nearly cut in half the number of entries in that directory. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

3 6

[libvirt] Regarding connecting to libvirt from windows
by anuj rampal 06 Oct '09

06 Oct '09

Hi, I can connect to libvirt over TCP/IP from windows and call all the functions. what i want is to do some kind of authentication like entering some kind of username/password. im using virConnectOpen(const char* name) function. i tried using virConnectOpenAuth().. but whatever parameter ii give it gets connected and there is no option to pass username/password. Thanks in advance. Regards Anuj

1 0

[libvirt] Creating Guest OS in Virtual Box
by santosh gandham 06 Oct '09

06 Oct '09

Hi, I am trying to create Guest OS (VM) in virtualBox throught libvirt. I installed libvirt 0.7.0 and created the template as <domain type='vbox'> <name>ubuntu</name> <os> <type>hvm</type> <boot dev='hd'/> </os> <memory>542720</memory> <vcpu>1</vcpu> <devices> <disk type='file' device='disk'> <source file='/home/santhosh/Desktop/ubuntu9.vdi'/> <target dev='hdd'/> </disk> <graphics type='rdp' port='3386'/> </devices> </domain> Then I issued this command $ virsh -c vbox:///session define /home/santhosh/Desktop/ubun.tmpl Domain ubuntu defined from /home/santhosh/Desktop/ubun.tmpl When I open VirtualBox, it has created a domain named "ubuntu". All the parameters specified in the template file are set except "Hard disks" . So, when I start the VM, I am left with boot errors 00:00:06.318 Guest Log: BIOS: int13_harddisk: function 02, unmapped device for ELDL=80 00:00:06.319 Guest Log: BIOS: Boot from Hard Disk 0 failed 00:00:06.321 Guest Log: Could not read from the boot medium! System halted. It is unable to boot from the disk image which I specified in the template and I am sure the path specified to disk image in the template is correct. I am unable to diagnize and find where is the error. Can any one help me here to know what might be the problem and what should I do. Thank you. Best Regards, Gandham Santhosh

2 3

[libvirt] USB node devices info
by Daniel Berteaud 05 Oct '09

05 Oct '09

Hi there. I have a problem with USB host device enumeration with libvirt. I'm running libvirt 0.7.0 on a fully updated CentOS 5.3 (libvirt was rebuilt unmodified from http://libvirt.org/sources/libvirt-0.7.0-1.fc11.src.rpm) I want to pas an USB key to one of my guest. lsusb see it just fine: (I want to pass the device 13fe:1d23) Bus 005 Device 001: ID 0000:0000 Bus 005 Device 002: ID 046b:ff10 American Megatrends, Inc. Bus 004 Device 001: ID 0000:0000 Bus 007 Device 001: ID 0000:0000 Bus 006 Device 001: ID 0000:0000 Bus 008 Device 001: ID 0000:0000 Bus 002 Device 006: ID 13fe:1d23 Kingston Technology Company Inc. Bus 002 Device 001: ID 0000:0000 Bus 001 Device 001: ID 0000:0000 Bus 003 Device 001: ID 0000:0000 Bus 003 Device 002: ID 413c:2003 Dell Computer Corp. Keyboard If I create a XML file for this device, I can attach it to my guest with virsh attach-device, no problem here. The problem is that libvirt cannot access the devices informations like pretty name, vendorID and productID, so for example, virt-manager only see devices with name "USB Raw Device Access" with 0000 as vendorID and productID. For example, for this particular USB key: virsh nodedev-dumpxml usb_device_13fe_1d23_07780BA646FD_0_usbraw <device> <name>usb_device_13fe_1d23_07780BA646FD_0_usbraw</name> <parent>usb_device_13fe_1d23_07780BA646FD_0</parent> <capability type='usb_device'> <bus>0</bus> <device>0</device> <product id='0x0000'>USB Raw Device Access</product> <vendor id='0x0000' /> </capability> </device> SELinux is in permissive mode. Anybody have an idea ? Where does libvirt get the info about host device ? (/sys ?) Regards, Daniel -- Daniel Berteaud FIREWALL-SERVICES SARL. Société de Services en Logiciels Libres Technopôle Montesquieu 33650 MARTILLAC Tel : 05 56 64 15 32 Fax : 05 56 64 15 32 Mail: daniel(a)firewall-services.com Web : http://www.firewall-services.com

1 0

[libvirt] regarding libvirtd start
by santosh gandham 05 Oct '09

05 Oct '09

Hi, I have installed libvirt 0.7.0. in fedora 11. When I start libvirt by using #libvirtd start , it is not showing any status message like libvirtd has started. The command line stuck up there itself. What is the problem? What should I do to run the libvirtd successfully? Thank you. -- Gandham Santhosh

2 1

[libvirt] [PATCH 0/5] test: Add node device driver
by Cole Robinson 05 Oct '09

05 Oct '09

The following patches implement a basic node device driver for the test hypervisor driver. This is useful for unit tests in libvirt users (particularly virtinst). Cole Robinson (5): nodedev: Add locking in nodeNumOfDevices nodedev: Break out virNodeDeviceHasCap to node_conf test: Implement node device driver. node conf: Make parsing routines consistent with other drivers test: Support loading node device info from file/XML examples/xml/test/testdev.xml | 16 ++ examples/xml/test/testnode.xml | 1 + examples/xml/test/testnodeinline.xml | 19 +++ src/conf/node_device_conf.c | 57 ++++++- src/conf/node_device_conf.h | 11 ++ src/libvirt_private.syms | 1 + src/node_device/node_device_driver.c | 22 +-- src/test/test_driver.c | 291 +++++++++++++++++++++++++++++++++- 8 files changed, 396 insertions(+), 22 deletions(-) create mode 100644 examples/xml/test/testdev.xml

2 11

[libvirt] [PATCH] test: Throw a proper error in GetBridgeName
by Cole Robinson 05 Oct '09

05 Oct '09

Throw error in GetBridgeName if net has no bridge. Signed-off-by: Cole Robinson <crobinso(a)redhat.com> --- src/test/test_driver.c | 10 ++++++++-- 1 files changed, 8 insertions(+), 2 deletions(-) diff --git a/src/test/test_driver.c b/src/test/test_driver.c index f57c92a..30014fc 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -2546,8 +2546,14 @@ static char *testNetworkGetBridgeName(virNetworkPtr network) { goto cleanup; } - if (privnet->def->bridge && - !(bridge = strdup(privnet->def->bridge))) { + if (!(privnet->def->bridge)) { + testError(network->conn, VIR_ERR_INTERNAL_ERROR, + _("network '%s' does not have a bridge name."), + privnet->def->name); + goto cleanup; + } + + if (!(bridge = strdup(privnet->def->bridge))) { virReportOOMError(network->conn); goto cleanup; } -- 1.6.5.rc2

3 3

[libvirt] [PATCHv2 0/8] Python related fixes
by Cole Robinson 05 Oct '09

05 Oct '09

The following series fixes various issues related to the python bindings. Cole Robinson (8): configure: Add explict --with-python option. python: Remove FastParser from generator. python: Remove use of xmllib in generator.py python: Don't generate conflicting conn.createXML functions. python: Don't generate bindings for vir*Ref python: Use a pure python implementation of 'vir*GetConnect' python: Fix generated virInterface method names python: Add a newline after custom classes configure.in | 6 ++- python/generator.py | 117 ++++++++++++++++++++++----------------------------- 2 files changed, 55 insertions(+), 68 deletions(-)

3 20

[libvirt] A first look at the node device libudev backend
by Dave Allan 05 Oct '09

