- Devel - libvirt List Archives

[libvirt] PATCH: 0/5: API for converting domain XML to/from native config formats
by Daniel P. Berrange 20 May '09

20 May '09

This is a long overdue update to the idea I described here http://www.redhat.com/archives/libvir-list/2009-April/msg00266.html And the initial patch posted here http://www.redhat.com/archives/libvir-list/2009-April/msg00401.html Daniel -- |: 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 :|

3 17

[libvirt] [PATCH] lxc: use new error code in lxcDomainUndefine
by Amy Griffis 20 May '09

20 May '09

Update lxcDomainUndefine() to use VIR_ERR_INVALID_OPERATION instead of VIR_ERR_INTERNAL_ERROR. Signed-off-by: Amy Griffis <amy.griffis(a)hp.com> diff --git a/src/lxc_driver.c b/src/lxc_driver.c index 5392af1..64c735d 100644 --- a/src/lxc_driver.c +++ b/src/lxc_driver.c @@ -329,13 +329,13 @@ static int lxcDomainUndefine(virDomainPtr dom) } if (virDomainIsActive(vm)) { - lxcError(dom->conn, dom, VIR_ERR_INTERNAL_ERROR, + lxcError(dom->conn, dom, VIR_ERR_OPERATION_INVALID, "%s", _("cannot delete active domain")); goto cleanup; } if (!vm->persistent) { - lxcError(dom->conn, dom, VIR_ERR_INTERNAL_ERROR, + lxcError(dom->conn, dom, VIR_ERR_OPERATION_INVALID, "%s", _("cannot undefine transient domain")); goto cleanup; }

2 1

[libvirt] virStorageVolLookupByPath fails with "libvir: Storage error : invalid storage volume pointer in no storage vol with matching path"
by Francesco Latino 19 May '09

19 May '09

Hello, I have a running KVM virtual guest "demo1" with a defined disk at path /guests/mydisk.img. But when I run conn = virConnectOpenReadOnly(NULL); path = "/guests/mydisk.img"; virStorageVolLookupByPath(conn,path); I got the error " ibvir: Storage error : invalid storage volume pointer in no storage vol with matching path " what am I missing ? Thanks, Francesco

2 2

[libvirt] [PATCH] python: Better error reporting if 'import libvirtmod' fails
by Cole Robinson 19 May '09

19 May '09

Don't squash a possibly legitimate libvirtmod error (e.g. some from clashing libvirt.so versions) with 'Cannot import cygvirtmod' --- python/libvir.py | 8 ++++++-- 1 files changed, 6 insertions(+), 2 deletions(-) diff --git a/python/libvir.py b/python/libvir.py index 8a16dd0..3cda8dc 100644 --- a/python/libvir.py +++ b/python/libvir.py @@ -8,8 +8,12 @@ # On cygwin, the DLL is called cygvirtmod.dll try: import libvirtmod -except: - import cygvirtmod as libvirtmod +except ImportError, lib_e: + try: + import cygvirtmod as libvirtmod + except ImportError, cyg_e: + if str(cyg_e).count("No module named"): + raise lib_e import types -- 1.6.2.2

3 23

[libvirt] [PATCH v2 0/2] FS backend VolCreateXMLFrom implementation.
by Cole Robinson 19 May '09

19 May '09

The following 2 patches implement virStorageVolCreateXMLFrom for the storage driver and the FS storage backend. Changes from v1: - Mark origvol as building in driver implementation - Make sure driver is locked when relocking pool object - Simplify sparse file detection with memcmp - Allocate read buffer on the heap. Thanks, Cole Cole Robinson (2): Storage driver implementation for CreateXMLFrom VolumeCreateXMLFrom FS storage backend implementation. src/storage_backend.h | 2 + src/storage_backend_fs.c | 247 +++++++++++++++++++++++++++++++++++++++------- src/storage_driver.c | 159 +++++++++++++++++++++++++++++- 3 files changed, 369 insertions(+), 39 deletions(-)

3 6

[libvirt] libvirt API to write sectors on a device
by Francesco Latino 19 May '09

19 May '09

Hello,***I am able to compile libvirt samples on my KVM hypervisor. (I edited and recompiled info.c code) *I am creating a sample code to read/write disk devices. With method *virDomainBlockPeek I can** read sectors from a given device attached to a KVM virtual guest.**I am not able to* find an API to write sectors on a disk: am I missing something ?Is there such method to write sectors from a given virtual disk ? Thanks,Francesco

2 2

[libvirt] [PATCH 0/5] Interface Config Public APIs and remote stubs, with MAC addresses in string format
by Laine Stump 19 May '09

19 May '09

These patches are based on those I sent last week (the public virInterface*() API definition, the local plumbing, and the RPC glue), but with suggestions incorporated: 1) MAC address is always used in null-terminated ASCII string format. This eliminates any potential problems with extra long addresses. 2) no comments in libvirt.h 3) flags arguments are all unsigned. Item (1) caused changes in most of the patches, so I just regenerated the entire set. Is this the proper/expected thing to do, or should I be making patches of patches?

3 9

[libvirt] PATCH: Enable migration with QEMU >= 0.10.0
by Daniel P. Berrange 19 May '09

19 May '09

The KVM migration code was added to QEMU for the 0.10.0 release, so we should enable this in libvirt now. Daniel diff -r be7993675e07 src/qemu_conf.c --- a/src/qemu_conf.c Thu Apr 30 14:49:27 2009 +0100 +++ b/src/qemu_conf.c Thu Apr 30 15:08:45 2009 +0100 @@ -473,16 +473,13 @@ int qemudExtractVersionInfo(const char * /* * Handling of -incoming arg with varying features - * -incoming tcp (kvm >= 79) - * -incoming exec (kvm >= 80) + * -incoming tcp (kvm >= 79, qemu >= 0.10.0) + * -incoming exec (kvm >= 80, qemu >= 0.10.0) * -incoming stdio (all earlier kvm) * * NB, there was a pre-kvm-79 'tcp' support, but it * was broken, because it blocked the monitor console * while waiting for data, so pretend it doesn't exist - * - * XXX when next QEMU release after 0.9.1 arrives, - * we'll need to add MIGRATE_QEMU_TCP/EXEC here too */ if (kvm_version >= 79) { flags |= QEMUD_CMD_FLAG_MIGRATE_QEMU_TCP; @@ -490,6 +487,9 @@ int qemudExtractVersionInfo(const char * flags |= QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC; } else if (kvm_version > 0) { flags |= QEMUD_CMD_FLAG_MIGRATE_KVM_STDIO; + } else if (version >= 10000) { + flags |= QEMUD_CMD_FLAG_MIGRATE_QEMU_TCP; + flags |= QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC; } if (retversion) -- |: 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 :|

3 8

[libvirt] PATCH: Declare QEMU migration capabilities
by Daniel P. Berrange 18 May '09

18 May '09

This patches adds the missing data in the capabilities XML about the QEMU driver's migration support. Specifically it declares support for both offline and live migrate, and registers the 'tcp' URI scheme. Daniel diff -r 00f2e113eca8 src/qemu_conf.c --- a/src/qemu_conf.c Mon May 18 12:38:18 2009 +0100 +++ b/src/qemu_conf.c Mon May 18 13:15:33 2009 +0100 @@ -371,7 +371,7 @@ virCapsPtr qemudCapsInit(void) { uname (&utsname); if ((caps = virCapabilitiesNew(utsname.machine, - 0, 0)) == NULL) + 1, 1)) == NULL) goto no_memory; /* Using KVM's mac prefix for QEMU too */ @@ -380,6 +380,9 @@ virCapsPtr qemudCapsInit(void) { if (virCapsInitNUMA(caps) < 0) goto no_memory; + virCapabilitiesAddHostMigrateTransport(caps, + "tcp"); + /* First the pure HVM guests */ for (i = 0 ; i < ARRAY_CARDINALITY(arch_info_hvm) ; i++) if (qemudCapsInitGuest(caps, -- |: 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] Domain Event Handlers
by Noah Zoschke 18 May '09

18 May '09

Hello, I have been playing around with the libvirt domain event handlers in Python and am seeing something surprising. I am using trunk libvirt and `event-test.py` with some more of the debugging strings uncommented. When I trigger my first event: virsh # define xppro-sp2-0.img.xml Domain xppro-sp2-0.img defined from xppro-sp2-0.img.xml The timeout and handle callbacks are called, but not my domainEventRegister handler: Invoking Timeout CB Invoking Handle CB When I trigger a second event: virsh # create xppro-sp2-0.img.xml Domain xppro-sp2-0.img created from xppro-sp2-0.img.xml my event handler for the first event is finally called: Invoking Timeout CB domainEventHandler EVENT: Domain xppro-sp2-0.img(-1) Added 1 Invoking Handle CB Likewise for a third event: virsh # destroy xppro-sp2-0.img Domain xppro-sp2-0.img destroyed Invoking Timeout CB domainEventHandler EVENT: Domain xppro-sp2-0.img(1) Started 0 Invoking Handle CB So my handler is always one event behind. This behavior is not intended is it? Am I missing either something in `event-test.py` or in how I am using virsh that would correct this behavior? Thank you, Noah

3 3

[libvirt] Usage of UUID
by Eduardo Otubo 18 May '09

18 May '09

Hello all, When I started to implement phyp driver, I use to handle the UUID as a normal 'unsigned char *' with a LPAR value called "logical_serial_num" that is a string in the format "067BE8B2", an identifier. In the last comments I got from danpb and DV, they called my attention to const string at UUID handling and some funtcions that must be implemented: domainDumpXML, listDefinedDomains and numOfDefinedDomains. My question is: Once I don't have information from my hypervisor about UUID, do I have to handle UUID at "list" or "list --all" commands? Or I should keep and handle this information only when defining a domaing at libvirt? I mean, in my point of view, there is no need to generate and store a UUID while just listing LPARs in the remote server. What I should do is to generate a random UUID (and store it) just when the user decides to define it at libvirt. Am I thinking right? Thanks for the help. []'s -- Eduardo Otubo Software Engineer Linux Technology Center IBM Systems & Technology Group Mobile: +55 19 8135 0885 otubo(a)linux.vnet.ibm.com

2 1

[libvirt] is qemu really needed for make rpm or is kvm sufficient?
by Gerrit Slomma 18 May '09

18 May '09

Hello I am curious if qemu is really needed to do "make rpm". As far as i have seen only the src.rpm has this dependency but neither of the install-rpms. What is this needed for in the src.rpm-package that do lack the install-packages? I haven't installed qemu on my hosts, only kvm and everything works fine, even live-migration works now without the extra qemu installed on the host i build the libvirt. But now "make rpm" fails due to the dependency for qemu. Couldn't either: A) the qemu-requirement be patched out of the requirements for the src.rpm (if possible) or B) the kvm-package has qemu and qemu-img added as "provides" (okay - this would be another mailing list) Gerrit

2 1

[libvirt] patch for virsh list
by Gerrit Slomma 16 May '09

16 May '09

Hello I thought it would be cool if virsh list behaves a bit more like virt-manager or xm list because as a system-administrator i want to know with one look what VMs are active, how much memory they commit and what their runtime is. I implemented this with the following patch which give such an sample output. What is your opinion and/or have i missed something? [root@rr016 ~]# virsh list --all Id Name Status Mem(kB) VCPUs Time(s) --------------------------------------------------------------- 1 rr019v3 laufend 1572864 2 15,3 2 rr019v4 laufend 1572864 2 11,7 - OpenSolaris08.11 ausschalten - Win7b_64 ausschalten --- a/src/virsh.c 2009-04-15 21:43:26.000000000 +0200 +++ b/src/virsh.c 2009-05-15 21:35:53.000000000 +0200 @@ -658,14 +658,16 @@ qsort(&names[0], maxname, sizeof(char*), namesorter); } } - vshPrintExtra(ctl, "%3s %-20s %s\n", _("Id"), _("Name"), _("State")); - vshPrintExtra(ctl, "----------------------------------\n"); + vshPrintExtra(ctl, "%3s %-20s %-15s %-8s %-5s %-7s\n", _("Id"), _("Name"), _("State"), + _("Mem(kB)"), _("VCPUs"), _("Time(s)")); + vshPrintExtra(ctl, "---------------------------------------------------------------\n"); for (i = 0; i < maxid; i++) { virDomainInfo info; virDomainPtr dom = virDomainLookupByID(ctl->conn, ids[i]); const char *state; - + double cpuUsed = 0; + /* this kind of work with domains is not atomic operation */ if (!dom) continue; @@ -675,10 +677,19 @@ else state = N_(vshDomainStateToString(info.state)); - vshPrint(ctl, "%3d %-20s %s\n", + if (info.cpuTime != 0) { + cpuUsed = info.cpuTime; + + cpuUsed /= 1000000000.0; + } + + vshPrint(ctl, "%3d %-20s %-15s %-8d %-5d %7.1f\n", virDomainGetID(dom), virDomainGetName(dom), - state); + state, + virDomainGetMaxMemory(dom), + info.nrVirtCpu, + cpuUsed); virDomainFree(dom); } for (i = 0; i < maxname; i++) {

2 2

[libvirt] USB device with product id 0x0000
by André Welter 16 May '09

16 May '09

Hi everyone! I'm new to this list and I joined because I stumbled upon a problem I could not find mentioned anywhere. I have a usb device (a bluetooth dongle) that reports a product id of 0x0000. When I try to attach this to a domain I get a "missing product" error. I checked the sources (0.6.3) and found in domain_conf.c lines 1645 following a check for a product id different from 0. If I remove this check and recompile everything works fine. I can attach the device to the virtual machine and it is recognized. I don't know the background of this check and if disabling it has any negative consequences so I would be happy about any comments. Regards André

2 1

[libvirt] (no subject)
by Eduardo Otubo 15 May '09

15 May '09

Hello all, I've been working on a libvirt extension to manage IBM's Power VMs (LPARs). The Power systems are managed through management console referred to as the HMC or using a management partition (IVM). Both HMC and IVM runs an SSH, then you can reach it via command line, and an HTTP server, then you can reach it via web browser. The protocol between the console and the partition (LPAR) is not disclosed, therefore I propose the driver to execute commands remoetly over an SSH connection to the consoles to manage IBM LPARs. The patch attached is the first scratch of the driver that will interact with HMC over a SSH connection. The URI model that is used in virsh command line is: virsh --conect phyp://$user@$server Some known issues are: * Next step is to make the URI like this: phyp://$user@ $HMC/@managed_system. Almost finished. What it takes now is $server = $HMC = $managed_system. * Next features in my TODO list are "resume", "stop" and "reboot" the LPAR. Any comments are welcome. Thanks, -- Eduardo Otubo Software Engineer Linux Technology Center IBM Systems & Technology Group Mobile: +55 19 8135 0885 otubo(a)linux.vnet.ibm.com

4 18

[libvirt] [PATCH 3/3] networking API for hostonly networks in VirtualBox driver in libvirt
by Pritesh Kothari 15 May '09