05 Oct '09

Hi all, Here is a first cut at the node device backend. The only devices implemented are part of storage and PCI devices, but it's conceptually finished. The remaining work is implementing the other device properties and types. Before I go do that, I'd like to get feedback on my approach. Since libudev doesn't have the concept of capabilities the way HAL does, Chris Lalancette and I thought it made more sense and rendered much more readable code to implement each device type in a single function and avoid the indirection created by the caps_tbl structure containing a function pointer. Monitoring of libudev for new devices is not implemented, but I'm not too concerned about that at this point. As Dan Berrange suggested on IRC, putting a call in the main loop to poll the monitor file descriptor will do the job. Dave >From 34fa8bf6ce1b72276dc73f292ae12a9dd1ab2433 Mon Sep 17 00:00:00 2001 From: David Allan <dallan(a)redhat.com> Date: Fri, 25 Sep 2009 10:59:28 -0400 Subject: [PATCH 1/1] A first look at the libudev node device backend. Thanks to Jiri for the build system changes and to Chris Lalancette for his code and suggestions on how this backend should be implemented. Storage device properties are partly populated and PCI devices are discovered, but all other devices are as yet unimplemented. --- configure.in | 47 ++- daemon/libvirtd.c | 3 +- src/Makefile.am | 16 +- src/node_device/node_device_driver.c | 3 + src/node_device/node_device_driver.h | 22 + src/node_device/node_device_hal.h | 19 - ...evice_hal_linux.c => node_device_linux_sysfs.c} | 0 src/node_device/node_device_udev.c | 627 ++++++++++++++++++++ src/node_device/node_device_udev.h | 27 + src/util/util.c | 46 ++ src/util/util.h | 4 + 11 files changed, 788 insertions(+), 26 deletions(-) rename src/node_device/{node_device_hal_linux.c => node_device_linux_sysfs.c} (100%) create mode 100644 src/node_device/node_device_udev.c create mode 100644 src/node_device/node_device_udev.h diff --git a/configure.in b/configure.in index cb5ce55..356abb7 100644 --- a/configure.in +++ b/configure.in @@ -1560,7 +1560,7 @@ test "$enable_shared" = no && lt_cv_objdir=. LV_LIBTOOL_OBJDIR=${lt_cv_objdir-.} AC_SUBST([LV_LIBTOOL_OBJDIR]) -dnl HAL or DeviceKit library for host device enumeration +dnl HAL, DeviceKit, or libudev library for host device enumeration HAL_REQUIRED=0.0 HAL_CFLAGS= HAL_LIBS= @@ -1654,8 +1654,46 @@ AM_CONDITIONAL([HAVE_DEVKIT], [test "x$with_devkit" = "xyes"]) AC_SUBST([DEVKIT_CFLAGS]) AC_SUBST([DEVKIT_LIBS]) +UDEV_REQUIRED=143 +UDEV_CFLAGS= +UDEV_LIBS= +AC_ARG_WITH([udev], + [ --with-udev use libudev for host device enumeration], + [], + [with_udev=check]) + +if test "$with_libvirtd" = "no" ; then + with_udev=no +fi +if test "x$with_udev" = "xyes" -o "x$with_udev" = "xcheck"; then + PKG_CHECK_MODULES(UDEV, libudev >= $UDEV_REQUIRED, + [with_udev=yes], [ + if test "x$with_udev" = "xcheck" ; then + with_udev=no + else + AC_MSG_ERROR( + [You must install udev-devel >= $UDEV_REQUIRED to compile libvirt]) + fi + ]) + if test "x$with_udev" = "xyes" ; then + AC_DEFINE_UNQUOTED([HAVE_UDEV], 1, + [use UDEV for host device enumeration]) + + old_CFLAGS=$CFLAGS + old_LDFLAGS=$LDFLAGS + CFLAGS="$CFLAGS $UDEV_CFLAGS" + LDFLAGS="$LDFLAGS $UDEV_LIBS" + AC_CHECK_FUNCS([udev_new],,[with_udev=no]) + CFLAGS="$old_CFLAGS" + LDFLAGS="$old_LDFLAGS" + fi +fi +AM_CONDITIONAL([HAVE_UDEV], [test "x$with_udev" = "xyes"]) +AC_SUBST([UDEV_CFLAGS]) +AC_SUBST([UDEV_LIBS]) + with_nodedev=no; -if test "$with_devkit" = "yes" -o "$with_hal" = "yes"; +if test "$with_devkit" = "yes" -o "$with_hal" = "yes" -o "$with_udev" = "yes"; then with_nodedev=yes AC_DEFINE_UNQUOTED([WITH_NODE_DEVICES], 1, [with node device driver]) @@ -1815,6 +1853,11 @@ AC_MSG_NOTICE([ devkit: $DEVKIT_CFLAGS $DEVKIT_LIBS]) else AC_MSG_NOTICE([ devkit: no]) fi +if test "$with_udev" = "yes" ; then +AC_MSG_NOTICE([ udev: $UDEV_CFLAGS $UDEV_LIBS]) +else +AC_MSG_NOTICE([ udev: no]) +fi if test "$with_netcf" = "yes" ; then AC_MSG_NOTICE([ netcf: $NETCF_CFLAGS $NETCF_LIBS]) else diff --git a/daemon/libvirtd.c b/daemon/libvirtd.c index 2bae782..c60a09f 100644 --- a/daemon/libvirtd.c +++ b/daemon/libvirtd.c @@ -832,8 +832,7 @@ static struct qemud_server *qemudInitialize(int sigread) { #ifdef WITH_STORAGE_DIR storageRegister(); #endif -#if defined(WITH_NODE_DEVICES) && \ - (defined(HAVE_HAL) || defined(HAVE_DEVKIT)) +#if defined(WITH_NODE_DEVICES) nodedevRegister(); #endif secretRegister(); diff --git a/src/Makefile.am b/src/Makefile.am index 9cbec47..b4328d6 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -237,16 +237,20 @@ SECURITY_DRIVER_SELINUX_SOURCES = \ NODE_DEVICE_DRIVER_SOURCES = \ - node_device/node_device_driver.c node_device/node_device_driver.h + node_device/node_device_driver.c \ + node_device/node_device_driver.h \ + node_device/node_device_linux_sysfs.c NODE_DEVICE_DRIVER_HAL_SOURCES = \ node_device/node_device_hal.c \ - node_device/node_device_hal.h \ - node_device/node_device_hal_linux.c + node_device/node_device_hal.h NODE_DEVICE_DRIVER_DEVKIT_SOURCES = \ node_device/node_device_devkit.c +NODE_DEVICE_DRIVER_UDEV_SOURCES = \ + node_device/node_device_udev.c + ######################### # @@ -626,6 +630,11 @@ libvirt_driver_nodedev_la_SOURCES += $(NODE_DEVICE_DRIVER_DEVKIT_SOURCES) libvirt_driver_nodedev_la_CFLAGS += $(DEVKIT_CFLAGS) libvirt_driver_nodedev_la_LDFLAGS += $(DEVKIT_LIBS) endif +if HAVE_UDEV +libvirt_driver_nodedev_la_SOURCES += $(NODE_DEVICE_DRIVER_UDEV_SOURCES) +libvirt_driver_nodedev_la_CFLAGS += $(UDEV_CFLAGS) +libvirt_driver_nodedev_la_LDFLAGS += $(UDEV_LIBS) +endif if WITH_DRIVER_MODULES libvirt_driver_nodedev_la_LDFLAGS += -module -avoid-version @@ -667,6 +676,7 @@ EXTRA_DIST += \ $(NODE_DEVICE_DRIVER_SOURCES) \ $(NODE_DEVICE_DRIVER_HAL_SOURCES) \ $(NODE_DEVICE_DRIVER_DEVKIT_SOURCES) \ + $(NODE_DEVICE_DRIVER_UDEV_SOURCES) \ $(SECURITY_DRIVER_SELINUX_SOURCES) \ $(SECRET_DRIVER_SOURCES) \ $(VBOX_DRIVER_EXTRA_DIST) diff --git a/src/node_device/node_device_driver.c b/src/node_device/node_device_driver.c index 93ca28c..3d06ed1 100644 --- a/src/node_device/node_device_driver.c +++ b/src/node_device/node_device_driver.c @@ -760,5 +760,8 @@ int nodedevRegister(void) { #ifdef HAVE_DEVKIT return devkitNodeRegister(); #endif +#ifdef HAVE_UDEV + return udevNodeRegister(); +#endif #endif } diff --git a/src/node_device/node_device_driver.h b/src/node_device/node_device_driver.h index db01624..5be0781 100644 --- a/src/node_device/node_device_driver.h +++ b/src/node_device/node_device_driver.h @@ -45,6 +45,9 @@ int halNodeRegister(void); #ifdef HAVE_DEVKIT int devkitNodeRegister(void); #endif +#ifdef HAVE_UDEV +int udevNodeRegister(void); +#endif void nodeDeviceLock(virDeviceMonitorStatePtr driver); void nodeDeviceUnlock(virDeviceMonitorStatePtr driver); @@ -53,4 +56,23 @@ void registerCommonNodeFuncs(virDeviceMonitorPtr mon); int nodedevRegister(void); +#ifdef __linux__ + +#define check_fc_host(d) check_fc_host_linux(d) +int check_fc_host_linux(union _virNodeDevCapData *d); + +#define check_vport_capable(d) check_vport_capable_linux(d) +int check_vport_capable_linux(union _virNodeDevCapData *d); + +#define read_wwn(host, file, wwn) read_wwn_linux(host, file, wwn) +int read_wwn_linux(int host, const char *file, char **wwn); + +#else /* __linux__ */ + +#define check_fc_host(d) +#define check_vport_capable(d) +#define read_wwn(host, file, wwn) + +#endif /* __linux__ */ + #endif /* __VIR_NODE_DEVICE_H__ */ diff --git a/src/node_device/node_device_hal.h b/src/node_device/node_device_hal.h index c859fe3..8ac8a35 100644 --- a/src/node_device/node_device_hal.h +++ b/src/node_device/node_device_hal.h @@ -22,23 +22,4 @@ #ifndef __VIR_NODE_DEVICE_HAL_H__ #define __VIR_NODE_DEVICE_HAL_H__ -#ifdef __linux__ - -#define check_fc_host(d) check_fc_host_linux(d) -int check_fc_host_linux(union _virNodeDevCapData *d); - -#define check_vport_capable(d) check_vport_capable_linux(d) -int check_vport_capable_linux(union _virNodeDevCapData *d); - -#define read_wwn(host, file, wwn) read_wwn_linux(host, file, wwn) -int read_wwn_linux(int host, const char *file, char **wwn); - -#else /* __linux__ */ - -#define check_fc_host(d) -#define check_vport_capable(d) -#define read_wwn(host, file, wwn) - -#endif /* __linux__ */ - #endif /* __VIR_NODE_DEVICE_HAL_H__ */ diff --git a/src/node_device/node_device_hal_linux.c b/src/node_device/node_device_linux_sysfs.c similarity index 100% rename from src/node_device/node_device_hal_linux.c rename to src/node_device/node_device_linux_sysfs.c diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c new file mode 100644 index 0000000..22669ff --- /dev/null +++ b/src/node_device/node_device_udev.c @@ -0,0 +1,627 @@ +/* + * node_device_udev.c: node device enumeration - libudev implementation + * + * Copyright (C) 2009 Red Hat + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Author: Dave Allan <dallan(a)redhat.com> + */ + +#include <config.h> +#include <libudev.h> +#include <regex.h> + +#include "node_device_udev.h" +#include "virterror_internal.h" +#include "node_device_conf.h" +#include "node_device_driver.h" +#include "driver.h" +#include "datatypes.h" +#include "uuid.h" +#include "logging.h" +#include "memory.h" +#include "util.h" + +#define VIR_FROM_THIS VIR_FROM_NODEDEV + +static virDeviceMonitorStatePtr driverState = NULL; + +static struct udev *udevGetUdev(void) +{ + return udev_monitor_get_udev(DRV_STATE_UDEV_MONITOR(driverState)); +} + + +static int udevGetSysfsData(char **data, const char *path) +{ + char *tmp = NULL; + FILE *fp = NULL; + int ret = -1; + + if (VIR_ALLOC_N(tmp, SYSFS_DATA_SIZE) != 0) { + goto out; + } + + memset(tmp, 0, SYSFS_DATA_SIZE); + + fp = fopen(path, "r"); + if (fp == NULL) { + goto out; + } + + if (fread(tmp, SYSFS_DATA_SIZE, 1, fp) == SYSFS_DATA_SIZE) { + goto out; + } + + /* strip off the trailing \n */ + tmp[strlen(tmp) - 1] = '\0'; + + /* return a block of memory containing only what was read; + * otherwise we will consume vast amounts of largely unused memory + * on systems with large numbers of devices. */ + *data = strdup(tmp); + if (*data == NULL) { + goto out; + } + + ret = 0; + +out: + VIR_FREE(tmp); + fclose(fp); + + return ret; +} + + +static int udevGetPCI(unsigned int *data, + const char *devpath, + const char *field) +{ + const char *syspath = NULL; + char *path = NULL, *tmp = NULL; + char *dummy; + int ret = 0; + + syspath = udev_get_sys_path(udevGetUdev()); + + if (virBuildPath(&path, syspath, devpath, field) != 0) { + goto out; + } + + if (udevGetSysfsData(&tmp, path) == 0) { + ret = virStrToLong_ui(tmp, &dummy, 16, data); + } + +out: + VIR_FREE(path); + return ret; +} + + +static int udevGetDMIData(char **data, + const char *field) +{ + const char *syspath = NULL; + char *path = NULL; + int ret = -1; + + syspath = udev_get_sys_path(udevGetUdev()); + + if (virBuildPath(&path, syspath, "class/dmi/id", field) == 0) { + ret = udevGetSysfsData(data, path); + VIR_FREE(path); + } + + return ret; +} + + +/* This function allocates memory from the heap for the property + * value. That memory must be later freed by some other code. */ +static int udevGetDeviceProperty(struct udev_device *udev_device, + const char *property_key, + char **property_value) +{ + const char *udev_value = NULL; + int ret = -1; + + udev_value = udev_device_get_property_value(udev_device, property_key); + if (udev_value == NULL) { + VIR_ERROR(_("udev reports property '%s' does not exist"), + property_key); + goto out; + } + + /* If this allocation is changed, the comment at the beginning + * of the function must also be changed. */ + *property_value = strdup(udev_value); + if (*property_value == NULL) { + VIR_ERROR(_("Failed to allocate memory for " + "property '%s' on device '%s'"), + property_key, udev_device_get_sysname(udev_device)); + goto out; + } + + ret = 0; + +out: + if (ret != 0) { + VIR_ERROR(_("Failed to get udev property '%s' for device '%s'"), + property_key, udev_device_get_sysname(udev_device)); + } + + return ret; +} + + +static int udevGetStringProperty(struct udev_device *udev_device, + const char *property_key, + char **value) +{ + return udevGetDeviceProperty(udev_device, property_key, value); +} + +/* +static int udevGetIntProperty(struct udev_device *udev_device, + const char *property_key, + int *value) +{ + char *udev_value = NULL; + int ret = 0; + + ret = udevGetDeviceProperty(udev_device, property_key, &udev_value); + + if (ret == 0) { + ret = virStrToLong_i(udev_value, NULL, 0, value); + } + + return