15 May '09

Hi All, As discussed on the list resending the networking patch's. the patch's are as below: [PATCH 1/3]: contains support for "Internal" network in libvirt [PATCH 2/3]: contains support for "Host only" and "Internal" networks in VirtualBox driver [PATCH 3/3]: contains networking API for hostonly networks in VirtualBox driver in libvirt (it contains all the fix's proposed on list along with two extra *DefinedNetworks functions) The patches work as per cvs checkin today. (git SHA1 ID: 9136ae2056b45ea83854d3fe31d860f645b8c883) Regards, Pritesh

3 8

[libvirt] [PATCH 1/1] Support for rdp/desktop types in graphics tag
by Pritesh Kothari 15 May '09

15 May '09

Hi All, As per the discussion earlier on the list I have modified the rdp type and added a new desktop type and posting the patch for same. Regards, Pritesh

3 14

[libvirt] [PATCH 0/2] Implement nodedev destroy API
by David Allan 15 May '09

15 May '09

Here is a patch that implements the node device destroy API call, as well as the suggestions that Dan made in his feedback on the patch implementing create. The patch also makes it not necessary for the caller of the node device create function to specify a name for the device and a host number for the adapter in the XML. The change to do so is kind of kludgy, though, IMO, as it involved changing the function signature of chain of internal function calls to differentiate a device that is being created from an existing device. I'm torn as to whether I think that's a better solution than requiring the caller of the nodedev create API to specify a name that will be ignored. One benefit of requiring the user to specify the name is that it is used in some error messages, so I can argue it both ways. Opinions very welcome. I'd particularly like someone in the Solaris world to make sure that the code builds as there is some #ifdef __linux__ in the patch. I tried to spin up Opensolaris to try it out but ran up against bug 6784591 and a lack of time. I've also tried to make it relatively easy for someone to add support for non-Linux OSes, so a patch there would be appreciated. Dave

3 4

[libvirt] how to configure keymap & keyboard layout with virt-manager
by libvirt＠pivert.org 14 May '09

14 May '09

Hi, I'm fighting to configure the keyboard on libvirt (kvm) (0.6.1-0ubuntu5). I'm using a belgian keyboard, so, my question is : - Which keyboard layout should I place in the xml file (or through the vm display adapter configuration). - Which keyboard layout should I use in the vm ? PS: When I use 'be' as keymap on the vnc line, the libvirtd just segfault at startup. Regards, -- François Delpierre

2 2

[libvirt] PATCH: Fix misc locking bugs, and add lock checker
by Daniel P. Berrange 14 May '09

14 May '09

This patch fixes a number of locking bugs, some serious, some not. It also adds the CIL lock checker I previously wrote. It is improved since last time, because it explicitly looks for the virDriver static variables and uses that to extract list of functions to be checked. This removes a huge number of false positives. It also now checks for a couple of dead lock scenarios, in addition to previous checks for lock ordering correctness. The serious locking bugs - qemudDomainMigratePrepare2: didn't unlock VM causing deadlock - storagePoolRefresh: unlocked the driver too early - storageVolumeCreateXML: locked the pool without having locked driver - umlDomainGetAutostart: missing unlock call with later deadlock The lock checker requires that you can ocaml and ocaml-cil installed. You then need to run configure using '--enable-test-locking' and do a full make clean, followed by make. This is because it has to set some special CFLAGS to disable booleans, and keep temporary files around. Once built, you can run 'tests/object-locking 2>/dev/null' to validate It should not report any errors, with this patch now applied .hgignore | 6 b/tests/object-locking.ml | 828 ++++++++++++++++++++++++++++++++++++++++++++++ configure.in | 16 src/.cvsignore | 2 src/.gitignore | 2 src/Makefile.am | 5 src/network_driver.c | 6 src/qemu_driver.c | 39 -- src/storage_driver.c | 7 src/test.c | 3 src/uml_driver.c | 4 tests/.cvsignore | 4 tests/.gitignore | 4 tests/Makefile.am | 27 + 14 files changed, 920 insertions(+), 33 deletions(-) Daniel diff -r 71a7d1d0ad9b .hgignore --- a/.hgignore Thu May 14 17:38:11 2009 +0100 +++ b/.hgignore Thu May 14 17:47:01 2009 +0100 @@ -235,9 +235,11 @@ src/*.exe src/*.gcda src/*.gcno src/*.gcov +src/*.i src/*.la src/*.lo src/*.loT +src/*.s src/.deps src/.libs src/Makefile @@ -264,6 +266,10 @@ tests/conftest tests/eventtest tests/nodedevxml2xmltest tests/nodeinfotest +tests/object-locking +tests/object-locking-files.txt +tests/object-locking.cmi +tests/object-locking.cmx tests/qemuxml2argvtest tests/qemuxml2xmltest tests/qparamtest diff -r 71a7d1d0ad9b configure.in --- a/configure.in Thu May 14 17:38:11 2009 +0100 +++ b/configure.in Thu May 14 17:47:01 2009 +0100 @@ -1132,6 +1132,22 @@ if test "${enable_oom}" = yes; then AC_DEFINE([TEST_OOM], 1, [Whether malloc OOM checking is enabled]) fi + +AC_ARG_ENABLE([test-locking], +[ --enable-test-locking thread locking tests using CIL], +[case "${enableval}" in + yes|no) ;; + *) AC_MSG_ERROR([bad value ${enableval} for test-locking option]) ;; + esac], + [enableval=no]) +enable_locking=$enableval + +if test "$enable_locking" = "yes"; then + LOCK_CHECKING_CFLAGS="-Dbool=char -D_Bool=char -save-temps" + AC_SUBST([LOCK_CHECKING_CFLAGS]) +fi +AM_CONDITIONAL([WITH_CIL],[test "$enable_locking" = "yes"]) + dnl Enable building the proxy? AC_ARG_WITH([xen-proxy], diff -r 71a7d1d0ad9b src/.cvsignore --- a/src/.cvsignore Thu May 14 17:38:11 2009 +0100 +++ b/src/.cvsignore Thu May 14 17:47:01 2009 +0100 @@ -16,3 +16,5 @@ libvirt_lxc virsh-net-edit.c virsh-pool-edit.c libvirt.syms +*.i +*.s diff -r 71a7d1d0ad9b src/.gitignore --- a/src/.gitignore Thu May 14 17:38:11 2009 +0100 +++ b/src/.gitignore Thu May 14 17:47:01 2009 +0100 @@ -16,3 +16,5 @@ libvirt_lxc virsh-net-edit.c virsh-pool-edit.c libvirt.syms +*.i +*.s diff -r 71a7d1d0ad9b src/Makefile.am --- a/src/Makefile.am Thu May 14 17:38:11 2009 +0100 +++ b/src/Makefile.am Thu May 14 17:47:01 2009 +0100 @@ -16,7 +16,8 @@ INCLUDES = \ -DLOCALEBASEDIR=\""$(datadir)/locale"\" \ -DLOCAL_STATE_DIR=\""$(localstatedir)"\" \ -DGETTEXT_PACKAGE=\"$(PACKAGE)\" \ - $(WARN_CFLAGS) + $(WARN_CFLAGS) \ + $(LOCK_CHECKING_CFLAGS) confdir = $(sysconfdir)/libvirt/ conf_DATA = qemu.conf @@ -662,5 +663,5 @@ if WITH_NETWORK endif -CLEANFILES = *.gcov .libs/*.gcda .libs/*.gcno *.gcno *.gcda +CLEANFILES = *.gcov .libs/*.gcda .libs/*.gcno *.gcno *.gcda *.i *.s DISTCLEANFILES = $(BUILT_SOURCES) diff -r 71a7d1d0ad9b src/network_driver.c --- a/src/network_driver.c Thu May 14 17:38:11 2009 +0100 +++ b/src/network_driver.c Thu May 14 17:47:01 2009 +0100 @@ -1217,7 +1217,6 @@ static int networkStart(virNetworkPtr ne networkDriverLock(driver); network = virNetworkFindByUUID(&driver->networks, net->uuid); - networkDriverUnlock(driver); if (!network) { networkReportError(net->conn, NULL, net, VIR_ERR_INVALID_NETWORK, @@ -1230,6 +1229,7 @@ static int networkStart(virNetworkPtr ne cleanup: if (network) virNetworkObjUnlock(network); + networkDriverUnlock(driver); return ret; } @@ -1240,7 +1240,6 @@ static int networkDestroy(virNetworkPtr networkDriverLock(driver); network = virNetworkFindByUUID(&driver->networks, net->uuid); - networkDriverUnlock(driver); if (!network) { networkReportError(net->conn, NULL, net, VIR_ERR_INVALID_NETWORK, @@ -1258,6 +1257,7 @@ static int networkDestroy(virNetworkPtr cleanup: if (network) virNetworkObjUnlock(network); + networkDriverUnlock(driver); return ret; } @@ -1349,7 +1349,6 @@ static int networkSetAutostart(virNetwor networkDriverLock(driver); network = virNetworkFindByUUID(&driver->networks, net->uuid); - networkDriverUnlock(driver); if (!network) { networkReportError(net->conn, NULL, net, VIR_ERR_INVALID_NETWORK, @@ -1399,6 +1398,7 @@ cleanup: VIR_FREE(autostartLink); if (network) virNetworkObjUnlock(network); + networkDriverUnlock(driver); return ret; } diff -r 71a7d1d0ad9b src/qemu_driver.c --- a/src/qemu_driver.c Thu May 14 17:38:11 2009 +0100 +++ b/src/qemu_driver.c Thu May 14 17:47:01 2009 +0100 @@ -1811,9 +1811,10 @@ static virDrvOpenStatus qemudOpen(virCon static int qemudClose(virConnectPtr conn) { struct qemud_driver *driver = conn->privateData; + qemuDriverLock(driver); /* Get rid of callbacks registered for this conn */ virDomainEventCallbackListRemoveConn(conn, driver->domainEventCallbacks); - + qemuDriverUnlock(driver); conn->privateData = NULL; return 0; @@ -2229,7 +2230,6 @@ static int qemudDomainSuspend(virDomainP qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - qemuDriverUnlock(driver); if (!vm) { char uuidstr[VIR_UUID_STRING_BUFLEN]; @@ -2264,11 +2264,9 @@ cleanup: if (vm) virDomainObjUnlock(vm); - if (event) { - qemuDriverLock(driver); - qemuDomainEventQueue(driver, event); - qemuDriverUnlock(driver); - } + if (event) + qemuDomainEventQueue(driver, event); + qemuDriverUnlock(driver); return ret; } @@ -2282,7 +2280,6 @@ static int qemudDomainResume(virDomainPt qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - qemuDriverUnlock(driver); if (!vm) { char uuidstr[VIR_UUID_STRING_BUFLEN]; @@ -2316,11 +2313,9 @@ static int qemudDomainResume(virDomainPt cleanup: if (vm) virDomainObjUnlock(vm); - if (event) { - qemuDriverLock(driver); - qemuDomainEventQueue(driver, event); - qemuDriverUnlock(driver); - } + if (event) + qemuDomainEventQueue(driver, event); + qemuDriverUnlock(driver); return ret; } @@ -3153,7 +3148,6 @@ static int qemudDomainGetSecurityLabel(v qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - qemuDriverUnlock(driver); if (!vm) { char uuidstr[VIR_UUID_STRING_BUFLEN]; @@ -3199,6 +3193,7 @@ static int qemudDomainGetSecurityLabel(v cleanup: if (vm) virDomainObjUnlock(vm); + qemuDriverUnlock(driver); return ret; } @@ -3617,7 +3612,6 @@ static int qemudDomainStart(virDomainPtr qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - qemuDriverUnlock(driver); if (!vm) { char uuidstr[VIR_UUID_STRING_BUFLEN]; @@ -3636,11 +3630,9 @@ static int qemudDomainStart(virDomainPtr cleanup: if (vm) virDomainObjUnlock(vm); - if (event) { - qemuDriverLock(driver); - qemuDomainEventQueue(driver, event); - qemuDriverUnlock(driver); - } + if (event) + qemuDomainEventQueue(driver, event); + qemuDriverUnlock(driver); return ret; } @@ -4144,7 +4136,6 @@ static int qemudDomainAttachDevice(virDo dev = virDomainDeviceDefParse(dom->conn, driver->caps, vm->def, xml, VIR_DOMAIN_XML_INACTIVE); - qemuDriverUnlock(driver); if (dev == NULL) goto cleanup; @@ -4197,6 +4188,7 @@ cleanup: virDomainDeviceDefFree(dev); if (vm) virDomainObjUnlock(vm); + qemuDriverUnlock(driver); return ret; } @@ -4296,7 +4288,6 @@ static int qemudDomainDetachDevice(virDo dev = virDomainDeviceDefParse(dom->conn, driver->caps, vm->def, xml, VIR_DOMAIN_XML_INACTIVE); - qemuDriverUnlock(driver); if (dev == NULL) goto cleanup; @@ -4320,6 +4311,7 @@ cleanup: virDomainDeviceDefFree(dev); if (vm) virDomainObjUnlock(vm); + qemuDriverUnlock(driver); return ret; } @@ -4359,7 +4351,6 @@ static int qemudDomainSetAutostart(virDo qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - qemuDriverUnlock(driver); if (!vm) { char uuidstr[VIR_UUID_STRING_BUFLEN]; @@ -4417,6 +4408,7 @@ cleanup: VIR_FREE(autostartLink); if (vm) virDomainObjUnlock(vm); + qemuDriverUnlock(driver); return ret; } @@ -4985,6 +4977,7 @@ qemudDomainMigratePrepare2 (virConnectPt vm->def->name); goto cleanup; } + virDomainObjUnlock(vm); } if (!(vm = virDomainAssignDef(dconn, diff -r 71a7d1d0ad9b src/storage_driver.c --- a/src/storage_driver.c Thu May 14 17:38:11 2009 +0100 +++ b/src/storage_driver.c Thu May 14 17:47:01 2009 +0100 @@ -792,7 +792,6 @@ storagePoolRefresh(virStoragePoolPtr obj storageDriverLock(driver); pool = virStoragePoolObjFindByUUID(&driver->pools, obj->uuid); - storageDriverUnlock(driver); if (!pool) { virStorageReportError(obj->conn, VIR_ERR_INVALID_STORAGE_POOL, @@ -834,6 +833,7 @@ storagePoolRefresh(virStoragePoolPtr obj cleanup: if (pool) virStoragePoolObjUnlock(pool); + storageDriverUnlock(driver); return ret; } @@ -939,7 +939,6 @@ storagePoolSetAutostart(virStoragePoolPt storageDriverLock(driver); pool = virStoragePoolObjFindByUUID(&driver->pools, obj->uuid); - storageDriverUnlock(driver); if (!pool) { virStorageReportError(obj->conn, VIR_ERR_INVALID_STORAGE_POOL, @@ -988,6 +987,7 @@ storagePoolSetAutostart(virStoragePoolPt cleanup: if (pool) virStoragePoolObjUnlock(pool); + storageDriverUnlock(driver); return ret; } @@ -1274,7 +1274,10 @@ storageVolumeCreateXML(virStoragePoolPtr buildret = backend->buildVol(obj->conn, buildvoldef); + storageDriverLock(driver); virStoragePoolObjLock(pool); + storageDriverUnlock(driver); + voldef->building = 0; pool->asyncjobs--; diff -r 71a7d1d0ad9b src/test.c --- a/src/test.c Thu May 14 17:38:11 2009 +0100 +++ b/src/test.c Thu May 14 17:47:01 2009 +0100 @@ -659,10 +659,12 @@ static virDrvOpenStatus testOpen(virConn if (ret == VIR_DRV_OPEN_SUCCESS) { testConnPtr privconn = conn->privateData; + testDriverLock(privconn); /* Init callback list */ if (VIR_ALLOC(privconn->domainEventCallbacks) < 0 || !(privconn->domainEventQueue = virDomainEventQueueNew())) { virReportOOMError(NULL); + testDriverUnlock(privconn); testClose(conn); return VIR_DRV_OPEN_ERROR; } @@ -671,6 +673,7 @@ static virDrvOpenStatus testOpen(virConn virEventAddTimeout(-1, testDomainEventFlush, privconn, NULL)) < 0) DEBUG0("virEventAddTimeout failed: No addTimeoutImpl defined. " "continuing without events."); + testDriverUnlock(privconn); } return (ret); diff -r 71a7d1d0ad9b src/uml_driver.c --- a/src/uml_driver.c Thu May 14 17:38:11 2009 +0100 +++ b/src/uml_driver.c Thu May 14 17:47:01 2009 +0100 @@ -1553,7 +1553,6 @@ static int umlDomainStart(virDomainPtr d umlDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - umlDriverUnlock(driver); if (!vm) { umlReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_DOMAIN, @@ -1672,6 +1671,7 @@ static int umlDomainGetAutostart(virDoma cleanup: if (vm) virDomainObjUnlock(vm); + umlDriverUnlock(driver); return ret; } @@ -1684,7 +1684,6 @@ static int umlDomainSetAutostart(virDoma umlDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); - umlDriverUnlock(driver); if (!vm) { umlReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_DOMAIN, @@ -1740,6 +1739,7 @@ cleanup: VIR_FREE(autostartLink); if (vm) virDomainObjUnlock(vm); + umlDriverUnlock(driver); return ret; } diff -r 71a7d1d0ad9b tests/.cvsignore --- a/tests/.cvsignore Thu May 14 17:38:11 2009 +0100 +++ b/tests/.cvsignore Thu May 14 17:47:01 2009 +0100 @@ -20,3 +20,7 @@ eventtest *.gcda *.gcno *.exe +object-locking +object-locking.cmi +object-locking.cmx +object-locking-files.txt diff -r 71a7d1d0ad9b tests/.gitignore --- a/tests/.gitignore Thu May 14 17:38:11 2009 +0100 +++ b/tests/.gitignore Thu May 14 17:47:01 2009 +0100 @@ -20,3 +20,7 @@ eventtest *.gcda *.gcno *.exe +object-locking +object-locking.cmi +object-locking.cmx +object-locking-files.txt diff -r 71a7d1d0ad9b tests/Makefile.am --- a/tests/Makefile.am Thu May 14 17:38:11 2009 +0100 +++ b/tests/Makefile.am Thu May 14 17:47:01 2009 +0100 @@ -68,6 +68,10 @@ if WITH_SECDRIVER_SELINUX noinst_PROGRAMS += seclabeltest endif +if WITH_CIL +noinst_PROGRAMS += object-locking +endif + noinst_PROGRAMS += nodedevxml2xmltest test_scripts = \ @@ -234,4 +238,25 @@ eventtest_SOURCES = \ eventtest_LDADD = -lrt $(LDADDS) endif -CLEANFILES = *.cov *.gcov .libs/*.gcda .libs/*.gcno *.gcno *.gcda +if WITH_CIL +CILOPTFLAGS = +CILOPTINCS = +CILOPTPACKAGES = -package unix,str,cil +CILOPTLIBS = -linkpkg + +object_locking_SOURCES = object-locking.ml + +%.cmx: %.ml + ocamlfind ocamlopt $(CILOPTFLAGS) $(CILOPTINCS) $(CILOPTPACKAGES) -c $< + +object-locking: object-locking.cmx object-locking-files.txt + ocamlfind ocamlopt $(CILOPTFLAGS) $(CILOPTINCS) $(CILOPTPACKAGES) $(CILOPTLIBS) $< -o $@ + +object-locking-files.txt: + find $(top_builddir)/src/ -name '*.i' > $@ + +else +EXTRA_DIST += object-locking.ml +endif + +CLEANFILES = *.cov *.gcov .libs/*.gcda .libs/*.gcno *.gcno *.gcda object-locking-files.txt diff -r 71a7d1d0ad9b tests/object-locking.ml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/object-locking.ml Thu May 14 17:47:01 2009 +0100 @@ -0,0 +1,828 @@ +(* + * Analyse libvirt driver API methods for mutex locking mistakes + * + * Copyright 2008-2009 Red Hat, Inc + * + * 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: Daniel P. Berrange <berrange(a)redhat.com> + *) + +open Pretty +open Cil + +(* + * Convenient routine to load the contents of a file into + * a list of strings + *) +let input_file filename = + let chan = open_in filename in + let lines = ref [] in + try while true; do lines := input_line chan :: !lines done; [] + with + End_of_file -> close_in chan; List.rev !lines + +module DF = Dataflow +module UD = Usedef +module IH = Inthash +module E = Errormsg +module VS = UD.VS + +let debug = ref false + + +let driverTables = [ + "virDriver"; + "virNetworkDriver"; + "virStorageDriver"; + "virDeviceMonitor"; +(* "virStateDriver"; Disable for now, since shutdown/startup have wierd locking rules *) +] + +(* + * This is the list of all libvirt methods which return + * pointers to locked objects + *) +let lockedObjMethods = [ + "virDomainFindByID"; + "virDomainFindByUUID"; + "virDomainFindByName"; + "virDomainAssignDef"; + + "virNetworkFindByUUID"; + "virNetworkFindByName"; + "virNetworkAssignDef"; + + "virNodeDeviceFindByName"; + "virNodeDeviceAssignDef"; + + "virStoragePoolObjFindByUUID"; + "virStoragePoolObjFindByName"; + "virStoragePoolObjAssignDef" +] + + +(* + * This is the list of all libvirt methods which + * can release an object lock. Technically we + * ought to pair them up correctly with previous + * ones, but the compiler can already complain + * about passing a virNetworkObjPtr to a virDomainObjUnlock + * method so lets be lazy + *) +let objectLockMethods = [ + "virDomainObjLock"; + "virNetworkObjLock"; + "virStoragePoolObjLock"; + "virNodeDevObjLock" +] + +(* + * This is the list of all libvirt methods which + * can release an object lock. Technically we + * ought to pair them up correctly with previous + * ones, but the compiler can already complain + * about passing a virNetworkObjPtr to a virDomainObjUnlock + * method so lets be lazy + *) +let objectUnlockMethods = [ + "virDomainObjUnlock"; + "virNetworkObjUnlock"; + "virStoragePoolObjUnlock"; + "virNodeDevObjUnlock" +] + +(* + * The data types that the previous two sets of + * methods operate on + *) +let lockableObjects = [ + "virDomainObjPtr"; + "virNetworkObjPtr"; + "virStoragePoolObjPtr"; + "virNodeDevObjPtr" +] + + + +(* + * The methods which globally lock an entire driver + *) +let driverLockMethods = [ + "qemuDriverLock"; + "openvzDriverLock"; + "testDriverLock"; + "lxcDriverLock"; + "umlDriverLock"; + "nodedevDriverLock"; + "networkDriverLock"; + "storageDriverLock" +] + +(* + * The methods which globally unlock an entire driver + *) +let driverUnlockMethods = [ + "qemuDriverUnlock"; + "openvzDriverUnlock"; + "testDriverUnlock"; + "lxcDriverUnlock"; + "umlDriverUnlock"; + "nodedevDriverUnlock"; + "networkDriverUnlock"; + "storageDriverUnlock" +] + +(* + * The data types that the previous two sets of + * methods operate on. These may be structs or + * typedefs, we don't care + *) +let lockableDrivers = [ + "qemud_driver"; + "openvz_driver"; + "testConnPtr"; + "lxc_driver_t"; + "uml_driver"; + "virStorageDriverStatePtr"; + "network_driver"; + "virDeviceMonitorState"; +] + + +let isFuncCallLval lval methodList = + match lval with + Var vi, o -> + List.mem vi.vname methodList + | _ -> false + +let isFuncCallExp exp methodList = + match exp with + Lval lval -> + isFuncCallLval lval methodList + | _ -> false + +let isFuncCallInstr instr methodList = + match instr with + Call (retval,exp,explist,srcloc) -> + isFuncCallExp exp methodList + | _ -> false + + + +let findDriverFunc init = + match init with + SingleInit (exp) -> ( + match exp with + AddrOf (lval) -> ( + match lval with + Var vi, o -> + true + | _ -> false + ) + | _ -> false + ) + | _ ->false + +let findDriverFuncs init = + match init with + CompoundInit (typ, list) -> + List.filter ( + fun l -> + match l with + (offset, init) -> + findDriverFunc init + + ) list; + | _ -> ([]) + + +let getDriverFuncs initinfo = + match initinfo.init with + Some (i) -> + let ls = findDriverFuncs i in + ls + | _ -> [] + +let getDriverFuncName init = + match init with + SingleInit (exp) -> ( + match exp with + AddrOf (lval) -> ( + match lval with + Var vi, o -> + + vi.vname + | _ -> "unknown" + ) + | _ -> "unknown" + ) + | _ -> "unknown" + + +let getDriverFuncNames initinfo = + List.map ( + fun l -> + match l with + (offset, init) -> + getDriverFuncName init + ) (getDriverFuncs initinfo) + + +(* + * Convenience methods which take a Cil.Instr object + * and ask whether its associated with one of the + * method sets defined earlier + *) +let isObjectFetchCall instr = + isFuncCallInstr instr lockedObjMethods + +let isObjectLockCall instr = + isFuncCallInstr instr objectLockMethods + +let isObjectUnlockCall instr = + isFuncCallInstr instr objectUnlockMethods + +let isDriverLockCall instr = + isFuncCallInstr instr driverLockMethods + +let isDriverUnlockCall instr = + isFuncCallInstr instr driverUnlockMethods + + +let isWantedType typ typeList = + match typ with + TNamed (tinfo, attrs) -> + List.mem tinfo.tname typeList + | TPtr (ptrtyp, attrs) -> + let f = match ptrtyp with + TNamed (tinfo2, attrs) -> + List.mem tinfo2.tname typeList + | TComp (cinfo, attrs) -> + List.mem cinfo.cname typeList + | _ -> + false in + f + | _ -> false + +(* + * Convenience methods which take a Cil.Varinfo object + * and ask whether it matches a variable datatype that + * we're interested in tracking for locking purposes + *) +let isLockableObjectVar varinfo = + isWantedType varinfo.vtype lockableObjects + +let isLockableDriverVar varinfo = + isWantedType varinfo.vtype lockableDrivers + +let isDriverTable varinfo = + isWantedType varinfo.vtype driverTables + + +(* + * Take a Cil.Exp object (ie an expression) and see whether + * the expression corresponds to a check for NULL against + * one of our interesting objects + * eg + * + * if (!vm) ... + * + * For a variable 'virDomainObjPtr vm' + *) +let isLockableThingNull exp funcheck = + match exp with + | UnOp (op,exp,typ) -> ( + match op with + LNot -> ( + match exp with + Lval (lhost, off) -> ( + match lhost with + Var vi -> + funcheck vi + | _ -> false + ) + | _ -> false + ) + | _ -> false + ) + | _ -> + false + +let isLockableObjectNull exp = + isLockableThingNull exp isLockableObjectVar + +let isLockableDriverNull exp = + isLockableThingNull exp isLockableDriverVar + + +(* + * Prior to validating a function, intialize these + * to VS.empty + * + * They contain the list of driver and object variables + * objects declared as local variables + * + *) +let lockableObjs: VS.t ref = ref VS.empty +let lockableDriver: VS.t ref = ref VS.empty + +(* + * Given a Cil.Instr object (ie a single instruction), get + * the list of all used & defined variables associated with + * it. Then caculate intersection with the driver and object + * variables we're interested in tracking and return four sets + * + * List of used driver variables + * List of defined driver variables + * List of used object variables + * List of defined object variables + *) +let computeUseDefState i = + let u, d = UD.computeUseDefInstr i in + let useo = VS.inter u !lockableObjs in + let defo = VS.inter d !lockableObjs in + let used = VS.inter u !lockableDriver in + let defd = VS.inter d !lockableDriver in + (used, defd, useo, defo) + + +(* Some crude helpers for debugging this horrible code *) +let printVI vi = + ignore(printf " | %a %s\n" d_type vi.vtype vi.vname) + +let printVS vs = + VS.iter printVI vs + + +let prettyprint2 stmdat () (_, ld, ud, lo, ui, uud, uuo, loud, ldlo, dead) = + text "" + + +type ilist = Cil.instr list + +(* + * This module implements the Cil.DataFlow.ForwardsTransfer + * interface. This is what 'does the interesting stuff' + * when walking over a function's code paths + *) +module Locking = struct + let name = "Locking" + let debug = debug + + (* + * Our state currently consists of + * + * The set of driver variables that are locked + * The set of driver variables that are unlocked + * The set of object variables that are locked + * The set of object variables that are unlocked + * + * Lists of Cil.Instr for: + * + * Instrs using an unlocked driver variable + * Instrs using an unlocked object variable + * Instrs locking a object variable while not holding a locked driver variable + * Instrs locking a driver variable while holding a locked object variable + * Instrs causing deadlock by fetching a lock object, while an object is already locked + * + *) + type t = (unit * VS.t * VS.t * VS.t * VS.t * ilist * ilist * ilist * ilist * ilist) + + (* This holds an instance of our state data, per statement *) + let stmtStartData = IH.create 32 + + let pretty = + prettyprint2 stmtStartData + + let copy (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + ((), ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) + + let computeFirstPredecessor stm (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + ((), ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) + + + (* + * Merge existing state for a statement, with new state + * + * If new and old state is the same, this returns None, + * If they are different, then returns the union. + *) + let combinePredecessors (stm:stmt) ~(old:t) ((_, ldn, udn, lon, uon, uudn, uuon, loudn, ldlon, deadn):t) = + match old with (_, ldo, udo, loo,uoo, uudo, uuoo, loudo, ldloo, deado)-> begin + let lde= (VS.equal ldo ldn) || ((VS.is_empty ldo) && (VS.is_empty ldn)) in + let ude= VS.equal udo udn || ((VS.is_empty udo) && (VS.is_empty udn)) in + let loe= VS.equal loo lon || ((VS.is_empty loo) && (VS.is_empty lon)) in + let uoe= VS.equal uoo uon || ((VS.is_empty uoo) && (VS.is_empty uon)) in + + if lde && ude && loe && uoe then + None + else ( + let ldret = VS.union ldo ldn in + let udret = VS.union udo udn in + let loret = VS.union loo lon in + let uoret = VS.union uoo uon in + Some ((), ldret, udret, loret, uoret, uudn, uuon, loudn, ldlon, deadn) + ) + end + + + (* + * This handles a Cil.Instr object. This is sortof a C level statement. + *) + let doInstr i (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + let transform (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + let used, defd, useo, defo = computeUseDefState i in + + + if isDriverLockCall i then ( + (* + * A driver was locked, so add to the list of locked + * driver variables, and remove from the unlocked list + *) + let retld = VS.union ld used in + let retud = VS.diff ud used in + + (* + * Report if any objects are locked already since + * thats a deadlock risk + *) + if VS.is_empty lo then + ((), retld, retud, lo, uo, uud, uuo, loud, ldlo, dead) + else + ((), retld, retud, lo, uo, uud, uuo, loud, List.append ldlo [i], dead) + ) else if isDriverUnlockCall i then ( + (* + * A driver was unlocked, so add to the list of unlocked + * driver variables, and remove from the locked list + *) + let retld = VS.diff ld used in + let retud = VS.union ud used in + + ((), retld, retud, lo, uo, uud, uuo, loud, ldlo, dead); + ) else if isObjectFetchCall i then ( + (* + * A object was fetched & locked, so add to the list of + * locked driver variables. Nothing to remove from unlocked + * list here. + * + * XXX, not entirely true. We should check if they're + * blowing away an existing non-NULL value in the lval + * really. + *) + let retlo = VS.union lo defo in + + (* + * Report if driver is not locked, since that's a safety + * risk + *) + if VS.is_empty ld then ( + if VS.is_empty lo then ( + ((), ld, ud, retlo, uo, uud, uuo, List.append loud [i], ldlo, dead) + ) else ( + ((), ld, ud, retlo, uo, uud, uuo, List.append loud [i], ldlo, List.append dead [i]) + ) + ) else ( + if VS.is_empty lo then ( + ((), ld, ud, retlo, uo, uud, uuo, loud, ldlo, dead) + ) else ( + ((), ld, ud, retlo, uo, uud, uuo, loud, ldlo, List.append dead [i]) + ) + ) + ) else if isObjectLockCall i then ( + (* + * A driver was locked, so add to the list of locked + * driver variables, and remove from the unlocked list + *) + let retlo = VS.union lo useo in + let retuo = VS.diff uo useo in + + (* + * Report if driver is not locked, since that's a safety + * risk + *) + if VS.is_empty ld then + ((), ld, ud, retlo, retuo, uud, uuo, List.append loud [i], ldlo, dead) + else + ((), ld, ud, retlo, retuo, uud, uuo, loud, ldlo, dead) + ) else if isObjectUnlockCall i then ( + (* + * A object was unlocked, so add to the list of unlocked + * driver variables, and remove from the locked list + *) + let retlo = VS.diff lo useo in + let retuo = VS.union uo useo in + ((), ld, ud, retlo, retuo, uud, uuo, loud, ldlo, dead); + ) else ( + (* + * Nothing special happened, at best an assignment. + * So add any defined variables to the list of unlocked + * object or driver variables. + * XXX same edge case as isObjectFetchCall about possible + * overwriting + *) + let retud = VS.union ud defd in + let retuo = VS.union uo defo in + + (* + * Report is a driver is used while unlocked + *) + let retuud = + if not (VS.is_empty used) && (VS.is_empty ld) then + List.append uud [i] + else + uud in + (* + * Report is a object is used while unlocked + *) + let retuuo = + if not (VS.is_empty useo) && (VS.is_empty lo) then + List.append uuo [i] + else + uuo in + + ((), ld, retud, lo, retuo, retuud, retuuo, loud, ldlo, dead) + ); + in + + DF.Post transform + + (* + * This handles a Cil.Stmt object. This is sortof a C code block + *) + let doStmt stm (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + DF.SUse ((), ld, ud, lo, uo, [], [], [], [], []) + + + (* + * This handles decision making for a conditional statement, + * ie an if (foo). It is called twice for each conditional + * ie, once per possible choice. + *) + let doGuard exp (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + (* + * If we're going down a branch where our object variable + * is set to NULL, then we must remove it from the + * list of locked objects. This handles the case of... + * + * vm = virDomainFindByUUID(..) + * if (!vm) { + * .... this code branch .... + * } else { + * .... leaves default handling for this branch ... + * } + *) + let lonull = UD.computeUseExp exp in + + let loret = + if isLockableObjectNull exp then + VS.diff lo lonull + else + lo in + let uoret = + if isLockableObjectNull exp then + VS.union uo lonull + else + uo in + let ldret = + if isLockableDriverNull exp then + VS.diff ld lonull + else + ld in + let udret = + if isLockableDriverNull exp then + VS.union ud lonull + else + ud in + + DF.GUse ((), ldret, udret, loret, uoret, uud, uuo, loud, ldlo, dead) + + (* + * We're not filtering out any statements + *) + let filterStmt stm = true + +end + +module L = DF.ForwardsDataFlow(Locking) + +let () = + (* Read the list of files from "libvirt-files". *) + let files = input_file "object-locking-files.txt" in + let f3iles = [ "../src/qemu_driver.i" ] in + + (* Load & parse each input file. *) + let files = + List.map ( + fun filename -> + (* Why does parse return a continuation? *) + let f = Frontc.parse filename in + f () + ) files in + + (* Merge them. *) + let file = Mergecil.merge files "test" in + + (* Do control-flow-graph analysis. *) + Cfg.computeFileCFG file; + + print_endline ""; + + let driverVars = List.filter ( + function + | GVar (varinfo, initinfo, loc) -> (* global variable *) + let name = varinfo.vname in + if isDriverTable varinfo then + true + else + false + | _ -> false + ) file.globals in + + let driverVarFuncs = List.map ( + function + | GVar (varinfo, initinfo, loc) -> (* global variable *) + let name = varinfo.vname in + if isDriverTable varinfo then + getDriverFuncNames initinfo + else + [] + | _ -> [] + ) driverVars in + + let driverFuncsAll = List.flatten driverVarFuncs in + let driverFuncsSkip = [ + "testClose"; + "openvzClose"; + ] in + let driverFuncs = List.filter ( + fun st -> + if List.mem st driverFuncsSkip then + false + else + true + ) driverFuncsAll in + + (* + * Now comes our fun.... iterate over every global symbol + * definition Cfg found..... but... + *) + List.iter ( + function + (* ....only care about functions *) + | GFun (fundec, loc) -> (* function definition *) + let name = fundec.svar.vname in + + if List.mem name driverFuncs then ( + (* Initialize list of driver & object variables to be empty *) + ignore (lockableDriver = ref VS.empty); + ignore (lockableObjs = ref VS.empty); + + (* + * Query all local variables, and figure out which correspond + * to interesting driver & object variables we track + *) + List.iter ( + fun var -> + if isLockableDriverVar var then + lockableDriver := VS.add var !lockableDriver + else if isLockableObjectVar var then + lockableObjs := VS.add var !lockableObjs; + ) fundec.slocals; + + List.iter ( + fun gl -> + match gl with + GVar (vi, ii, loc) -> + if isLockableDriverVar vi then + lockableDriver := VS.add vi !lockableDriver + | _ -> () + ) file.globals; + + (* + * Initialize the state for each statement (ie C code block) + * to be empty + *) + List.iter ( + fun st -> + IH.add Locking.stmtStartData st.sid ((), + VS.empty, VS.empty, VS.empty, VS.empty, + [], [], [], [], []) + ) fundec.sallstmts; + + (* + * This walks all the code paths in the function building + * up the state for each statement (ie C code block) + * ie, this is invoking the "Locking" module we created + * earlier + *) + L.compute fundec.sallstmts; + + (* + * Find all statements (ie C code blocks) which have no + * successor statements. This means they are exit points + * in the function + *) + let exitPoints = List.filter ( + fun st -> + List.length st.succs = 0 + ) fundec.sallstmts in + + (* + * For each of the exit points, check to see if there are + * any with locked driver or object variables & grab them + *) + let leaks = List.filter ( + fun st -> + let (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + IH.find Locking.stmtStartData st.sid in + let leakDrivers = not (VS.is_empty ld) in + let leakObjects = not (VS.is_empty lo) in + leakDrivers or leakObjects + ) exitPoints in + + let mistakes = List.filter ( + fun st -> + let (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + IH.find Locking.stmtStartData st.sid in + let lockDriverOrdering = (List.length ldlo) > 0 in + let lockObjectOrdering = (List.length loud) > 0 in + + let useDriverUnlocked = (List.length uud) > 0 in + let useObjectUnlocked = (List.length uuo) > 0 in + + let deadLocked = (List.length dead) > 0 in + + lockDriverOrdering or lockObjectOrdering or useDriverUnlocked or useObjectUnlocked or deadLocked + ) fundec.sallstmts in + + if (List.length leaks) > 0 || (List.length mistakes) > 0 then ( + print_endline "================================================================"; + ignore (printf "Function: %s\n" name); + print_endline "----------------------------------------------------------------"; + ignore (printf " - Total exit points with locked vars: %d\n" (List.length leaks)); + + (* + * Finally tell the user which exit points had locked varaibles + * And show them the line number and code snippet for easy fixing + *) + List.iter ( + fun st -> + ignore (Pretty.printf " - At exit on %a\n^^^^^^^^^\n" d_stmt st); + let (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + IH.find Locking.stmtStartData st.sid in + print_endline " variables still locked are"; + printVS ld; + printVS lo + ) leaks; + + + ignore (printf " - Total blocks with lock ordering mistakes: %d\n" (List.length mistakes)); + List.iter ( + fun st -> + let (_, ld, ud, lo, uo, uud, uuo, loud, ldlo, dead) = + IH.find Locking.stmtStartData st.sid in + List.iter ( + fun i -> + ignore (Pretty.printf " - Driver locked while object is locked on %a\n" d_instr i); + ) ldlo; + List.iter ( + fun i -> + ignore (Pretty.printf " - Object locked while driver is unlocked on %a\n" d_instr i); + ) loud; + List.iter ( + fun i -> + ignore (Pretty.printf " - Driver used while unlocked on %a\n" d_instr i); + ) uud; + List.iter ( + fun i -> + ignore (Pretty.printf " - Object used while unlocked on %a\n" d_instr i); + ) uuo; + List.iter ( + fun i -> + ignore (Pretty.printf " - Object fetched while locked objects exist %a\n" d_instr i); + ) dead; + ) mistakes; + print_endline "================================================================"; + print_endline ""; + print_endline ""; + ); + + () + ) + | _ -> () + ) file.globals; + + -- |: 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 2