ret; +} + + +static int udevGetUint64Property(struct udev_device *udev_device, + const char *property_key, + uint64_t *value) +{ + char *udev_value = NULL; + int ret = 0; + + ret = udevGetDeviceProperty(udev_device, property_key, &udev_value); + + if (ret == 0) { + ret = virStrToLong_ull(udev_value, NULL, 0, value); + } + + return ret; +} +*/ + +static void udevLogFunction(struct udev *udev ATTRIBUTE_UNUSED, + int priority ATTRIBUTE_UNUSED, + const char *file, + int line, + const char *fn, + const char *fmt, + va_list args) +{ + VIR_ERROR_INT(file, fn, line, fmt, args); +} + + +static void virUdevGetDeviceProperties(struct udev *udev, const char *name) +{ + struct udev_device *device = NULL; + struct udev_list_entry *list_entry = NULL; + + device = udev_device_new_from_syspath(udev, name); + + udev_list_entry_foreach(list_entry, + udev_device_get_properties_list_entry(device)) { + + VIR_ERROR(" name: '%s' value: '%s'\n", + udev_list_entry_get_name(list_entry), + udev_list_entry_get_value(list_entry)); + } + + udev_device_unref(device); + + return; +} + + +static int udevProcessPCI(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + const char *devpath = NULL, *syspath = NULL; + int ret = -1; + + devpath = udev_device_get_devpath(device); + syspath = udev_device_get_syspath(device); + + if (VIR_ALLOC(def->caps) != 0) { + goto out; + } + + def->caps->type = VIR_NODE_DEV_CAP_PCI_DEV; + + char *p = strrchr(devpath, '/'); + if (p) { + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.domain); + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.bus); + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.slot); + virStrToLong_ui(p+1, &p, 16, &def->caps->data.pci_dev.function); + } + + if (udevGetPCI(&def->caps->data.pci_dev.vendor, devpath, "vendor") || + udevGetPCI(&def->caps->data.pci_dev.product, devpath, "device")) { + goto out; + } + + /* FIXME: to do the vendor name and product name, we have to + * parse /usr/share/hwdata/pci.ids. Look in hal/hald/ids.c + */ + + ret = 0; + +out: + return ret; +} + + +static int udevProcessStorage(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + const char *devpath = NULL; + struct udev_list_entry *list_entry = NULL; + int ret = -1; + + devpath = udev_device_get_devpath(device); + + /* Can we get libudev to add device subtypes? */ + if (STRPREFIX(devpath, "/devices/virtual/block")) { + goto out; + } + + if (VIR_ALLOC(def->caps) != 0) { + goto out; + } + + def->caps->type = VIR_NODE_DEV_CAP_STORAGE; + + def->caps->data.storage.block = strdup(udev_device_get_devnode(device)); + /* Can we get libudev to add subtypes? */ + def->caps->data.storage.drive_type = strdup("disk"); + + udev_list_entry_foreach(list_entry, udev_device_get_devlinks_list_entry(device)) { + if (STREQ(udev_list_entry_get_name(list_entry), "/dev/cdrom")) { + VIR_FREE(def->caps->data.storage.drive_type); + def->caps->data.storage.drive_type = strdup("cdrom"); + } + } + + /* These lines are somewhat more than 80 col, but I think the code + * is more readable if they aren't broken up. */ + if (udevGetStringProperty(device, "DEVNAME", &def->caps->data.storage.block)) { + /* None of these properties are available through libudev... + udevGetStringProperty(device, "ID_MODEL", &def->caps->data.storage.model) || + udevGetStringProperty(device, "ID_VENDOR", &def->caps->data.storage.vendor) || + udevGetStringProperty(device, "ID_SERIAL", &def->caps->data.storage.serial)) { + */ + goto out; + } + ret = 0; + +out: + return ret; +} + + +static int udevProcessOneDevice(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + virNodeDeviceObjPtr dev = NULL; + const char *devtype = NULL, *devpath = NULL; + regex_t pci_rec; + int ret = -1; + + /* How to handle parent devices? Shall we change the meaning of + * the parent pointer for the udev backend? What does client code + * use the parent relationship for? We really want to preserve + * the behavior of the HAL backend, but that may be difficult in + * practice. */ + //parent = udev_device_get_parent(device); + + devtype = udev_device_get_devtype(device); + devpath = udev_device_get_devpath(device); + + regcomp(&pci_rec, + "^/devices/pci[0-9]{4}:[0-9]{2}/[0-9]{4}:[0-9]{2}:[0-9]{2}.[0-9]$", + REG_EXTENDED | REG_NOSUB); + + if (devtype != NULL && STREQ(devtype, "disk")) { + ret = udevProcessStorage(device, def); + } else if (regexec(&pci_rec, devpath, 0, NULL, 0) == 0) { + /* This is puzzling...why isn't there a devtype for pci in libudev? */ + ret = udevProcessPCI(device, def); + } + + if (ret == 0) { + dev = virNodeDeviceAssignDef(NULL, &driverState->devs, def); + virNodeDeviceObjUnlock(dev); + } + + return ret; +} + + +static int udevProcessDeviceListEntry(struct udev *udev, + struct udev_list_entry *list_entry) +{ + struct udev_device *device; + virNodeDeviceDefPtr def = NULL; + const char *name = NULL; + int ret = -1; + + name = udev_list_entry_get_name(list_entry); + + if (VIR_ALLOC(def) != 0) { + goto out; + } + + def->name = strdup(name); + if (def->name == NULL) { + goto out; + } + + device = udev_device_new_from_syspath(udev, name); + if (device != NULL) { + udevProcessOneDevice(device, def); + udev_device_unref(device); + ret = 0; + } + +out: + return ret; +} + + +static int udevEnumerateDevices(struct udev *udev) +{ + struct udev_enumerate *udev_enumerate = NULL; + struct udev_list_entry *list_entry = NULL; + const char *name = NULL; + int ret = 0; + + udev_enumerate = udev_enumerate_new(udev); + + ret = udev_enumerate_scan_devices(udev_enumerate); + if (0 != ret) { + VIR_ERROR("udev scan devices returned %d\n", ret); + goto out; + } + + udev_list_entry_foreach(list_entry, + udev_enumerate_get_list_entry(udev_enumerate)) { + + udevProcessDeviceListEntry(udev, list_entry); + name = udev_list_entry_get_name(list_entry); + virUdevGetDeviceProperties(udev, name); + } + +out: + udev_enumerate_unref(udev_enumerate); + return ret; +} + + +static int udevDeviceMonitorShutdown(void) +{ + int ret = 0; + + struct udev_monitor *udev_monitor = NULL; + struct udev *udev = NULL; + + if (driverState) { + + nodeDeviceLock(driverState); + udev_monitor = DRV_STATE_UDEV_MONITOR(driverState); + + if (udev_monitor != NULL) { + udev = udev_monitor_get_udev(udev_monitor); + udev_monitor_unref(udev_monitor); + } + + if (udev != NULL) { + udev_unref(udev); + } + + virNodeDeviceObjListFree(&driverState->devs); + nodeDeviceUnlock(driverState); + virMutexDestroy(&driverState->lock); + VIR_FREE(driverState); + + } else { + ret = -1; + } + + return ret; +} + + +static int udevSetupSystemDev(void) +{ + virNodeDeviceDefPtr def = NULL; + virNodeDeviceObjPtr dev = NULL; + char *tmp = NULL; + int ret = -1; + + if (VIR_ALLOC(def) != 0) { + goto out; + } + + def->name = strdup("system"); + if (def->name == NULL) { + goto out; + } + + if (VIR_ALLOC(def->caps) != 0) { + goto out; + } + + /* These lines are somewhat more than 80 col, but I think the code + * is more readable if they aren't broken up. */ + if (udevGetDMIData(&def->caps->data.system.product_name, "product_name") || + udevGetDMIData(&def->caps->data.system.hardware.vendor_name, "sys_vendor") || + udevGetDMIData(&def->caps->data.system.hardware.version, "product_version") || + udevGetDMIData(&def->caps->data.system.hardware.serial, "product_serial")) { + goto out; + } + + udevGetDMIData(&tmp, "product_uuid"); + virUUIDParse(tmp, def->caps->data.system.hardware.uuid); + VIR_FREE(tmp); + + if (udevGetDMIData(&def->caps->data.system.firmware.vendor_name, "bios_vendor") || + udevGetDMIData(&def->caps->data.system.firmware.version, "bios_version") || + udevGetDMIData(&def->caps->data.system.firmware.release_date, "bios_date")) { + goto out; + } + + dev = virNodeDeviceAssignDef(NULL, &driverState->devs, def); + virNodeDeviceObjUnlock(dev); + + ret = 0; + +out: + return ret; +} + + +static int udevDeviceMonitorStartup(int privileged ATTRIBUTE_UNUSED) +{ + struct udev *udev = NULL; + struct udev_monitor *udev_monitor = NULL; + int ret = 0; + + if (VIR_ALLOC(driverState) < 0) { + ret = -1; + goto out; + } + + if (virMutexInit(&driverState->lock) < 0) { + VIR_FREE(driverState); + ret = -1; + goto out; + } + + nodeDeviceLock(driverState); + + /* + * http://www.kernel.org/pub/linux/utils/kernel/hotplug/libudev/libudev-udev.h… + * + * indicates no return value other than success, so we don't check + * its return value. + */ + udev = udev_new(); + udev_set_log_fn(udev, udevLogFunction); + + udev_monitor = udev_monitor_new_from_netlink(udev, "udev"); + if (udev_monitor == NULL) { + ret = -1; + goto out; + } + + /* udev can be retrieved from udev_monitor */ + driverState->privateData = udev_monitor; + nodeDeviceUnlock(driverState); + + /* Populate with known devices */ + + if (udevSetupSystemDev() != 0) { + ret = -1; + goto out; + } + + if (udevEnumerateDevices(udev) != 0) { + ret = -1; + goto out; + } + +out: + if (ret == -1) { + udevDeviceMonitorShutdown(); + } + return ret; +} + + +static int udevDeviceMonitorReload(void) +{ + return 0; +} + + +static int udevDeviceMonitorActive(void) +{ + /* Always ready to deal with a shutdown */ + return 0; +} + + +static virDrvOpenStatus udevNodeDrvOpen(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) +{ + if (driverState == NULL) { + return VIR_DRV_OPEN_DECLINED; + } + + conn->devMonPrivateData = driverState; + + return VIR_DRV_OPEN_SUCCESS; +} + +static int udevNodeDrvClose(virConnectPtr conn) +{ + conn->devMonPrivateData = NULL; + return 0; +} + +static virDeviceMonitor udevDeviceMonitor = { + .name = "udevDeviceMonitor", + .open = udevNodeDrvOpen, + .close = udevNodeDrvClose, +}; + +static virStateDriver udevStateDriver = { + .initialize = udevDeviceMonitorStartup, + .cleanup = udevDeviceMonitorShutdown, + .reload = udevDeviceMonitorReload, + .active = udevDeviceMonitorActive, +}; + +int udevNodeRegister(void) +{ + VIR_ERROR0("Registering udev node device backend\n"); + + registerCommonNodeFuncs(&udevDeviceMonitor); + if (virRegisterDeviceMonitor(&udevDeviceMonitor) < 0) { + return -1; + } + + return virRegisterStateDriver(&udevStateDriver); +} diff --git a/src/node_device/node_device_udev.h b/src/node_device/node_device_udev.h new file mode 100644 index 0000000..bac628d --- /dev/null +++ b/src/node_device/node_device_udev.h @@ -0,0 +1,27 @@ +/* + * node_device_udev.h: node device enumeration - libudev implementation + * + * Copyright (C) 2009 Red Hat + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Author: Dave Allan <dallan(a)redhat.com> + */ + +#include <libudev.h> +#include <stdint.h> + +#define SYSFS_DATA_SIZE 4096 +#define DRV_STATE_UDEV_MONITOR(ds) ((struct udev_monitor *)((ds)->privateData)) diff --git a/src/util/util.c b/src/util/util.c index ef07a6a..a1fe0d5 100644 --- a/src/util/util.c +++ b/src/util/util.c @@ -2088,3 +2088,49 @@ void virFileWaitForDevices(virConnectPtr conn) void virFileWaitForDevices(virConnectPtr conn ATTRIBUTE_UNUSED) {} #endif #endif + +/* Don't call this function directly, call virBuildPath instead. */ +int virBuildPathInternal(char **path, ...) +{ + char *path_component = NULL; + va_list ap; + size_t used = 0; + int ret = 0; + + if (VIR_ALLOC_N(*path, PATH_MAX) != 0) { + ret = -1; + goto out; + } + + va_start(ap, *path); + + while ((path_component = va_arg(ap, char *)) != NULL) + { + used += snprintf(*path + used, PATH_MAX - used, "%s/", path_component); + + /* From the snprintf man page: + + The functions snprintf() and vsnprintf() do not write more + than size bytes (including the trailing '\0'). If the + output was truncated due to this limit then the return + value is the number of characters (not including the + trailing '\0') which would have been written to the final + string if enough space had been available. Thus, a return + value of size or more means that the output was truncated. + + We therefore set used to the size of the buffer to avoid + the size being passed to snprintf from becoming negative. + */ + + if (used > PATH_MAX) { + used = PATH_MAX; + } + } + + va_end(ap); + + *(*path + used - 1) = '\0'; /* kill trailing slash */ + +out: + return ret; +} diff --git a/src/util/util.h b/src/util/util.h index dbfdc45..aa466af 100644 --- a/src/util/util.h +++ b/src/util/util.h @@ -244,4 +244,8 @@ char *virFileFindMountPoint(const char *type); void virFileWaitForDevices(virConnectPtr conn); +#define virBuildPath(path, ...) virBuildPathInternal(path, __VA_ARGS__, NULL) +/* Don't call virBuildPathInternal directly, use virBuildPath */ +int virBuildPathInternal(char **path, ...); + #endif /* __VIR_UTIL_H__ */ -- 1.6.4.4