[libvirt] Remote virConnect.listDomainsID() returns invalid IDs (Python, Xen)
by Russell Haering 14 May '09

14 May '09

I'm working with the Libvirt Python bindings and have discovered the following: If I run virConnect.listDomainsID() locally I get a list of valid IDs. For example: >>> import libvirt >>> conn = libvirt.open('xen://') >>> conn.listDomainsID() [0, 35, 36, 37, 38, 39, 40, 61] If I run the exact same thing remotely I get a list including numerous invalid IDs. This example shows the same host as shown above, about 20 seconds earlier (everything essentially identical): >>> import libvirt >>> conn = libvirt.open('xen+tls://xxxx.xxxxx') >>> conn.listDomainsID() [0, 35, 36, 37, 38, 39, 40, 51, 52, 53, 54, 60, 61] IDs 51, 52, 53, 54 and 60 are invalid. I first noticed this problem working with libvirt 0.6.0 on the client and 0.6.2 on the server, and noticed that with those versions the problem also occurred with virsh. Locally(0.6.0): local ~ # virsh list Id Name State ---------------------------------- 0 Domain-0 running 35 xxxxxxx idle 36 xxxxx idle 37 xxxxxx idle 38 xxxxxxxxx idle 39 xxxxxx idle 40 xxxxx idle 61 xxxxxxxxxx idle Whereas remotely(0.6.0): remote # virsh -c xen+tls://xxxxx.xxxxx list Id Name State ---------------------------------- 0 Domain-0 running 35 xxxxxxx idle 36 xxxxx idle 37 xxxxxx idle 38 xxxxxxxxx idle 39 xxxxxx idle 40 xxxxx idle libvir: Xen Daemon error : GET operation failed: xend_get: error from xen daemon: libvir: Xen Daemon error : GET operation failed: xend_get: error from xen daemon: libvir: Xen Daemon error : GET operation failed: xend_get: error from xen daemon: libvir: Xen Daemon error : GET operation failed: xend_get: error from xen daemon: libvir: Xen Daemon error : GET operation failed: xend_get: error from xen daemon: 61 xxxxxxxxxx idle However with virsh 0.6.2 on the client this appears to be fixed. Thanks, -- Russell Haering Systems Administrator OSU Open Source Lab

1 0

[libvirt] Has anyone ever gotten the client or full to compile on OS X?
by Schley Andrew Kutz 14 May '09

14 May '09

It's me. The original instigator of the long Leopard thread from early 2008. I once again need to do libvirt development from my Mac, and I still cannot seem to get it to compile :( It is currently bugging out at: gcc -DHAVE_CONFIG_H -I. -I.. -I../gnulib/lib -I../gnulib/lib -I../ include -I../include -I../qemud -I/usr/include/libxml2 -DLIBDIR=\"/opt/ libvirt//lib\" -DBINDIR=\"/opt/libvirt//libexec\" -DSBINDIR=\"/opt/ libvirt//sbin\" -DSYSCONF_DIR=\"/opt/libvirt//etc\" -DLOCALEBASEDIR=\"/ opt/libvirt//share/locale\" -DLOCAL_STATE_DIR=\"/opt/libvirt//var\" - DGETTEXT_PACKAGE=\"libvirt\" -Wall -Wformat -Wmissing-prototypes - Wnested-externs -Wpointer-arith -Wextra -Wshadow -Wcast-align -Wwrite- strings -Waggregate-return -Wstrict-prototypes -Winline -Wredundant- decls -Wno-sign-compare -Wp,-D_FORTIFY_SOURCE=2 -fexceptions - fasynchronous-unwind-tables -I/opt/local/include -g -O2 -MT pci.lo -MD -MP -MF .deps/pci.Tpo -c pci.c -fno-common -DPIC -o .libs/pci.o pci.c: In function 'pciFindStubDriver': pci.c:652: error: 'MODPROBE' undeclared (first use in this function) pci.c:652: error: (Each undeclared identifier is reported only once pci.c:652: error: for each function it appears in.) make[3]: *** [pci.lo] Error 1 make[2]: *** [all] Error 2 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2 Any help is appreciated. -- -a "Ideally, a code library must be immediately usable by naive developers, easily customized by more sophisticated developers, and readily extensible by experts." -- L. Stein