2 1

[libvirt] cann't connect kvm hypervisor
by Liu, Zhentao 05 Oct '09

05 Oct '09

Hello, I'm new of Opennebula. I have installed KVM im my remote node. But when I want to check the KVM in the node, I can't connect to the hypervisor. -------------------------------------------------------------------- root@forest:/var/run/libvirt# virsh -c qemu:///localhost/system list Connecting to uri: qemu:///system error: failed to connect to the hypervisor -------------------------------------------------------------------- Can somebody tell me why virsh doesn't work? And how I can fix it? I think that it's maybe the problem of libvirt. Any helps would be very appreciated? Thanks Zhentao -----Original Message----- From: tinova79(a)gmail.com on behalf of Tino Vazquez Sent: Mon 10/5/2009 4:33 PM To: Shi Jin Cc: Liu, Zhentao; users(a)lists.opennebula.org Subject: Re: [one-users] cann't connect kvm hypervisor Hi there, I would suggest forwarding this question to the libvirt mailing list, other people there may know the best procedure to deal with this. Regards, -Tino -- Constantino Vázquez, Grid Technology Engineer/Researcher: http://www.dsa-research.org/tinova DSA Research Group: http://dsa-research.org Globus GridWay Metascheduler: http://www.GridWay.org OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org On Mon, Oct 5, 2009 at 4:30 PM, Shi Jin <jinzishuai(a)gmail.com> wrote: > I had similar problems earlier. > The only solution I found without rebooting the machine is to killall > kvm on the node. > I would also appreciate any suggestion to better handle this problem. > Thanks. > > Shi > > On Mon, Oct 5, 2009 at 5:56 AM, Liu, Zhentao > <Zhentao.Liu(a)fokus.fraunhofer.de> wrote: >> Hi Tino >> >> I have checked that libvirtd deaemon is running. I have also restarted >> libvirtd deaemon, but I can't still connect to kvm hypervisor. >> ----------------------------------------------------------- >> root@forest:/var/run/libvirt# /etc/init.d/libvirt-bin start >> * Starting libvirt management daemon libvirtd [ >> OK ] >> root@forest:/var/run/libvirt# virsh -c qemu:///system list >> >> Connecting to uri: qemu:///system >> error: failed to connect to the hypervisor >> ------------------------------------------------------------- >> >> other ideas? >> >> Thanks >> >> Zhentao >> >> -----Original Message----- >> From: tinova79(a)gmail.com on behalf of Tino Vazquez >> Sent: Mon 10/5/2009 1:43 PM >> To: Liu, Zhentao >> Cc: users(a)lists.opennebula.org >> Subject: Re: [one-users] cann't connect kvm hypervisor >> >> Hi Zhentao, >> >> Can you check that the libvirtd deaemon is running in the remote node? >> >> Regards, >> >> -Tino >> >> -- >> Constantino Vázquez, Grid Technology Engineer/Researcher: >> http://www.dsa-research.org/tinova >> DSA Research Group: http://dsa-research.org >> Globus GridWay Metascheduler: http://www.GridWay.org >> OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org >> >> >> >> On Mon, Oct 5, 2009 at 1:30 PM, Liu, Zhentao >> <Zhentao.Liu(a)fokus.fraunhofer.de> wrote: >>> Hello, >>> >>> I'm new of Opennebula. I have installed KVM im my remote node. But when I >>> want to check the KVM in the node, I can't connect to the hypervisor. >>> >>> -------------------------------------------------------------------- >>> root@forest:/var/run/libvirt# virsh -c qemu:///localhost/system list >>> Connecting to uri: qemu:///system >>> error: failed to connect to the hypervisor >>> -------------------------------------------------------------------- >>> >>> Can somebody tell me why virsh doesn't work? And how I can fix it? >>> >>> Any helps would be very appreciated? >>> >>> Thanks >>> >>> Zhentao >>> >>> _______________________________________________ >>> Users mailing list >>> Users(a)lists.opennebula.org >>> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org >>> >>> >> >> >> _______________________________________________ >> Users mailing list >> Users(a)lists.opennebula.org >> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org >> >> > > > > -- > Shi Jin, Ph.D. >

2 1

Re: [libvirt] Creating Guest OS in Virtual Box
by Pritesh Kothari 05 Oct '09

05 Oct '09

> >> > >> <os> > >> <type>hvm</type> > >> <boot dev='hd'/> > >> </os> > >> > >> <memory>542720</memory> > >> <vcpu>1</vcpu> > >> > >> <devices> > >> <disk type='file' device='disk'> > >> <source file='/home/santhosh/Desktop/ubuntu9.vdi'/> > >> <target dev='hdd'/> ok the actual error is not in the code and the patch failed cause you are using the git HEAD which already has it. The error actually is that you forgot to specify the target type, i.e the target tag should be: <target dev='hdd' bus='ide'/> and not the one which you specified. the documents on the libvirt.org are pretty much old i guess and someone should update them someday ;). Regards, Pritesh

1 0

[libvirt] getting notified of events
by Dan Kenigsberg 05 Oct '09

05 Oct '09

I am trying to understand how to use the virEvent* API. I've built latest libvirt (with "Fix emission of domain events messages" commit) and run the domain-events example. events-c gives me mostly silence. nothing is printed when a domain changes state. events-python is more bizarre: it prints a line on every event - except for the first one, where "Updating Timeout 0 0 Poll sleep 1" is printed. Only when the second event occurs, does the callback for the first event gets executed. What am I missing? Could someone more familiar take a look? Regards, Dan.

2 1

[libvirt] Safe? /etc/libvirt/qemu on shared storage?
by James Brackinshaw 05 Oct '09

05 Oct '09

Hello, I have a directory /mnt/shared-storage where the disk files for kvm live. Live migration works well, but I have a problem: after a migration I see two guests, one on the original host "shutoff" and one on the new host "running". I'd like to see the guest only on the host it is running on. Would it be safe/recommended to put the /etc/libvirt/qemu directory on shared storage? What would the best way of doing this be? Is this lock safe? Thanks JB

5 7

[libvirt] Update libvirt API and refs XML files
by Matthias Bolte 02 Oct '09

02 Oct '09

Commit 47c8709564ac3f2ec5b4a3830ac213e69b4a44d5 "Fix up a few typos in the tree" triggered changes in docs/libvirt-api.xml and docs/libvirt-refs.xml. The attached patch contains this changes. The files docs/libvirt-api.xml and docs/libvirt-refs.xml are generated by apibuild.py, why are this generated files under version control? Matthias

6 8

[libvirt] Performance statistics
by James Brackinshaw 02 Oct '09

02 Oct '09

Hello, Is it possible to get at the (historical) performance data provided by libvirtd? Is it stored for a period of time, or only available to clients polling it? Thanks JB

3 3

[libvirt] [PATCH 00/11] Generic data stream handling
by Daniel P. Berrange 02 Oct '09

02 Oct '09

The following series of patches introduce support for generic data streams in the libvirt API, the remote protocol, client & daemon. The public API takes the form of a new object virStreamPtr and methods to read/write/close it The remote protocol was the main hard bit. Since the protocol allows multiple concurrent API calls on a single connection, this needed to also allow concurrent data streams. It is also not acceptable for a large data stream to block other traffic while it is transferring. Thus, we introduce a new protocol message type 'REMOTE_STREAM' to handle transfer for the stream data. A method involving a data streams starts off in the normal way with a REMOTE_CALL to the server, and a REMOTE_REPLY response message. If this was successful, there now follows the data stream traffic. For outgoing streams (data from client to server), the client will send zero or more REMOTE_STREAM packets containing the data with status == REMOTE_CONTINUE. These are asynchronous and not acknowledged by the server. At any time the server may send an async message with a type of REMOTE_STREAM and status of REMOTE_ERROR. This indicates to the client that the transfer is aborting at server request. If the client wishes to abort, it can send the server a REMOTE_STREAM+REMOTE_ERROR message. If the client finishes its data transfer, it will send a final REMOTE_STREAM+REMOTE_OK message, and the server will respond with the same. This full roundtrip handshake ensures any async error messages are guarenteed to be flushed For incoming data streams (data from server to client), the server sends zero or more REMOTE_STREAM packets containing the data with status == REMOTE_CONTINUE. These are asynchronous and not acknowledged by the client. At any time the client may send an async message with a type of REMOTE_STREAM and status of REMOTE_ERROR. This indicates to the server that the transfer is aborting at client request. If the server wishes to abort, it can send the server a REMOTE_STREAM+REMOTE_ERROR message. When the server finishes its data transfer, it will send a final REMOTE_STREAM+REMOTE_CONTINUE message ewith a data length of zero (ie EOF). The client will then send a REMOTE_STREAM+REMOTE_OK packet and the server will respond with the same. This full roundtrip handshake ensures any async error messages are guarenteed to be flushed This all ensures that multiple data streams can be active in parallel, and with a maximum data packet size of 256 KB, no single stream can cause too much latency on the connection for other API calls/streams. The only thing it does not allow for is one API method to use two or more streams. These may be famous last words, but I don't think that use case will be neccessary for any of our APIs... The last 5 patches with a subject of [DEMO] are *NOT* intended to be committed to the repository. They merely demonstrate the use of data streams for a couple of hypothetical file upload and download APIs. Actually they were mostly to allow me to test the code streams code without messing around with the QEMU migration code. The immediate use case for this data stream code is Chris' QEMU migration patchset. The next use case is to allow serial console access to be tunnelled over libvirtd, eg to make 'virsh console GUEST' work remotely. This use case is why I included the support for non-blocking data streams and event loop integration (not required for Chris' migration use case) Anyway, assuming Chris confirms that I've not broken his code, then patches 1-6 are targetted for this next release. .x-sc_avoid_write | 3 include/libvirt/libvirt.h | 101 ++++ include/libvirt/libvirt.h.in | 101 ++++ po/POTFILES.in | 1 qemud/Makefile.am | 1 qemud/dispatch.c | 131 ++++++ qemud/dispatch.h | 17 qemud/event.c | 54 +- qemud/qemud.c | 56 +- qemud/qemud.h | 33 + qemud/remote.c | 237 ++++++++++- qemud/remote.h | 1 qemud/remote_dispatch_args.h | 2 qemud/remote_dispatch_prototypes.h | 153 +++++++ qemud/remote_dispatch_table.h | 10 qemud/remote_generate_stubs.pl | 1 qemud/remote_protocol.c | 18 qemud/remote_protocol.h | 20 qemud/remote_protocol.x | 57 ++ qemud/stream.c | 605 +++++++++++++++++++++++++++++ qemud/stream.h | 53 ++ src/datatypes.c | 59 ++ src/datatypes.h | 33 + src/driver.h | 40 + src/esx/esx_driver.c | 2 src/libvirt.c | 769 +++++++++++++++++++++++++++++++++++++ src/libvirt_private.syms | 2 src/libvirt_public.syms | 18 src/lxc_driver.c | 2 src/opennebula/one_driver.c | 2 src/openvz_driver.c | 2 src/qemu_driver.c | 2 src/remote_internal.c | 601 ++++++++++++++++++++++++++++ src/test.c | 477 ++++++++++++++++++++++ src/uml_driver.c | 2 src/vbox/vbox_tmpl.c | 3 src/virsh.c | 164 +++++++ src/xen_unified.c | 2 tests/eventtest.c | 19 39 files changed, 3783 insertions(+), 71 deletions(-) Regards, Daniel

5 43

[libvirt] [PATCH 1/2] Fix ordering of <exports> in API description file
by Daniel P. Berrange 02 Oct '09

02 Oct '09

* docs/apibuild.py: Make uniq() function sort keys before returning them to ensure consitent ordering of <exports> in XML file --- docs/apibuild.py | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/docs/apibuild.py b/docs/apibuild.py index b00619f..c5257f3 100755 --- a/docs/apibuild.py +++ b/docs/apibuild.py @@ -55,7 +55,9 @@ def uniq(items): d = {} for item in items: d[item]=1 - return d.keys() + k = d.keys() + k.sort() + return k class identifier: def __init__(self, name, header=None, module=None, type=None, lineno = 0, -- 1.6.2.5

2 3

[libvirt] Tunnelled migration
by Chris Lalancette 02 Oct '09

02 Oct '09

At long last, now that the stream stuff is committed to the tree, I can repost the tunnelled migration patches. Note that these patches will conflict in some places with danpb's previously posted "PEER2PEER" patches, but it should be easy to resolve the differences. V2: Rework the monitor patch to be an API Use the new monitor API Remove silly use of qemuEscapeMonitorArgs()