4 16

[libvirt] [PATCH] Fix the style of argument("cpumap") in op_pincpu()
by Tatsuro Enokura 14 May '09

14 May '09

Hi all, The form of cpumap argument in op_pincpu method on xend change by following a patch. http://xenbits.xensource.com/xen-unstable.hg?rev/a63d20d7a941 xenDaemonDomainPinCpu() sends the string for cpumap argument like python's list style now. This patch is following the change of cpumap argument. "[0,1,2]" ---> "0,1,2" Signed-off-by: Tatsuro Enokura <fj2026af(a)aa.jp.fujitsu.com> # my e-mail address has changed. Thanks Tatsuro Enokura

3 9

[libvirt] LibvirtException: socket closed unexpectedly
by Marcel Michelmann 14 May '09

14 May '09

Hi *, I wrote a program in Java communicating with libvirt/XEN over SSH, with libvirt-java-0.2.1. I poll the CPU and RAM usage quite often (about 5 times a second). My program runs well about between 5-8 minutes, but then I get LibvirtException: socket closed unexpectedly. Although I try to catch the LibvirtExceptions with special System.out.println messages I have no idea where this is coming from. Now my question: what leads to this exception? exception caught: org.libvirt.LibvirtException: socket closed unexpectedly level:VIR_ERR_ERROR code:VIR_ERR_RPC domain:VIR_FROM_REMOTE hasConn:false hasDom:false hasNet:false message:socket closed unexpectedly str1:%s str2:socket closed unexpectedly str3:null int1:0 int2:0 After that I can start my program again, and it is the same again (runs some minutes and then throws the exception). Perhaps someone knows? I would be very thankful for any help! Best regards, Marcel

3 4

[libvirt] Latest git tree doesn't compile
by Florian Vichot 13 May '09

13 May '09

Hello everyone, I just got the latest and grooviest libvirt from the git tree at : git://git.et.redhat.com/libvirt.git but compiling fails: /bin/bash ../mylibtool --tag=CC --mode=compile gcc -DHAVE_CONFIG_H -I. -I.. -I../gnulib/lib -I../gnulib/lib -I../include -I../include -I../qemud -I/usr/include/libxml2 -DLIBDIR=\""/usr/local/lib"\" -DBINDIR=\""/usr/local/libexec"\" -DSBINDIR=\""/usr/local/sbin"\" -DSYSCONF_DIR="\"/usr/local/etc\"" -DLOCALEBASEDIR=\""/usr/local/share/locale"\" -DLOCAL_STATE_DIR=\""/usr/local/var"\" -DGETTEXT_PACKAGE=\"libvirt\" -Wall -Wformat -Wformat-security -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wextra -Wshadow -Wcast-align -Wwrite-strings -Waggregate-return -Wstrict-prototypes -Winline -Wredundant-decls -Wno-sign-compare -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fasynchronous-unwind-tables -g -O2 -MT lxc_driver.lo -MD -MP -MF .deps/lxc_driver.Tpo -c -o lxc_driver.lo lxc_driver.c (CC) -o lxc_driver.lo lxc_driver.c lxc_container.c: In function 'lxcContainerDropCapabilities': lxc_container.c:657: error: 'PR_CAPBSET_DROP' undeclared (first use in this function) lxc_container.c:657: error: (Each undeclared identifier is reported only once lxc_container.c:657: error: for each function it appears in.) Apparently, my machine doesn't have the PR_CAPBSET_DROP flag defined. It's inexistent in sys/capability.h and "man prctl". I run a Ubuntu 8.04, kernel is 2.6.24-24-openvz. Am I missing headers that are not tested for in the ./configure ? If not, any way that functionality could be disabled on distros that don't support it ? Thanks, Florian

2 2

[libvirt] [PATCH] avoid XML attributes escaping problems
by Daniel Veillard 13 May '09

13 May '09

The associated patch solves https://bugzilla.redhat.com/show_bug.cgi?id=499791 where ' or " character in attribute values may lead to non well-formed XML. This takes the simplest approach to always escape then to ' and " since this is completely equivalent from an XML point of view and avoid the trouble. 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] create domain for existing xp image
by Gunnar Thielebein 13 May '09

13 May '09

I want to create a domain for an existing xp machine via virt-install. The command is taken from this wiki article http://wiki.libvirt.org/page/QEMUSwitchToLibvirt: > virt-install --accelerate --hvm --connect qemu:///system -n xp -r 512 > --cdrom=/dev/null --file=WinXP.img --vnc --noautoconsole --os-type > windows --os-variant winxp --force When I start and connect I see message: > We apologize for the inconvenience, > ... > A recent hardware or software change might have caused this. When I look at the features I saw that it lacks acpi so it changes image HAL table. Is it intentional missing for that os/variant?

1 0

[libvirt] [PATCH 0/5] Interface Config public APIs and remote stubs
by Laine Stump 13 May '09

13 May '09

These 5 patches contain the public virInterface*() API definition, the local plumbing, and the RPC glue. The test driver and real driver are still TBI, but having this in will allow other people to work on related stuff. Other things still needed: python/java bindings cli commands in virsh I'm sure lots of other things I don't know about yet. Note that I have added a couple things to the API that I previously didn't realize were needed (virInterfaceRef and virInterfaceFree - I had figured the client side would just pass around a cookie, not actually have storage associated with it).

3 17

[libvirt] PATCH: Fix virsh reporting open errors
by Daniel P. Berrange 13 May '09

13 May '09

The error handling refactor we did in virsh a few months back, accidentally means we never report the real libvirt error message for the initial connection attempt. This patch fixes it Daniel Index: src/virsh.c =================================================================== RCS file: /data/cvs/libvirt/src/virsh.c,v retrieving revision 1.202 diff -u -p -r1.202 virsh.c --- src/virsh.c 28 Apr 2009 10:55:45 -0000 1.202 +++ src/virsh.c 8 May 2009 16:36:30 -0000 @@ -6876,6 +6876,7 @@ vshInit(vshControl *ctl) * such as "help". */ if (!ctl->conn) { + virshReportError(ctl); vshError(ctl, FALSE, "%s", _("failed to connect to the hypervisor")); return FALSE; } -- |: 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] qemu error: package not installed
by Nigel Goddard 13 May '09

13 May '09

I get the following error on Ubuntu 8.04 when tryint to install the libvirt-bin package: Setting up libvirt-bin (0.4.0-2ubuntu8.1) ... * Starting libvirt management daemon libvirtd libvir: QEMU error : cannot create bridge 'vnet%d' : Package not installed Failed to autostart network 'default': cannot create bridge 'vnet%d' : Package not installed Can anyone advise which package I need to install? thanks Nigel -- ============================================ Nigel Goddard Axiope Limited ngoddard(a)axiope.com Tel: +44 787 967 1811 http://www.axiope.com Fax: +44 871 251 4870 -------------------------------------------- Registered in Scotland. Registered address: Quartermile One, 15 Lauriston Place Edinburgh EH3 9EP, United Kingdom ============================================

4 3

[libvirt] [PATCH 0/7] virStorageVolCreateXMLFrom API
by Cole Robinson 12 May '09

12 May '09

Hi all, The following patches implement a new API call: virStorageVolCreateXMLFrom. virStorageVolPtr virStorageVolCreateXMLFrom(virStoragePoolPtr pool, const char *xmldesc, unsigned int flags, virStorageVolPtr clonevol) Arguments function similar to the regular volume create API, except the parameter 'clonevol' is used as the input source. The API allows for volume cloning, possibly across pools, or within the same pool but converting volume formats. Tests I performed (and ensured that a VM booted without complaints from the new image): Raw -> Raw (with successful sparse detection) Raw -> QCOW2 QCOW2 -> Raw QCOW2 -> QCOW2 QCOW2 -> VMDK Raw partition ('disk' volume) -> Raw file Thanks, Cole Cole Robinson (7): Add public API stubs for virStorageVolCreateXMLFrom Remote driver plumbing for virStorageVolCreateXMLFrom Test driver implementation of virStorageVolCreateXMLFrom Virsh commands vol-clone and vol-create-from Storage driver implementation for CreateXMLFrom Break out FS volume build routines to their own functions. VolumeCreateXMLFrom FS storage backend implementation. include/libvirt/libvirt.h | 4 + include/libvirt/libvirt.h.in | 4 + qemud/remote.c | 34 +++ qemud/remote_dispatch_args.h | 1 + qemud/remote_dispatch_prototypes.h | 7 + qemud/remote_dispatch_ret.h | 1 + qemud/remote_dispatch_table.h | 5 + qemud/remote_protocol.c | 24 ++ qemud/remote_protocol.h | 18 ++ qemud/remote_protocol.x | 15 +- src/driver.h | 6 + src/libvirt.c | 61 +++++- src/libvirt_public.syms | 5 + src/remote_internal.c | 33 +++ src/storage_backend.h | 2 + src/storage_backend_fs.c | 491 +++++++++++++++++++++++++----------- src/storage_driver.c | 153 +++++++++++- src/test.c | 95 +++++++ src/virsh.c | 177 +++++++++++++- 19 files changed, 985 insertions(+), 151 deletions(-)

3 19

[libvirt] [PATCH] Fix error squashing when refreshing file volumes
by Cole Robinson 12 May '09

12 May '09

When refreshing a file based pool, errors hit when determining a volume's format were being squashed, reporting OOM instead. The attached patch fixes the error reporting here. Reported by Jason Guiditta. Thanks, Cole

3 3

[libvirt] PATCH: Another attempt to fix vbox driver open
by Daniel P. Berrange 12 May '09

12 May '09

The patches we just applied for the VirtualBox open method still were not quite right. It would return VIR_DRV_OPEN_DECLINED when uri==NULL, but before doing so it would have set conn->uri to vbox:///session. So even though it declined the connection, all the later drivers would now ignore it. Also, it now returns DECLINED for some real errors that should be reported to the user. Here's an alternative idea I've had for trying to address this. Some goals: - If the user gives a URI with a vbox:/// prefix, we should always handle it, unless a 'server' is set when we leave it to the remote driver - If an invalid path is given we must give back a real error code - If after deciding the URI is for us, any initialization fails we must raise an error. - If the vbox glue layer is missing, we should still raise errors for requested URIs, so user knows their URI is correct. To do this, I've taken the approach of registering a dummy vbox driver if the glue layer is missing. This just parses the URI and always returns an error for any vbox:// URIs that would otherwise work Daniel Index: src/vbox/vbox_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/vbox/vbox_driver.c,v retrieving revision 1.2 diff -u -p -r1.2 vbox_driver.c --- src/vbox/vbox_driver.c 6 May 2009 13:51:19 -0000 1.2 +++ src/vbox/vbox_driver.c 8 May 2009 16:35:57 -0000 @@ -34,6 +34,7 @@ #include "logging.h" #include "vbox_driver.h" #include "vbox_XPCOMCGlue.h" +#include "virterror_internal.h" #define VIR_FROM_THIS VIR_FROM_VBOX @@ -43,15 +44,25 @@ extern virDriver vbox22Driver; extern virDriver vbox25Driver; #endif +static virDriver vboxDriverDummy; + +#define VIR_FROM_THIS VIR_FROM_VBOX + +#define vboxError(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_VBOX, code, __FILE__, \ + __FUNCTION__, __LINE__, fmt) int vboxRegister(void) { virDriverPtr driver; uint32_t uVersion; - /* vboxRegister() shouldn't fail as that will render libvirt unless. - * So, we use the v2.2 driver as a fallback/dummy. + /* + * If the glue layer won' initialize, we register a driver + * with a dummy open method, so we can report nicer errors + * if the user requests a vbox:// URI which we know won't + * ever work */ - driver = &vbox22Driver; + driver = &vboxDriverDummy; /* Init the glue and get the API version. */ if (VBoxCGlueInit() == 0) { @@ -79,7 +90,7 @@ int vboxRegister(void) { } } else { - DEBUG0("VBoxCGlueInit failed"); + DEBUG0("VBoxCGlueInit failed, using dummy driver"); } if (virRegisterDriver(driver) < 0) @@ -87,3 +98,46 @@ int vboxRegister(void) { return 0; } + +static virDrvOpenStatus vboxOpenDummy(virConnectPtr conn, + virConnectAuthPtr auth ATTRIBUTE_UNUSED, + int flags ATTRIBUTE_UNUSED) { + uid_t uid = getuid(); + + if (conn->uri == NULL || + conn->uri->scheme == NULL || + STRNEQ (conn->uri->scheme, "vbox") || + conn->uri->server != NULL) + return VIR_DRV_OPEN_DECLINED; + + if (conn->uri->path == NULL || STREQ(conn->uri->path, "")) { + vboxError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("no VirtualBox drviver path specified (try vbox:///session)")); + return VIR_DRV_OPEN_ERROR; + } + + if (uid != 0) { + if (STRNEQ (conn->uri->path, "/session")) { + vboxError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown driver path '%s' specified (try vbox:///session)"), conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } + } else { /* root */ + if (STRNEQ (conn->uri->path, "/system") && + STRNEQ (conn->uri->path, "/session")) { + vboxError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown driver path '%s' specified (try vbox:///system)"), conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } + } + + vboxError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("unable to initialize VirtualBox driver API")); + return VIR_DRV_OPEN_ERROR; +} + +static virDriver vboxDriverDummy = { + VIR_DRV_VBOX, + "VBOX", + .open = vboxOpenDummy, +}; Index: src/vbox/vbox_tmpl.c =================================================================== RCS file: /data/cvs/libvirt/src/vbox/vbox_tmpl.c,v retrieving revision 1.5 diff -u -p -r1.5 vbox_tmpl.c --- src/vbox/vbox_tmpl.c 8 May 2009 10:18:26 -0000 1.5 +++ src/vbox/vbox_tmpl.c 8 May 2009 16:35:57 -0000 @@ -216,16 +216,6 @@ no_memory: } static int vboxInitialize(virConnectPtr conn, vboxGlobalData *data) { - - if (VBoxCGlueInit() != 0) { - vboxError(conn, VIR_ERR_INTERNAL_ERROR, "Can't Initialize VirtualBox, VBoxCGlueInit failed."); - goto cleanup; - } - - /* This is for when glue init failed and we're serving as dummy driver. */ - if (g_pfnGetFunctions == NULL) - goto cleanup; - /* Get the API table for out version, g_pVBoxFuncs is for the oldest version of the API that we support so we cannot use that. */ data->pFuncs = g_pfnGetFunctions(VBOX_XPCOMC_VERSION); @@ -291,13 +281,13 @@ static int vboxExtractVersion(virConnect } static void vboxUninitialize(vboxGlobalData *data) { + if (!data) + return; + if (data->pFuncs) data->pFuncs->pfnComUninitialize(); VBoxCGlueTerm(); - if (!data) - return; - virDomainObjListFree(&data->domains); virCapabilitiesFree(data->caps); VIR_FREE(data); @@ -306,52 +296,62 @@ static void vboxUninitialize(vboxGlobalD static virDrvOpenStatus vboxOpen(virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, int flags ATTRIBUTE_UNUSED) { - vboxGlobalData *data; + vboxGlobalData *data = NULL; uid_t uid = getuid(); if (conn->uri == NULL) { conn->uri = xmlParseURI(uid ? "vbox:///session" : "vbox:///system"); if (conn->uri == NULL) { + virReportOOMError(conn); return VIR_DRV_OPEN_ERROR; } - } else if (conn->uri->scheme == NULL || - conn->uri->path == NULL ) { - return VIR_DRV_OPEN_DECLINED; } - if (STRNEQ (conn->uri->scheme, "vbox")) + if (conn->uri->scheme == NULL || + STRNEQ (conn->uri->scheme, "vbox")) + return VIR_DRV_OPEN_DECLINED; + + /* Leave for remote driver */ + if (conn->uri->server != NULL) return VIR_DRV_OPEN_DECLINED; + if (conn->uri->path == NULL || STREQ(conn->uri->path, "")) { + vboxError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("no VirtualBox drviver path specified (try vbox:///session)")); + return VIR_DRV_OPEN_ERROR; + } + if (uid != 0) { - if (STRNEQ (conn->uri->path, "/session")) - return VIR_DRV_OPEN_DECLINED; + if (STRNEQ (conn->uri->path, "/session")) { + vboxError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown driver path '%s' specified (try vbox:///session)"), conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } } else { /* root */ if (STRNEQ (conn->uri->path, "/system") && - STRNEQ (conn->uri->path, "/session")) - return VIR_DRV_OPEN_DECLINED; + STRNEQ (conn->uri->path, "/session")) { + vboxError(conn, VIR_ERR_INTERNAL_ERROR, + _("unknown driver path '%s' specified (try vbox:///system)"), conn->uri->path); + return VIR_DRV_OPEN_ERROR; + } } if (VIR_ALLOC(data) < 0) { virReportOOMError(conn); - goto cleanup; + return VIR_DRV_OPEN_ERROR; } - if (!(data->caps = vboxCapsInit())) - goto cleanup; - - if (vboxInitialize(conn, data) < 0) - goto cleanup; - - if (vboxExtractVersion(conn, data) < 0) - goto cleanup; + if (!(data->caps = vboxCapsInit()) || + vboxInitialize(conn, data) < 0 || + vboxExtractVersion(conn, data) < 0) { + vboxUninitialize(data); + return VIR_DRV_OPEN_ERROR; + } conn->privateData = data; DEBUG0("in vboxOpen"); return VIR_DRV_OPEN_SUCCESS; -cleanup: - vboxUninitialize(data); - return VIR_DRV_OPEN_DECLINED; } static int vboxClose(virConnectPtr conn) { -- |: 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 :|

4 5

[libvirt] [PATCH] also enable bridges without IP address
by Ludwig Nussel 12 May '09

12 May '09

Bridges that are not up won't forward packets Signed-off-by: Ludwig Nussel <ludwig.nussel(a)suse.de> cu Ludwig --- src/network_driver.c | 9 +++------ 1 files changed, 3 insertions(+), 6 deletions(-) diff --git a/src/network_driver.c b/src/network_driver.c index a17a769..a163b15 100644 --- a/src/network_driver.c +++ b/src/network_driver.c @@ -836,8 +836,7 @@ static int networkStartNetworkDaemon(virConnectPtr conn, goto err_delbr; } - if (network->def->ipAddress && - (err = brSetInterfaceUp(driver->brctl, network->def->bridge, 1))) { + if ((err = brSetInterfaceUp(driver->brctl, network->def->bridge, 1))) { virReportSystemError(conn, err, _("failed to bring the bridge '%s' up"), network->def->bridge); @@ -878,8 +877,7 @@ static int networkStartNetworkDaemon(virConnectPtr conn, networkRemoveIptablesRules(driver, network); err_delbr1: - if (network->def->ipAddress && - (err = brSetInterfaceUp(driver->brctl, network->def->bridge, 0))) { + if ((err = brSetInterfaceUp(driver->brctl, network->def->bridge, 0))) { char ebuf[1024]; networkLog(NETWORK_WARN, _("Failed to bring down bridge '%s' : %s\n"), network->def->bridge, virStrerror(err, ebuf, sizeof ebuf)); @@ -920,8 +918,7 @@ static int networkShutdownNetworkDaemon(virConnectPtr conn, networkRemoveIptablesRules(driver, network); char ebuf[1024]; - if (network->def->ipAddress && - (err = brSetInterfaceUp(driver->brctl, network->def->bridge, 0))) { + if ((err = brSetInterfaceUp(driver->brctl, network->def->bridge, 0))) { networkLog(NETWORK_WARN, _("Failed to bring down bridge '%s' : %s\n"), network->def->bridge, virStrerror(err, ebuf, sizeof ebuf)); } -- 1.6.2.1

3 2

[libvirt] PATCH: ALlow for quiet build
by Daniel P. Berrange 12 May '09

12 May '09

This is a simpler version of http://www.redhat.com/archives/libvir-list/2009-April/msg00523.html just overriding the LIBTOOL variable in configure.in, instead of in each Makefile.am As before you still need 'make -s' to get full effect Daniel diff -r f94c0624da5c configure.in --- a/configure.in Fri May 08 11:41:29 2009 +0100 +++ b/configure.in Mon May 11 16:46:08 2009 +0100 @@ -57,6 +57,9 @@ dnl Support building Win32 DLLs (must ap AC_LIBTOOL_WIN32_DLL AM_PROG_LIBTOOL +dnl Override normal libtool in favour of our quiet version +LIBTOOL='$(SHELL) $(top_srcdir)/mylibtool' +AC_SUBST([LIBTOOL]) AM_PROG_CC_C_O diff -r f94c0624da5c mylibtool --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/mylibtool Mon May 11 16:46:08 2009 +0100 @@ -0,0 +1,58 @@ +#!/bin/sh + +mode=libtool +cfiles="" +ofiles="" +afiles="" + +wantnext=0 +for v in "$@" +do + case $v + in + --mode=compile) + mode=CC + ;; + --mode=link) + mode=LD + ;; + esac + + case $v + in + *.c) + cfiles="$cfiles $v" + ;; + *.o) + if [ "$mode" = "LD" -o "$wantnext" = "1" ]; then + ofiles="$ofiles $v" + fi + ;; + *.lo) + if [ "$mode" = "LD" -o "$wantnext" = "1" ]; then + ofiles="$ofiles $v" + fi + ;; + esac + + if [ "$mode" = "LD" -a "$wantnext" = "1" ]; then + afiles="$afiles $v" + fi + + if [ "$v" = "-o" ]; then + wantnext=1 + else + wantnext=0 + fi +done + +args="" +test -n "$afiles" && args="$args -o$afiles" +test -n "$ofiles" -a "$mode" = "CC" && args="$args -o" +test -n "$ofiles" && args="$args$ofiles" +test -n "$cfiles" && args="$args$cfiles" + +echo "($mode)$args" + +here=`dirname $0` +exec $here/libtool --silent "$@" -- |: 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 :|

3 2

[libvirt] Java binding
by Zvi Dubitzky 12 May '09

12 May '09

Who is the contact for libvirt Java binding ? In a late mail I said that the Java binding crashes upon live migrate when the dname=null . It does not crash if dname = source domain name even though works slow. The C code of libvirt / libvirt.c /virDomainMigrate () says in its header that : "If a hypervisor supports renaming domains during migration, * then you may set the dname parameter to the new name (otherwise * it keeps the same name). If this is not supported by the * hypervisor, dname must be NULL or else you will get an error." So in the kvm case dname = NULL works for the C binding but not for the JAVA binding as it should . Any idea ? thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] PATCH: 0/4: Fix the event loop (again) no really, this time its fixed.
by Daniel P. Berrange 12 May '09

12 May '09

Back in March I attempted to fix the event loop handling of deleted file handles http://www.redhat.com/archives/libvir-list/2009-March/msg00246.html but have merely succeeded in screwing it up in a different fun & exciting way. That patch fixed a case where there was a deleted file handle in the list at the time virEventRunOnce() starts. Which was nice. Not nice, though was that it then breaks the case where a file handle is deleted during the course of dispatching events afer poll(). Investigating this also identified that when we were marking a file handle as deleted, we were forgetting to run virEventInterrupt(), which meant that in some cases the main thread would not be woken up until the next event triggers. Not very critical, but at the same time, not good because it delays cleanup & release of resources. This series of patches fixes the problems, and more importantly, adds a unit test which cover these nasty edge cases. Daniel -- |: 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 :|

3 10

[libvirt] QEMU/KVM auto-detection bug
by Chris Lalancette 12 May '09

12 May '09

All, Even with Guido/Pritesh's recent changes to the vbox open routine, the auto-detection in libvirt is currently broken. What happens is that in src/libvirt.c:do_open(), at the start of the loop to auto-detect drivers, ret->uri is NULL. As each driver declines, it remains NULL. However, the very first thing the vboxOpen() routine does is: if (conn->uri == NULL) { conn->uri = xmlParseURI(uid ? "vbox:///session" : "vbox:///system"); if (conn->uri == NULL) { return VIR_DRV_OPEN_ERROR; } So, any driver that is trying to auto-probe after vboxOpen sees a conn->uri with a scheme of vbox and a path of system/session, and they all fail (including Qemu, which is what I'm mostly concerned with at the moment). In point of fact, this whole auto-detection thing is fragile. We got away with it before because the qemuOpen() routine bombs out before doing the whole conn->uri == NULL check, but in general, it's very easy for one buggy driver to mess up auto-detection for all drivers. Any thoughts about where to go here? We should obviously fix the vboxOpen() routine to not mess with the URI if it's going to fail. Additionally, I was thinking that maybe we should reset conn->uri to NULL in each iteration of the auto-detection loop. While that might leak memory for buggy drivers, it should at least make this whole thing less fragile. Thoughts? -- Chris Lalancette

3 5

[libvirt] [PATCH 1/6] Allow multiple monitor devices (v2)
by Anthony Liguori 12 May '09

12 May '09

Right now only one monitor device can be enabled at a time. In order to support asynchronous notification of events, I would like to introduce a 'wait' command that waits for an event to occur. This implies that we need an additional monitor session to allow commands to still be executed while waiting for an asynchronous notification. Signed-off-by: Anthony Liguori <aliguori(a)us.ibm.com> diff --git a/vl.c b/vl.c index 4bd173f..f78cabb 100644 --- a/vl.c +++ b/vl.c @@ -184,6 +184,9 @@ int main(int argc, char **argv) /* Max number of bluetooth switches on the commandline. */ #define MAX_BT_CMDLINE 10 +/* Maximum number of monitor devices */ +#define MAX_MONITOR_DEVICES 10 + /* XXX: use a two level table to limit memory usage */ #define MAX_IOPORTS 65536 @@ -4252,8 +4255,9 @@ int main(int argc, char **argv, char **envp) int hda_index; int optind; const char *r, *optarg; - CharDriverState *monitor_hd = NULL; - const char *monitor_device; + CharDriverState *monitor_hds[MAX_MONITOR_DEVICES]; + const char *monitor_devices[MAX_MONITOR_DEVICES]; + int monitor_device_index; const char *serial_devices[MAX_SERIAL_PORTS]; int serial_device_index; const char *parallel_devices[MAX_PARALLEL_PORTS]; @@ -4324,7 +4328,6 @@ int main(int argc, char **argv, char **envp) kernel_cmdline = ""; cyls = heads = secs = 0; translation = BIOS_ATA_TRANSLATION_AUTO; - monitor_device = "vc:80Cx24C"; serial_devices[0] = "vc:80Cx24C"; for(i = 1; i < MAX_SERIAL_PORTS; i++) @@ -4340,6 +4343,11 @@ int main(int argc, char **argv, char **envp) virtio_consoles[i] = NULL; virtio_console_index = 0; + monitor_devices[0] = "vc:80Cx24C"; + for (i = 1; i < MAX_MONITOR_DEVICES; i++) + monitor_devices[i] = NULL; + monitor_device_index = 0; + usb_devices_index = 0; nb_net_clients = 0; @@ -4723,7 +4731,12 @@ int main(int argc, char **argv, char **envp) break; } case QEMU_OPTION_monitor: - monitor_device = optarg; + if (monitor_device_index >= MAX_MONITOR_DEVICES) { + fprintf(stderr, "qemu: too many monitor devices\n"); + exit(1); + } + monitor_devices[monitor_device_index] = optarg; + monitor_device_index++; break; case QEMU_OPTION_serial: if (serial_device_index >= MAX_SERIAL_PORTS) { @@ -4974,8 +4987,8 @@ int main(int argc, char **argv, char **envp) serial_devices[0] = "stdio"; if (parallel_device_index == 0) parallel_devices[0] = "null"; - if (strncmp(monitor_device, "vc", 2) == 0) - monitor_device = "stdio"; + if (strncmp(monitor_devices[0], "vc", 2) == 0) + monitor_devices[0] = "stdio"; } #ifndef _WIN32 @@ -5184,14 +5197,14 @@ int main(int argc, char **argv, char **envp) #endif /* Maintain compatibility with multiple stdio monitors */ - if (!strcmp(monitor_device,"stdio")) { + if (!strcmp(monitor_devices[0],"stdio")) { for (i = 0; i < MAX_SERIAL_PORTS; i++) { const char *devname = serial_devices[i]; if (devname && !strcmp(devname,"mon:stdio")) { - monitor_device = NULL; + monitor_devices[0] = NULL; break; } else if (devname && !strcmp(devname,"stdio")) { - monitor_device = NULL; + monitor_devices[0] = NULL; serial_devices[i] = "mon:stdio"; break; } @@ -5208,11 +5221,20 @@ int main(int argc, char **argv, char **envp) } } - if (monitor_device) { - monitor_hd = qemu_chr_open("monitor", monitor_device, NULL); - if (!monitor_hd) { - fprintf(stderr, "qemu: could not open monitor device '%s'\n", monitor_device); - exit(1); + for (i = 0; i < MAX_MONITOR_DEVICES; i++) { + const char *devname = monitor_devices[i]; + if (devname && strcmp(devname, "none")) { + char label[32]; + if (i == 0) + snprintf(label, sizeof(label), "monitor"); + else + snprintf(label, sizeof(label), "monitor%d", i); + monitor_hds[i] = qemu_chr_open(label, devname, NULL); + if (!monitor_hds[i]) { + fprintf(stderr, "qemu: could not open monitor device '%s'\n", + devname); + exit(1); + } } } @@ -5335,8 +5357,13 @@ int main(int argc, char **argv, char **envp) text_consoles_set_display(display_state); qemu_chr_initial_reset(); - if (monitor_device && monitor_hd) - monitor_init(monitor_hd, MONITOR_USE_READLINE | MONITOR_IS_DEFAULT); + for (i = 0; i < MAX_MONITOR_DEVICES; i++) { + if (monitor_devices[i] && monitor_hds[i]) { + monitor_init(monitor_hds[i], + MONITOR_USE_READLINE | + ((i == 0) ? MONITOR_IS_DEFAULT : 0)); + } + } for(i = 0; i < MAX_SERIAL_PORTS; i++) { const char *devname = serial_devices[i];