2 6

[libvirt] Silently ignored virDomainRestore failures
by Charles Duffy 02 Oct '09

02 Oct '09

Howdy, all. I maintain a test infrastructure which makes heavy use of virDomainSave and virDomainRestore, and have been seeing occasional cases where my saved images are for some reason not restored correctly -- and, indeed, the incoming migration streams are not even read in their entirety. While this generally appears to be caused by issues outside of libvirt's purview, one unfortunate issue is that libvirt can report success performing a restore even when the operation is effectively an abject failure. Consider the following snippet, taken from one of my /var/log/libvirt/qemu/<domain>.log files: LC_ALL=C PATH=/sbin:/usr/sbin:/bin:/usr/bin USER=root LOGNAME=root /usr/bin/qemu-kvm -S -M pc-0.11 -m 512 -smp 1 <...lots of arguments here...> -incoming exec:cat cat: write error: Broken pipe This leaves a running qemu hosting a catatonic guest -- but the libvirt client (connecting through the Python bindings) received a status of success for the operation given here. libvirt's mechanism for validating a successful restore consists of running a "cont" command on the guest, and then checking virGetLastError(); AIUI, it is expected that the "cont" will not be able to run until the restore is completed, as the monitor should not be responsive until that time. Browsing through qemudMonitorSendCont (and qemudMonitorCommandWithHandler, which it calls), I don't see anything which looks at the log file with the stderr output from qemu to determine whether an error actually occurred. (As an aside, "info history" on the guest's monitor socket indicates that it was indeed issued this "cont"). Should the existing cont+virGetLastError() approach be sufficient to handle this class of error? If not, is there any guidance on what would comprise a better system? (I suppose we could add something to the exec: to affirmatively indicate on stderr that the decompressor [or cat, if not using one] exited successfully, and check for that marker in the log file... but that seems quite a dirty hack). Thanks!

2 7

[libvirt] [PATCH 00/27] Refactor QEMU monitor command handling
by Daniel P. Berrange 01 Oct '09

01 Oct '09

In the QEMU driver source code the methods which talk to the QEMU monitor currently all just call qemudMonitorCommand() directly with the raw command string, and then parse the raw reply. In the not too distant future QEMU is introducing a new machine parsable monitor syntax. With the current way the code is structured supporting this new mode will be seriously unpleasant. This large series of patches, moves all the monitor command formatting and parsing code out into a separate source file src/qemu/qemu_monitor_text.c. There is one API in that file for each logical monitor command we wish to issue, accepting (mostly) strongly typed arguments. The exception is the NIC hotplug method which still takes the raw NIC string for now. The main qemu_driver.c file now directly calls the appropriate monitor command APIs, making logic there much cleaner. When we add support for the new QEMU monitor syntax we'll gain another file src/qemu/qemu_monitor_json.c which implements all the same APIs as src/qemu/qemu_monitor_text.c, but using the new JSON syntax instead of raw text strings This patch series is soooooo large, because I did it in many small steps, one command at a time. It is also now much easier to debug the monitor by just setting the env variables LIBVIRT_LOG_OUTPUTS="1:stderr" LIBVIRT_LOG_FILTERS="1:qemu_monitor" And you'll get the command & reply of each monitor interaction printed I've tested basic handling of every new method with the exception of the migration ones, since I don't have a convenient target host when on my laptop. Overall we get a small increase in code size, but huge increase in readability ! Makefile.am | 1 libvirt_private.syms | 1 qemu/qemu_conf.c | 20 qemu/qemu_conf.h | 2 qemu/qemu_driver.c | 1669 +++++--------------------------------------- qemu/qemu_monitor_text.c | 1777 +++++++++++++++++++++++++++++++++++++++++++++++ qemu/qemu_monitor_text.h | 173 ++++ 7 files changed, 2171 insertions(+), 1472 deletions(-) Daniel

4 63

[libvirt] Re: [virt-tools-list] Some VMs not starting at host boot
by Cole Robinson 01 Oct '09

01 Oct '09

On 10/01/2009 11:07 AM, Daniel Huhardeaux wrote: > Hi, > > 3 VMs on a host, all in autostart: the both where > > <disk type='block' device='disk'> > <source dev='/dev/vg0/DomU1'/> > <target dev='vda' bus='virtio'/> > </disk> > > are not started automagically on host startup. They start without > problem from virsh. The one having > > <disk type='file' device='disk'> > <source file='/media/vm2/vdisk.qcow2'/> > <target dev='hda' bus='ide'/> > </disk> > > has no problem. Any idea why? FYI, libvirt/virsh specific questions should go to libvirt-list(a)redhat.com. CCing Try 'virsh dominfo <vmname>' for each VM to ensure autostart is actually enabled. To see why autostart is failing, you want to launch libvirtd by hand. Stop all your VMs, do a service libvirtd stop, then (as root) run 'libvirtd'. Any VM startup error messages should be printed on the console. - Cole

2 1

[libvirt] [PATCH] Fix build in separate build directory
by Jiri Denemark 01 Oct '09

01 Oct '09

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- cfg.mk | 2 +- examples/domain-events/events-c/Makefile.am | 2 +- examples/hellolibvirt/Makefile.am | 2 +- tools/Makefile.am | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/cfg.mk b/cfg.mk index c217332..44cfd17 100644 --- a/cfg.mk +++ b/cfg.mk @@ -101,7 +101,7 @@ sc_prohibit_VIR_ERR_NO_MEMORY: msg='use virReportOOMError, not V'IR_ERR_NO_MEMORY \ $(_prohibit_regexp) -include Makefile.nonreentrant +include $(srcdir)/Makefile.nonreentrant sc_prohibit_nonreentrant: @fail=0 ; \ for i in $(NON_REENTRANT) ; \ diff --git a/examples/domain-events/events-c/Makefile.am b/examples/domain-events/events-c/Makefile.am index 60b1589..64457c2 100644 --- a/examples/domain-events/events-c/Makefile.am +++ b/examples/domain-events/events-c/Makefile.am @@ -1,4 +1,4 @@ -INCLUDES = -I@top_srcdir@/include +INCLUDES = -I$(top_builddir)/include -I$(top_srcdir)/include noinst_PROGRAMS = event-test event_test_CFLAGS = $(WARN_CFLAGS) event_test_SOURCES = event-test.c diff --git a/examples/hellolibvirt/Makefile.am b/examples/hellolibvirt/Makefile.am index 6ef2cc8..8d108a7 100644 --- a/examples/hellolibvirt/Makefile.am +++ b/examples/hellolibvirt/Makefile.am @@ -1,4 +1,4 @@ -INCLUDES = -I@top_srcdir@/include +INCLUDES = -I$(top_builddir)/include -I$(top_srcdir)/include noinst_PROGRAMS = hellolibvirt hellolibvirt_CFLAGS = $(WARN_CFLAGS) hellolibvirt_SOURCES = hellolibvirt.c diff --git a/tools/Makefile.am b/tools/Makefile.am index c48b203..941e93e 100644 --- a/tools/Makefile.am +++ b/tools/Makefile.am @@ -43,7 +43,7 @@ virsh_LDADD = \ $(VIRSH_LIBS) virsh_CFLAGS = \ -I$(top_srcdir)/gnulib/lib -I../gnulib/lib \ - -I../include \ + -I../include -I$(top_srcdir)/include \ -I$(top_srcdir)/src \ -I$(top_srcdir)/src/util \ -DGETTEXT_PACKAGE=\"$(PACKAGE)\" \ -- 1.6.5.rc2

2 1

[libvirt] [PATCH] Fix documentation and comment typos
by Paolo Bonzini 01 Oct '09