11 60

[libvirt] PATCH: Fix QEMU arg detection for kvm >= 85
by Daniel P. Berrange 11 May '09

11 May '09

The -help output from KVM >= 85 is now large than we expected. This patch increases the buffer from 8k to 64k when reading help. It also provides more detailed error messages should something go wrong in the future. Two examples of improved error reporting # virsh -c qemu:///system start VirtTest error: Failed to start domain VirtTest error: Unable to read QEMU help output: Value too large for defined data type # virsh -c qemu:///system start VirtTest error: Failed to start domain VirtTest error: internal error cannot parse QEMU version number in 'QEMU PC emulator version 0.9.1 (kvm-79), Copyright (c) 2003-2008 Fabrice Bellard' Daniel Index: src/qemu_conf.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_conf.c,v retrieving revision 1.145 diff -u -p -r1.145 qemu_conf.c --- src/qemu_conf.c 8 May 2009 10:07:16 -0000 1.145 +++ src/qemu_conf.c 8 May 2009 10:37:22 -0000 @@ -431,18 +431,28 @@ int qemudExtractVersionInfo(const char * return -1; char *help = NULL; - enum { MAX_HELP_OUTPUT_SIZE = 8192 }; + enum { MAX_HELP_OUTPUT_SIZE = 1024*64 }; int len = virFileReadLimFD(newstdout, MAX_HELP_OUTPUT_SIZE, &help); - if (len < 0) + if (len < 0) { + virReportSystemError(NULL, errno, "%s", + _("Unable to read QEMU help output")); goto cleanup2; + } if (sscanf(help, "QEMU PC emulator version %u.%u.%u (kvm-%u)", &major, &minor, &micro, &kvm_version) != 4) kvm_version = 0; - if (!kvm_version && sscanf(help, "QEMU PC emulator version %u.%u.%u", - &major, &minor, &micro) != 3) + if (!kvm_version && + sscanf(help, "QEMU PC emulator version %u.%u.%u", + &major, &minor, &micro) != 3) { + char *eol = strchr(help, '\n'); + if (eol) *eol = '\0'; + qemudReportError(NULL, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + _("cannot parse QEMU version number in '%s'"), + help); goto cleanup2; + } version = (major * 1000 * 1000) + (minor * 1000) + micro; Index: src/qemu_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_driver.c,v retrieving revision 1.236 diff -u -p -r1.236 qemu_driver.c --- src/qemu_driver.c 8 May 2009 10:11:14 -0000 1.236 +++ src/qemu_driver.c 8 May 2009 10:37:22 -0000 @@ -1379,12 +1379,8 @@ static int qemudStartVMDaemon(virConnect if (qemudExtractVersionInfo(emulator, NULL, - &qemuCmdFlags) < 0) { - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, - _("Cannot determine QEMU argv syntax %s"), - emulator); + &qemuCmdFlags) < 0) goto cleanup; - } if (qemuPrepareHostDevices(conn, vm->def) < 0) goto cleanup; @@ -3703,12 +3699,8 @@ static int qemudDomainChangeEjectableMed if (qemudExtractVersionInfo(vm->def->emulator, NULL, - &qemuCmdFlags) < 0) { - qemudReportError(conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, - _("Cannot determine QEMU argv syntax %s"), - vm->def->emulator); + &qemuCmdFlags) < 0) return -1; - } if (qemuCmdFlags & QEMUD_CMD_FLAG_DRIVE) { if (!(devname = qemudDiskDeviceName(conn, newdisk))) -- |: 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] [PATCH v2 0/4] Improve QEMU process startup error
by Cole Robinson 11 May '09

11 May '09

Error reporting for failed QEMU processes isn't too great at the moment. A couple cases I have hit: - Trying to start a VM with a cdrom pointing to a nonexistent file. The svirt hook function choked on this and indicated as much in the domain log, but I still had to wait for the pidfile timeout (10 seconds), and the only error reported was 'Domain did not show up'. - QEMU dies quickly at startup. Happened with the pulse audio issue, or if we build cli args incorrectly. The reporting here varies: if QEMU dies before writing out the pidfile, the above case applies. If the pidfile is written, we timeout in ReadLogOutput, and it's hit or miss whether we even report the logfile output to the user. The following patches improve the above cases by ensuring we never have to wait for a timeout, and that the logfile output is always reported in the error message. Changes since v1: Drop patch 1 + 2 (applied separately) Patch 7 is pretty isolated, so was resent to list separately. Don't waitpid in virExecDaemonize if virExec fails. Remove needless pidfile loop in qemudStartVMDaemon. Thanks, Cole Cole Robinson (4): Add helper function virExecDaemonize Report qemu log data if we fail to daemonize the process. Add pidfile argument to __virExec Check that QEMU is still alive while reading startup output. src/libvirt_private.syms | 2 +- src/proxy_internal.c | 16 +--- src/qemu_conf.c | 7 -- src/qemu_driver.c | 89 +++++++++++-------- src/remote_internal.c | 15 +--- src/uml_driver.c | 11 +-- src/util.c | 90 +++++++++++++++++--- src/util.h | 15 +++- .../qemuxml2argvdata/qemuxml2argv-boot-cdrom.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-boot-floppy.args | 2 +- .../qemuxml2argv-boot-network.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-bootloader.args | 2 +- .../qemuxml2argv-clock-localtime.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-clock-utc.args | 2 +- .../qemuxml2argv-console-compat.args | 2 +- .../qemuxml2argv-disk-cdrom-empty.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-cdrom.args | 2 +- .../qemuxml2argv-disk-drive-boot-cdrom.args | 2 +- .../qemuxml2argv-disk-drive-boot-disk.args | 2 +- .../qemuxml2argv-disk-drive-cache-v1-none.args | 2 +- .../qemuxml2argv-disk-drive-cache-v1-wb.args | 2 +- .../qemuxml2argv-disk-drive-cache-v1-wt.args | 2 +- .../qemuxml2argv-disk-drive-cache-v2-none.args | 2 +- .../qemuxml2argv-disk-drive-cache-v2-wb.args | 2 +- .../qemuxml2argv-disk-drive-cache-v2-wt.args | 2 +- .../qemuxml2argv-disk-drive-fmt-qcow.args | 2 +- .../qemuxml2argv-disk-drive-shared.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-floppy.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-disk-many.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-disk-usb.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-virtio.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-xenvbd.args | 2 +- .../qemuxml2argv-graphics-sdl-fullscreen.args | 2 +- .../qemuxml2argv-graphics-sdl.args | 2 +- .../qemuxml2argv-graphics-vnc-sasl.args | 2 +- .../qemuxml2argv-graphics-vnc-tls.args | 2 +- .../qemuxml2argv-graphics-vnc.args | 2 +- .../qemuxml2argv-hostdev-pci-address.args | 2 +- .../qemuxml2argv-hostdev-usb-address.args | 2 +- .../qemuxml2argv-hostdev-usb-product.args | 2 +- .../qemuxml2argv-input-usbmouse.args | 2 +- .../qemuxml2argv-input-usbtablet.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-input-xen.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-migrate.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-minimal.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-misc-acpi.args | 2 +- .../qemuxml2argv-misc-no-reboot.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-misc-uuid.args | 2 +- .../qemuxml2argv-net-eth-ifname.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-net-eth.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-net-user.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-net-virtio.args | 2 +- .../qemuxml2argv-parallel-tcp.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-restore-v1.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-restore-v2.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-dev.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-file.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-many.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-pty.args | 2 +- .../qemuxml2argv-serial-tcp-telnet.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-tcp.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-udp.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-unix.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-serial-vc.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-sound.args | 2 +- 65 files changed, 211 insertions(+), 148 deletions(-)

2 9

[libvirt] [PATCH 2/3] support for "Host only" and "Internal" networks in VirtualBox driver
by Pritesh Kothari 11 May '09

11 May '09

Hi All, As discussed on the list resending the networking patch's. the patch's are as below: [PATCH 1/3]: contains support for "Internal" network in libvirt [PATCH 2/3]: contains support for "Host only" and "Internal" networks in VirtualBox driver [PATCH 3/3]: contains networking API for hostonly networks in VirtualBox driver in libvirt (it contains all the fix's proposed on list along with two extra *DefinedNetworks functions) The patches work as per cvs checkin today. (git SHA1 ID: 9136ae2056b45ea83854d3fe31d860f645b8c883) Regards, Pritesh

3 2

[libvirt] [PATCH 1/3] support for "Internal" network in libvirt
by Pritesh Kothari 11 May '09

11 May '09

Hi All, As discussed on the list resending the networking patch's. the patch's are as below: [PATCH 1/3]: contains support for "Internal" network in libvirt [PATCH 2/3]: contains support for "Host only" and "Internal" networks in VirtualBox driver [PATCH 3/3]: contains networking API for hostonly networks in VirtualBox driver in libvirt (it contains all the fix's proposed on list along with two extra *DefinedNetworks functions) The patches work as per cvs checkin today. (git SHA1 ID: 9136ae2056b45ea83854d3fe31d860f645b8c883) Regards, Pritesh

3 2

[libvirt] Status and plans for next release
by Daniel Veillard 11 May '09

11 May '09

Seems I didn't send any reminder about the next release plans on Friday, so here is a recap: - feature freeze on the 22nd so we have 2 more week to work on reviewing and adding OpenNebula/Power drivers and try to get the NPIV, netcf and secure migration patches in. It's likely not everything will make the release cut but we can try ! - target for next release is Friday 29th So far we have mostly a lot of bug fixes and VirtualBox driver updates commited since 0.6.3, 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] Java binding blockStats bug (libvirt-java-0.2.1)
by Zvi Dubitzky 11 May '09

11 May '09

running blcokStats() with Java binding (libvirt-java-0.2.1 ) I get an exception Inspecting the src/jni/org_libvirt_Domain.c is see the following lines of code : JNIEXPORT jobject JNICALL Java_org_libvirt_Domain__1blockStats (JNIEnv *env, jobject obj, jlong VDP, jstring j_path){ struct _virDomainBlockStats stats; jobject j_stats; jclass stats_cls=(*env)->FindClass(env, "org/libvirt/DomainInterfaceStats"); const char *path = (*env)->GetStringUTFChars(env, j_path, NULL); if(virDomainBlockStats((virDomainPtr)VDP, path, &stats, sizeof(struct _virDomainBlockStats))<0){ (*env)->ReleaseStringUTFChars(env, j_path, path); return NULL; } . . . The bold line should be replaced by : jclass stats_cls=(*env)->FindClass(env, "org/libvirt/DomainBlockStats") as we deal here with blockStats and not network interface Stats ! thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] Fw: Java api for domain migration
by Zvi Dubitzky 10 May '09

10 May '09

Sorry , clarification: Bottom line is that the Java binding of libvirt should accept 'dname' parameter = null as the C libvirt case does without crashing The below stament that " If I try to specify a meaningful 'dname' then if it is the same name as the original host domain 'name' I will get a message saying : invalid domain pointer in no domain with matching id -1." is not correct. The migrate works under Java in this case as in the C case i.e the domain is migrated to the destination host but remains paused there for a long time until resumed there. If the 'dname' is null then the migration is faster but that works ok only with the libvirt C library. It crashes with the Java binding . Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL ----- Forwarded by Zvi Dubitzky/Haifa/IBM on 10/05/2009 13:29 ----- Zvi Dubitzky/Haifa/IBM 10/05/2009 13:14 To libvir-list(a)redhat.com cc Subject Fw: Java api for domain migration Hi I think I detected the origin of the libvirt Java binding crash in the case of domain migration though I do not know how to fix it yet . The fix should be in the src/jni/org_libvirt_Domain.c / migrate function. First: It turns out the that the C libvirt interface for migrate ( libvirt 0.6.1. + kvm-84) has a limitation with specifying the 'dname' parameter . It should be NULL to work ok , meaning the domain is migrated from the old host to the new host under the same domain name . If I try to specify a meaningful 'dname' then if it is the same name as the original host domain 'name' I will get a message saying : invalid domain pointer in no domain with matching id -1. If I am using 'dname' different than the original host domain 'name' it will complain that no such domain is found on the target host. So using NULL for 'dname' is the only one that makes the migrate work Using 'dname' = NULL with the Java binding causes it though to crash as below (see .log file). Fixing the jni code as below (see added 'if') does not help: src/jni/org_libvirt_Domain.c -------------------------------- JNIEXPORT jlong JNICALL Java_org_libvirt_Domain__1migrate (JNIEnv *env, jobject obj, jlong VDP, jobject dconn, jlong flags, jstring j_dname, jstring j_uri, jlong bandwidth){ virConnectPtr destVCP; const char *dname=NULL; const char *uri=NULL; //if String="", we pass NULL to the library if((*env)->GetStringLength(env, j_dname)>0) dname=(*env)->GetStringUTFChars(env, j_dname, NULL); //if String="", we pass NULL to the library if((*env)->GetStringLength(env, j_uri)>0) uri=(*env)->GetStringUTFChars(env, j_uri, NULL); //Extract the destination Conn Ptr destVCP=(virConnectPtr)(*env)->GetLongField(env, dconn, (*env)->GetFieldID(env, (*env)->GetObjectClass(env, dconn), "VCP", "J")); jlong retval = (jlong)virDomainMigrate((virDomainPtr)VDP, destVCP, flags, dname, uri, bandwidth) ; if (dname) (*env)->ReleaseStringUTFChars(env, j_dname, dname); if (uri) (*env)->ReleaseStringUTFChars(env, j_uri, uri); return retval; } Please your advice on how to solve the migrate for the Java binding thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL ----- Forwarded by Zvi Dubitzky/Haifa/IBM on 10/05/2009 12:59 ----- Zvi Dubitzky/Haifa/IBM 07/05/2009 13:49 To libvir-list(a)redhat.com cc Subject Java api for libvirt migration I am trying to do a migrate of VM with libvirt 0.6.1 and kvm-84 . I am successful in doing it many times via the C interface . I can do most of the basic libvirt activity via the libvirt JAVA library (libvirt-java-0.2.1). When it comes to a migrate the JVM fails wit a log file attached below . Can someone help. My memory consumption in my JAVA test code is small. The migrate parameters I am sending with the Java are identical to those I use in the C case and is: destDomain = localDomain.migrate (dcon, // destination host Connect object VIR_MIGRATE_LIVE, null, //const char *dname destination domain name target_uri, //"tcp:9.148.28.13:4444" 0 // unsigned long bandwidth) - do not specify ); thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