01 Oct '09

Fix a few mispellings :-) of "successfully" and regenerate docs/libvirt-*.xml. The regeneration also includes the changes in 47c8709. * src/libvirt.c: Fix typos. * src/secret/secret_driver.c: Fix typos. * docs/libvirt-api.xml: Regenerate. * docs/libvirt-refs.xml: Regenerate. --- docs/libvirt-api.xml | 6 +++--- docs/libvirt-refs.xml | 10 +++------- src/libvirt.c | 4 ++-- src/secret/secret_driver.c | 4 ++-- 4 files changed, 10 insertions(+), 14 deletions(-) diff --git a/docs/libvirt-api.xml b/docs/libvirt-api.xml index 795701c..88dc246 100644 --- a/docs/libvirt-api.xml +++ b/docs/libvirt-api.xml @@ -2847,7 +2847,7 @@ API looks like ...report an error... virStreamFree(st); close(fd);]]></info> - <return type='int' info='0 if all the data was succesfully received. The caller should invoke virStreamFinish(st) to flush the stream upon success and then virStreamFree Returns -1 upon any error, with virStreamAbort() already having been called, so the caller need only call virStreamFree()'/> + <return type='int' info='0 if all the data was successfully received. The caller should invoke virStreamFinish(st) to flush the stream upon success and then virStreamFree Returns -1 upon any error, with virStreamAbort() already having been called, so the caller need only call virStreamFree()'/> <arg name='stream' type='virStreamPtr' info='pointer to the stream object'/> <arg name='handler' type='virStreamSinkFunc' info='sink callback for writing data to application'/> <arg name='opaque' type='void *' info='application defined data'/> @@ -2864,7 +2864,7 @@ the caller no longer needs the reference to this object.]]></info> <info><![CDATA[Write a series of bytes to the stream. This method may block the calling application for an arbitrary amount of time. Once an application has finished sending data -it should call virStreamFinish to wait for succesful +it should call virStreamFinish to wait for successful confirmation from the driver, or detect any error This method may not be used if a stream source has been @@ -2939,7 +2939,7 @@ API looks like ...report an error... virStreamFree(st); close(fd);]]></info> - <return type='int' info='0 if all the data was succesfully sent. The caller should invoke virStreamFinish(st) to flush the stream upon success and then virStreamFree Returns -1 upon any error, with virStreamAbort() already having been called, so the caller need only call virStreamFree()'/> + <return type='int' info='0 if all the data was successfully sent. The caller should invoke virStreamFinish(st) to flush the stream upon success and then virStreamFree Returns -1 upon any error, with virStreamAbort() already having been called, so the caller need only call virStreamFree()'/> <arg name='stream' type='virStreamPtr' info='pointer to the stream object'/> <arg name='handler' type='virStreamSourceFunc' info='source callback for reading data from application'/> <arg name='opaque' type='void *' info='application defined data'/> diff --git a/docs/libvirt-refs.xml b/docs/libvirt-refs.xml index 428f847..08034e4 100644 --- a/docs/libvirt-refs.xml +++ b/docs/libvirt-refs.xml @@ -7176,21 +7176,17 @@ <ref name='virNetworkCreate'/> <ref name='virStreamEventUpdateCallback'/> </word> - <word name='succesful'> - <ref name='virStreamSend'/> - </word> - <word name='succesfully'> - <ref name='virStreamRecvAll'/> - <ref name='virStreamSendAll'/> - </word> <word name='successful'> <ref name='virDomainMigrate'/> <ref name='virDomainSave'/> <ref name='virNodeDeviceCreateXML'/> + <ref name='virStreamSend'/> </word> <word name='successfully'> <ref name='virConnectOpen'/> <ref name='virStreamFinish'/> + <ref name='virStreamRecvAll'/> + <ref name='virStreamSendAll'/> </word> <word name='such'> <ref name='virConnCopyLastError'/> diff --git a/src/libvirt.c b/src/libvirt.c index 4cc19ec..dacf8c4 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -9638,7 +9638,7 @@ error: * virStreamFree(st); * close(fd); * - * Returns 0 if all the data was succesfully sent. The caller + * Returns 0 if all the data was successfully sent. The caller * should invoke virStreamFinish(st) to flush the stream upon * success and then virStreamFree * @@ -9735,7 +9735,7 @@ cleanup: * virStreamFree(st); * close(fd); * - * Returns 0 if all the data was succesfully received. The caller + * Returns 0 if all the data was successfully received. The caller * should invoke virStreamFinish(st) to flush the stream upon * success and then virStreamFree * diff --git a/src/secret/secret_driver.c b/src/secret/secret_driver.c index 9c4102e..d61c24a 100644 --- a/src/secret/secret_driver.c +++ b/src/secret/secret_driver.c @@ -751,7 +751,7 @@ secretDefineXML(virConnectPtr conn, const char *xml, if (secretDeleteSaved(conn, driver, secret) < 0) goto restore_backup; } - /* Saved succesfully - drop old values */ + /* Saved successfully - drop old values */ new_attrs = NULL; virSecretDefFree(backup); @@ -843,7 +843,7 @@ secretSetValue(virSecretPtr obj, const unsigned char *value, if (secretSaveValue(obj->conn, driver, secret) < 0) goto restore_backup; } - /* Saved succesfully - drop old value */ + /* Saved successfully - drop old value */ if (old_value != NULL) { memset(old_value, 0, old_value_size); VIR_FREE(old_value); -- 1.6.2.5

2 1

[libvirt] [PATCH] Allows for a <description> tag for domains
by Daniel Veillard 01 Oct '09

01 Oct '09

https://bugzilla.redhat.com/show_bug.cgi?id=523639 feature request which makes sense to me, the simple patch attached seems to be sufficient, one can define and have the description back in the dump. Doesn't try to keep the location of the tag, it always get serialized after <uuid>. The only drawbacks I can think of are: - others XML formats may require the same, but honnestly it's trivial - machine generated description (for example if the history log of a domain gets stored there) could grow a lot and I wonder if we have a hard limit on the size when transmitting xml descriptions Oh and I didn't found a good place to test the feature, i.e. no test seems to parse and reserialize XML it's always about conversions. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

4 8

[libvirt] [PATCH] Avoid a libvirtd crash on broken input 523418
by Daniel Veillard 01 Oct '09

01 Oct '09

A simple typo in an XML domain file could lead to a crash, because we called STRPREFIX() on the looked up value without checking it was non-null. Patch is trivial, and should go in but I wonder if we shouldn't make those STR macros safer too like turning #define STRPREFIX(a,b) (strncmp((a),(b),strlen((b))) == 0) into #define STRPREFIX(a,b) ((a) && (b) && (strncmp((a),(b),strlen((b))) == 0)) anyway I think we should apply the patch independently, without it libvirtd crash, with it we get the proper error report: [root@paphio ~]# virsh define RHEL-5.4-64.xml 18:32:21.982: warning : processCallDispatchReply:7570 : Method call error error: Failed to define domain from RHEL-5.4-64.xml error: internal error No <source> 'dev' attribute specified with <interface type='bridge'/> Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] [PATCH] Fix USB device re-labelling
by Mark McLoughlin 01 Oct '09

01 Oct '09

A simple misplaced break out of a switch results in: libvir: error : Failed to open file '/sys/bus/pci/devices/0000:00:54c./vendor': No such file or directory libvir: error : Failed to open file '/sys/bus/pci/devices/0000:00:54c./device': No such file or directory libvir: error : this function is not supported by the hypervisor: Failed to read product/vendor ID for 0000:00:54c. when trying to passthrough a USB host device to qemu. * src/security_selinux.c: fix a switch/break thinko --- src/security/security_selinux.c | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/src/security/security_selinux.c b/src/security/security_selinux.c index bdd9eb2..d08d502 100644 --- a/src/security/security_selinux.c +++ b/src/security/security_selinux.c @@ -492,12 +492,11 @@ SELinuxSetSecurityHostdevLabel(virConnectPtr conn, ret = usbDeviceFileIterate(conn, usb, SELinuxSetSecurityUSBLabel, vm); usbFreeDevice(conn, usb); - - break; } else { /* XXX deal with product/vendor better */ ret = 0; } + break; } case VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI: { -- 1.6.2.5

2 1

[libvirt] [OT] Relocation lookup
by Rakotomandimby Mihamina 01 Oct '09

01 Oct '09

Hi all, Given the Social and Political reality in my country (Madagascar), I am obliged to look for a relocation. This is my public profile: http://www.linkedin.com/in/mihaminarakotomandimby Would you be aware of a position I could fit in? Thank you. -- Architecte Informatique chez Blueline/Gulfsat: Administration Systeme, Recherche & Developpement +261 34 29 155 34

1 0

[libvirt] [PATCH] incorrect error message in virDomainNetDefParseXML
by Florian Vichot 01 Oct '09

01 Oct '09

When declaring a <interface type="bridge"> tag, <source> needs a "bridge" attribute, but the parser complains about a missing "dev" attribute. Florian

3 4

[libvirt] "virsh destroy" deletes extra VMs
by Shi Jin 30 Sep '09

30 Sep '09

Hi there, I am running Ubuntu 9.04 X64 server with KVM/libvirt. I recently found that the "virsh destroy" command actually delete VMs it should not delete. For example, please see the following screen copy: root@node1:/srv/cloud/ImgRep/WinXP# virsh list Connecting to uri: qemu:///system Id Name State ---------------------------------- 27 one-80 running 28 one-81 running 29 one-83 running 30 one-82 running 31 one-79 running root@node1:/srv/cloud/ImgRep/WinXP# virsh destroy 29 Connecting to uri: qemu:///system Domain 29 destroyed root@node1:/srv/cloud/ImgRep/WinXP# virsh list Connecting to uri: qemu:///system Id Name State ---------------------------------- 27 one-80 running 28 one-81 running 31 one-79 running Please note that VM 30 was deleted when we tried to delete VM 29. The /var/log/messages showed the following information: ep 30 08:58:59 node1 kernel: [238101.304619] br1: port 5(vnet4) entering disabled state Sep 30 08:58:59 node1 kernel: [238101.342851] device vnet4 left promiscuous mode Sep 30 08:58:59 node1 kernel: [238101.342854] br1: port 5(vnet4) entering disabled state Sep 30 08:58:59 node1 kernel: [238101.424851] br1: port 4(vnet3) entering disabled state Sep 30 08:59:00 node1 kernel: [238101.463031] device vnet3 left promiscuous mode Sep 30 08:59:00 node1 kernel: [238101.463035] br1: port 4(vnet3) entering disabled state Sep 30 09:10:53 node1 -- MARK -- Sep 30 09:30:53 node1 -- MARK -- Sep 30 09:50:53 node1 -- MARK -- Sep 30 09:51:31 node1 kernel: [241253.266955] device vnet3 entered promiscuous mode Sep 30 09:51:31 node1 kernel: [241253.268857] br1: port 4(vnet3) entering learning state Sep 30 09:51:40 node1 kernel: [241262.260147] br1: topology change detected, propagating Sep 30 09:51:40 node1 kernel: [241262.260152] br1: port 4(vnet3) entering forwarding state Sep 30 09:53:59 node1 kernel: [241401.204570] br1: port 4(vnet3) entering disabled state Sep 30 09:53:59 node1 kernel: [241401.232395] device vnet3 left promiscuous mode Sep 30 09:53:59 node1 kernel: [241401.232399] br1: port 4(vnet3) entering disabled state Sep 30 09:54:20 node1 kernel: [241421.694803] br1: port 2(vnet0) entering disabled state Sep 30 09:54:20 node1 kernel: [241421.733095] device vnet0 left promiscuous mode Sep 30 09:54:20 node1 kernel: [241421.733099] br1: port 2(vnet0) entering disabled state Sep 30 09:54:20 node1 kernel: [241421.815209] br2: port 2(vnet1) entering disabled state Sep 30 09:54:20 node1 kernel: [241421.853728] device vnet1 left promiscuous mode Sep 30 09:54:20 node1 kernel: [241421.853732] br2: port 2(vnet1) entering disabled state Sep 30 09:54:20 node1 kernel: [241421.935254] br1: port 3(vnet2) entering disabled state Sep 30 09:54:20 node1 kernel: [241421.963527] device vnet2 left promiscuous mode Sep 30 09:54:20 node1 kernel: [241421.963531] br1: port 3(vnet2) entering disabled state Sep 30 09:55:29 node1 kernel: [241490.644204] br1: port 6(vnet5) entering disabled state Sep 30 09:55:29 node1 kernel: [241490.673195] device vnet5 left promiscuous mode Sep 30 09:55:29 node1 kernel: [241490.673199] br1: port 6(vnet5) entering disabled state Sep 30 10:10:53 node1 -- MARK -- Sep 30 10:11:49 node1 kernel: [242470.955142] device vnet0 entered promiscuous mode Sep 30 10:11:49 node1 kernel: [242470.957051] br1: port 2(vnet0) entering learning state Sep 30 10:11:52 node1 kernel: [242474.180036] device vnet1 entered promiscuous mode Sep 30 10:11:52 node1 kernel: [242474.181943] br1: port 3(vnet1) entering learning state Sep 30 10:11:55 node1 kernel: [242477.431693] device vnet2 entered promiscuous mode Sep 30 10:11:55 node1 kernel: [242477.433558] br1: port 4(vnet2) entering learning state Sep 30 10:11:58 node1 kernel: [242479.950082] br1: topology change detected, propagating Sep 30 10:11:58 node1 kernel: [242479.950086] br1: port 2(vnet0) entering forwarding state Sep 30 10:12:00 node1 kernel: [242482.059146] device vnet3 entered promiscuous mode Sep 30 10:12:00 node1 kernel: [242482.060330] br1: port 5(vnet3) entering learning state Sep 30 10:12:01 node1 kernel: [242483.180170] br1: topology change detected, propagating Sep 30 10:12:01 node1 kernel: [242483.180173] br1: port 3(vnet1) entering forwarding state Sep 30 10:12:04 node1 kernel: [242485.710948] device vnet4 entered promiscuous mode Sep 30 10:12:04 node1 kernel: [242485.712109] br1: port 6(vnet4) entering learning state Sep 30 10:12:04 node1 kernel: [242486.430098] br1: topology change detected, propagating Sep 30 10:12:04 node1 kernel: [242486.430104] br1: port 4(vnet2) entering forwarding state Sep 30 10:12:09 node1 kernel: [242491.051857] br1: topology change detected, propagating Sep 30 10:12:09 node1 kernel: [242491.051862] br1: port 5(vnet3) entering forwarding state Sep 30 10:12:13 node1 kernel: [242494.711792] br1: topology change detected, propagating Sep 30 10:12:13 node1 kernel: [242494.711797] br1: port 6(vnet4) entering forwarding state Sep 30 10:22:00 node1 kernel: [243082.274008] br1: port 4(vnet2) entering disabled state Sep 30 10:22:00 node1 kernel: [243082.312108] device vnet2 left promiscuous mode Sep 30 10:22:00 node1 kernel: [243082.312112] br1: port 4(vnet2) entering disabled state Sep 30 10:22:00 node1 kernel: [243082.365331] br1: port 5(vnet3) entering disabled state Sep 30 10:22:00 node1 kernel: [243082.393430] device vnet3 left promiscuous mode Sep 30 10:22:00 node1 kernel: [243082.393433] br1: port 5(vnet3) entering disabled state I have 4 NICs and the ifconfig shows root@node1:/srv/cloud/ImgRep/WinXP# ifconfig bond0 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57 inet6 addr: fe80::224:e8ff:fe6b:d57/64 Scope:Link UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1 RX packets:7799931 errors:0 dropped:0 overruns:0 frame:0 TX packets:6879357 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:4360166425 (4.3 GB) TX bytes:4350243659 (4.3 GB) br0 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57 inet addr:192.168.1.128 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::224:e8ff:fe6b:d57/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:7641913 errors:0 dropped:0 overruns:0 frame:0 TX packets:6412260 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:4169876739 (4.1 GB) TX bytes:4291182908 (4.2 GB) br1 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5b inet6 addr: fe80::224:e8ff:fe6b:d5b/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:152606 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:11276026 (11.2 MB) TX bytes:468 (468.0 B) br2 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5d inet6 addr: fe80::224:e8ff:fe6b:d5d/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:152596 errors:0 dropped:0 overruns:0 frame:0 TX packets:6 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:11286132 (11.2 MB) TX bytes:468 (468.0 B) eth0 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1 RX packets:3622205 errors:0 dropped:0 overruns:0 frame:0 TX packets:3443103 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1758147112 (1.7 GB) TX bytes:2174389369 (2.1 GB) Interrupt:36 Memory:d6000000-d6012700 eth1 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57 UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1 RX packets:4177726 errors:0 dropped:0 overruns:0 frame:0 TX packets:3436254 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:2602019313 (2.6 GB) TX bytes:2175854290 (2.1 GB) Interrupt:48 Memory:d8000000-d8012700 eth2 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5b inet6 addr: fe80::224:e8ff:fe6b:d5b/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:256014 errors:0 dropped:0 overruns:0 frame:0 TX packets:91549 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:20680139 (20.6 MB) TX bytes:17317087 (17.3 MB) Interrupt:32 Memory:da000000-da012700 eth3 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5d inet6 addr: fe80::224:e8ff:fe6b:d5d/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:245189 errors:0 dropped:0 overruns:0 frame:0 TX packets:32669 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:20866257 (20.8 MB) TX bytes:8265329 (8.2 MB) Interrupt:42 Memory:dc000000-dc012700 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:324787 errors:0 dropped:0 overruns:0 frame:0 TX packets:324787 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:571707504 (571.7 MB) TX bytes:571707504 (571.7 MB) vnet0 Link encap:Ethernet HWaddr 52:c3:a3:d0:09:84 inet6 addr: fe80::50c3:a3ff:fed0:984/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:72 errors:0 dropped:0 overruns:0 frame:0 TX packets:533 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:500 RX bytes:10410 (10.4 KB) TX bytes:49306 (49.3 KB) vnet1 Link encap:Ethernet HWaddr 02:0d:4c:1e:5c:82 inet6 addr: fe80::d:4cff:fe1e:5c82/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:36 errors:0 dropped:0 overruns:0 frame:0 TX packets:563 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:500 RX bytes:6023 (6.0 KB) TX bytes:53245 (53.2 KB) vnet4 Link encap:Ethernet HWaddr a6:58:67:2b:37:8a inet6 addr: fe80::a458:67ff:fe2b:378a/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:42 errors:0 dropped:0 overruns:0 frame:0 TX packets:552 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:500 RX bytes:6733 (6.7 KB) TX bytes:52189 (52.1 KB) Note that there is no IP address assigned to the br1 device which is used here. I don't think that should be a problem. Also the first 2 NICs are bonded first and then created as a bridge br0 and I don't think it is a problem either. root@node1:/srv/cloud/ImgRep/WinXP# virsh --version Connecting to uri: qemu:///system 0.6.1 root@node1:/srv/cloud/ImgRep/WinXP# kvm -h QEMU PC emulator version 0.9.1 (kvm-84), Copyright (c) 2003-2008 Fabrice Bellard Has anyone had similar problem before? How can I fix it? Thanks a lot. Shi -- Shi Jin, PhD

2 2

[libvirt] [PATCH] fixes to allow 'make rpm' to work again
by Daniel Veillard 30 Sep '09

30 Sep '09

Just a few oversights: - examples/domain-events has no makefile, but domain-events/events-c does - we should still add examples/domain-events/events-python to the tarballs in make dist - python/libvirtclass.txt seems to have vanished from dist and builds not a big deal but don't assume it's there in the spec file. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 3

[libvirt] [PATCH] Fix emission of domain events messages
by Daniel P. Berrange 30 Sep '09

30 Sep '09

The code which updated the message length after writing the payload wrote the updated length word in the wrong place since the XDR object was given a buffer pointing to the start of the header payload, rather than message start. * daemon/remote.c: Fix updating of event message length so that we actually send the payload, not just the header --- daemon/remote.c | 14 ++++++++------ 1 files changed, 8 insertions(+), 6 deletions(-) diff --git a/daemon/remote.c b/daemon/remote.c index 6d4cc1c..2b58045 100644 --- a/daemon/remote.c +++ b/daemon/remote.c @@ -4559,17 +4559,19 @@ remoteDispatchDomainEventSend (struct qemud_client *client, /* Serialise the return header and event. */ xdrmem_create (&xdr, - msg->buffer + msg->bufferOffset, - msg->bufferLength - msg->bufferOffset, + msg->buffer, + msg->bufferLength, XDR_ENCODE); - if (!xdr_remote_domain_event_msg(&xdr, data)) + /* Skip over the header we just wrote */ + if (xdr_setpos (&xdr, msg->bufferOffset) == 0) goto xdr_error; + if (!xdr_remote_domain_event_msg(&xdr, data)) + goto xdr_error; - /* Update length word */ - msg->bufferOffset += xdr_getpos (&xdr); - len = msg->bufferOffset; + /* Update length word to include payload*/ + len = msg->bufferOffset = xdr_getpos (&xdr); if (xdr_setpos (&xdr, 0) == 0) goto xdr_error; -- 1.6.2.5

2 3

[libvirt] Tunnelled migration
by Chris Lalancette 30 Sep '09

30 Sep '09

At long last, now that the stream stuff is committed to the tree, I can repost the tunnelled migration patches. Note that these patches will conflict in some places with danpb's previously posted "PEER2PEER" patches, but it should be easy to resolve the differences.

2 15

[libvirt] Fix apibuild.py warnings
by Matthias Bolte 30 Sep '09

30 Sep '09

apibuild.py warns about misformatted function comment for virStreamEventAddCallback, virStreamEventUpdateCallback and virStreamEventRemoveCallback. The attached patch adds the missing ':' and contains the resulting changes in docs/libvirt-api.xml and docs/libvirt-refs.xml Matthias

3 2

[libvirt] [PATCH] minor fix's
by Pritesh Kothari 30 Sep '09

30 Sep '09

Hi All, Just minor fix's. Regards, Pritesh

3 2