1 0

[libvirt] Fw: Java api for domain migration
by Zvi Dubitzky 10 May '09

10 May '09

Hi I think I detected the origin of the libvirt Java binding crash in the case of domain migration though I do not know how to fix it yet . The fix should be in the src/jni/org_libvirt_Domain.c / migrate function. First: It turns out the that the C libvirt interface for migrate ( libvirt 0.6.1. + kvm-84) has a limitation with specifying the 'dname' parameter . It should be NULL to work ok , meaning the domain is migrated from the old host to the new host under the same domain name . If I try to specify a meaningful 'dname' then if it is the same name as the original host domain 'name' I will get a message saying : invalid domain pointer in no domain with matching id -1. If I am using 'dname' different than the original host domain 'name' it will complain that no such domain is found on the target host. So using NULL for 'dname' is the only one that makes the migrate work Using 'dname' = NULL with the Java binding causes it though to crash as below (see .log file). Fixing the jni code as below (see added 'if') does not help: src/jni/org_libvirt_Domain.c -------------------------------- JNIEXPORT jlong JNICALL Java_org_libvirt_Domain__1migrate (JNIEnv *env, jobject obj, jlong VDP, jobject dconn, jlong flags, jstring j_dname, jstring j_uri, jlong bandwidth){ virConnectPtr destVCP; const char *dname=NULL; const char *uri=NULL; //if String="", we pass NULL to the library if((*env)->GetStringLength(env, j_dname)>0) dname=(*env)->GetStringUTFChars(env, j_dname, NULL); //if String="", we pass NULL to the library if((*env)->GetStringLength(env, j_uri)>0) uri=(*env)->GetStringUTFChars(env, j_uri, NULL); //Extract the destination Conn Ptr destVCP=(virConnectPtr)(*env)->GetLongField(env, dconn, (*env)->GetFieldID(env, (*env)->GetObjectClass(env, dconn), "VCP", "J")); jlong retval = (jlong)virDomainMigrate((virDomainPtr)VDP, destVCP, flags, dname, uri, bandwidth) ; if (dname) (*env)->ReleaseStringUTFChars(env, j_dname, dname); if (uri) (*env)->ReleaseStringUTFChars(env, j_uri, uri); return retval; } Please your advice on how to solve the migrate for the Java binding thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL ----- Forwarded by Zvi Dubitzky/Haifa/IBM on 10/05/2009 12:59 ----- Zvi Dubitzky/Haifa/IBM 07/05/2009 13:49 To libvir-list(a)redhat.com cc Subject Java api for libvirt migration I am trying to do a migrate of VM with libvirt 0.6.1 and kvm-84 . I am successful in doing it many times via the C interface . I can do most of the basic libvirt activity via the libvirt JAVA library (libvirt-java-0.2.1). When it comes to a migrate the JVM fails wit a log file attached below . Can someone help. My memory consumption in my JAVA test code is small. The migrate parameters I am sending with the Java are identical to those I use in the C case and is: destDomain = localDomain.migrate (dcon, // destination host Connect object VIR_MIGRATE_LIVE, null, //const char *dname destination domain name target_uri, //"tcp:9.148.28.13:4444" 0 // unsigned long bandwidth) - do not specify ); thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

1 0

[libvirt] migration fails
by jens 09 May '09

09 May '09

Hello, I tried to migrate a kvm domain to a second identical machine with the command: virsh migrate --live dom1 qemu+ssh://second.machine.org/system which leads me to an unresponsive virsh on the second machine - 'virsh list' just takes forever and qemu wasn't started (the domain was not migrated), finally I have to kill libvirtd on the second machine. Because I'm new to libvirt I searched the web to find a solution for this behavior but without success. Please can someone on the list point me to the right direction how I can use libvirt to migrate kvm domains? I use kernel 2.6.29.1 and kvm-85 with libvirt 0.6.2 on an debian unstable system. Attached are debug logs generated by the source- and destination-machine. Thank you very much Jens

5 7

[libvirt] [PATCH] change permissions of directories in cgroups
by Ryota Ozaki 08 May '09

08 May '09

Hi, This patch creates a directory in cgroups with an ordinary permission 0755 (rwxr-xr-x) instead of 0655 (rw-r-xr-x). I guess 0655 is not expected and just a mistake, or is there a special reason? Thanks, ozaki-r Signed-off-by: Ryota Ozaki <ozaki.ryota(a)gmail.com> >From 4829855bc0baa3c75806f106bc0e54eb1da75eea Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Fri, 8 May 2009 03:23:50 +0900 Subject: [PATCH] change permissions of directories in cgroups This patch creates a directory in cgroups with an ordinary permission 0755 (rwxr-xr-x) instead of 0655 (rw-r-xr-x). --- src/cgroup.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/cgroup.c b/src/cgroup.c index d1d44a2..50517e2 100644 --- a/src/cgroup.c +++ b/src/cgroup.c @@ -436,7 +436,7 @@ static int virCgroupMakeGroup(const char *name) virCgroupFree(&root); if (access(path, F_OK) != 0) { - if (mkdir(path, 0655) < 0) { + if (mkdir(path, 0755) < 0) { rc = -errno; VIR_FREE(path); break; -- 1.6.0.6

3 3

[libvirt] [PATCH] lxc: fix mknod file type
by Ryota Ozaki 08 May '09

08 May '09

Hi, This patch fixes the 2nd argument of mknod syscall. The argument should include a file type, i.e., S_IFCHR in this case, otherwise created files will be regular files. Thanks, ozaki-r Signed-off-by: Ryota Ozaki <ozaki.ryota(a)gmail.com> >From 60009b725da855f131e8272f0397455ae27bafc3 Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Wed, 6 May 2009 03:47:04 +0900 Subject: [PATCH] lxc: fix mknod file type This patch fixes the 2nd argument of mknod syscall. The argument should include a file type, i.e., S_IFCHR in this case, otherwise created files will be regular files. --- src/lxc_container.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/lxc_container.c b/src/lxc_container.c index 3946b84..c30daec 100644 --- a/src/lxc_container.c +++ b/src/lxc_container.c @@ -440,7 +440,7 @@ static int lxcContainerPopulateDevices(void) /* Populate /dev/ with a few important bits */ for (i = 0 ; i < ARRAY_CARDINALITY(devs) ; i++) { dev_t dev = makedev(devs[i].maj, devs[i].min); - if (mknod(devs[i].path, 0, dev) < 0 || + if (mknod(devs[i].path, S_IFCHR, dev) < 0 || chmod(devs[i].path, devs[i].mode)) { virReportSystemError(NULL, errno, _("failed to make device %s"), @@ -457,7 +457,7 @@ static int lxcContainerPopulateDevices(void) } } else { dev_t dev = makedev(LXC_DEV_MAJ_TTY, LXC_DEV_MIN_PTMX); - if (mknod("/dev/ptmx", 0, dev) < 0 || + if (mknod("/dev/ptmx", S_IFCHR, dev) < 0 || chmod("/dev/ptmx", 0666)) { virReportSystemError(NULL, errno, "%s", _("failed to make device /dev/ptmx")); -- 1.6.0.6

2 4

[libvirt] [PATCH] vbox: don't fail on missing vbox
by Guido Günther 08 May '09

08 May '09

Hi, virsh fails with: 10:47:37.104: debug : do_open:922 : no name, allowing driver auto-select 10:47:37.104: debug : do_open:930 : trying driver 0 (Test) ... 10:47:37.104: debug : do_open:936 : driver 0 Test returned DECLINED 10:47:37.104: debug : do_open:930 : trying driver 1 (OPENVZ) ... 10:47:37.105: debug : do_open:936 : driver 1 OPENVZ returned DECLINED 10:47:37.105: debug : do_open:930 : trying driver 2 (VBOX) ... 10:47:37.106: debug : do_open:936 : driver 2 VBOX returned ERROR 10:47:37.106: debug : virUnrefConnect:210 : unref connection 0x10032b10 1 10:47:37.106: debug : virReleaseConnect:171 : release connection 0x10032b10 error: failed to connect to the hypervisor if one has built with vbox support but vbox is not installed. Attached patch calls VBoxCGlueInit() early so we can return 'declined' instead of 'error'. O.k. to apply? -- Guido

3 4

[libvirt] [PATCH] vbox: don't crash if init fails early
by Guido Günther 08 May '09

08 May '09

Hi, not having set up pfuncs already results in a core dump. O.k. to apply? Cheers, -- Guido

3 3

[libvirt] Libvirt-Java Binding
by Marcel Michelmann 07 May '09

07 May '09

Hello, I created a program in Java for communicating with a libvirt/XEN host over SSH. For every update I create a new connection. It just runs about 1 1/2 minutes. I guess it is because I do not close any of these connections. When I tried to close them after having used them (Connect.close()) the program hangs. But after pressing Ctrl+C the program stops and the results are shown. Is that a bug in the libvirt-Java binding? Does anybody else have the same problem with Connect.close()? Thanks, Marcel

3 3

[libvirt] using public vir*Free() (instead of virUnref*()) from vir*FreeName() destructors.
by Laine Stump 07 May '09

07 May '09

I just noticed in datatypes.c that the vir(Network|StoragePool|StorageVol)FreeName destructors call the public API vir*Free() functions rather than the local virUnref*(). vir*Free() all clear any errors, which seems like it might not be the right thing to do (eg, if we're cleaning out the hashtables as part of some error-handling path, and haven't yet logged the error). On the other hand, the public API function does some sanity checking on the object to make sure it really is what we think it is before calling the unref. So it's got that going for it. (5 points to anyone who catches the reference without going to Google) But I don't know enough to know whether or not that's important in this case. So which is more important: the extra sanity check, or not clearing existing error codes?

2 3

[libvirt] PCI Passthrough
by Aaron Clausen 07 May '09

07 May '09

I'm having a heck of a time trying to figure out how to do PCI passthrough for a PCI SCSI controller. I need, if at all possible, to access it so that I can use a SCSI tape drive with a Windows Server 2000 guest. This is xml config I have: <domain type='kvm'> <name>testserver</name> <uuid>f75a2f63-e1bd-72a5-6134-3fcafbcdb970</uuid> <memory>524288</memory> <currentMemory>524288</currentMemory> <vcpu>1</vcpu> <os> <type arch='x86_64' machine='pc'>hvm</type> <boot dev='hd'/> </os> <features> <pae/> </features> <clock offset='localtime'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/bin/kvm</emulator> <disk type='file' device='disk'> <source file='/var/lib/libvirt/images/testserver.img'/> <target dev='hda' bus='ide'/> </disk> <disk type='block' device='cdrom'> <source dev='/dev/cdrom'/> <target dev='hdc' bus='ide'/> <readonly/> </disk> <interface type='bridge'> <mac address='54:52:00:50:03:08'/> <source bridge='br0'/> <target dev='vnet2'/> </interface> <serial type='pty'> <source path='/dev/pts/7'/> <target port='0'/> </serial> <console type='pty' tty='/dev/pts/7'> <source path='/dev/pts/7'/> <target port='0'/> </console> <input type='tablet' bus='usb'/> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='-1' autoport='yes' keymap='en-us'/> </devices> </domain> -- Aaron Clausen mightymartianca(a)gmail.com

2 4

[libvirt] Virsh list on xen after stop/restart not working properly
by Thomas Graves 07 May '09

07 May '09

Hello, I am seeing an issue on libvirt 0.6.1 and 0.6.3 with xen3.3.1 where virsh list -all isn't reporting all vms that are active. It happens after you shutdown and then restart a vm. As a side note, it works fine on libvirt 0.3.3. It appears that the domid of the vm is changed when you shutdown/restart a vm on xen. Libvirt is checking xenstore to find domids and it is finding the old ones that are marked as cancelled and not finding the higher number ones that are active/valid. Has anyone seen this issue and have a patch for it? It looks like the old libvirt 0.3.3 went to the hypervisor and xend before the xenstore and that is working properly. Thanks, Tom

2 1

[libvirt] KVM processes -- should we be able to attach them to the libvirtd process?
by Hugh O. Brock 07 May '09

07 May '09

Not too long ago we took a patch that allowed QEMU VMs to keep running even if libvirtd died or was restarted. I was talking to Matt Farellee (cc'd) this afternoon about manageability, and he feels fairly strongly that this behavior should be optional -- in other words, it should be possible to guarantee that if libvirtd dies, it will take all the VMs with the "die-with-libvirtd" flag set down with it. I'm not sure this API is portable to Xen, but it would work on any hypervisor that represents the VM as a normal process. Does this strike anyone else as useful behavior? Thanks, --Hugh

5 6

[libvirt] Java api for libvirt migration
by Zvi Dubitzky 07 May '09

07 May '09

I am trying to do a migrate of VM with libvirt 0.6.1 and kvm-84 . I am successful in doing it many times via the C interface . I can do most of the basic libvirt activity via the libvirt JAVA library (libvirt-java-0.2.1). When it comes to a migrate the JVM fails wit a log file attached below . Can someone help. My memory consumption in my JAVA test code is small. The migrate parameters I am sending with the Java are identical to those I use in the C case and is: destDomain = localDomain.migrate (dcon, // destination host Connect object VIR_MIGRATE_LIVE, null, //const char *dname destination domain name target_uri, //"tcp:9.148.28.13:4444" 0 // unsigned long bandwidth) - do not specify ); thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

1 0

[libvirt] [PATCH] return a sensible error on bogus paths
by Guido Günther 07 May '09

07 May '09

Hi, attached patch makes it easier to detect bogus paths when creating/removing pidfiles. O.k. to apply? Cheers, -- Guido

2 2

[libvirt] [PATCH]: don't crash when calling libvirt_lxc without arguments
by Guido Günther 07 May '09

07 May '09

In this case we reference def with is NULL. O.k. to apply? Cheers, -- Guido

2 2

[libvirt] PATCH: Add LSB init headers
by Frederik Himpe 07 May '09

07 May '09

The following patch is needed to make sure libvirt is not started too early in the boot sequence on Mandriva. --- libvirt-0.6.1/qemud/libvirtd.init.in.orig 2009-03-05 19:34:05.000000000 +0100 +++ libvirt-0.6.1/qemud/libvirtd.init.in 2009-03-05 19:34:54.000000000 +0100 @@ -5,7 +5,11 @@ # ### BEGIN INIT INFO # Provides: libvirtd +# Required-Start: $network messagebus +# Should-Start: $named # Should-Start: xend +# Required-Stop: $network messagebus +# Should-Stop: $named # Default-Start: 3 4 5 # Short-Description: daemon for libvirt virtualization API # Description: This is a daemon for managing guest instances

4 5

[libvirt] [PATCH 2/2] support for multiple graphics devices
by Pritesh Kothari 07 May '09

07 May '09

Hi All, I have added support for multiple graphics devices, the patches are as below. I have checked them against current cvs head and works fine PATCH 1/2: contains changes in libvirt for multiple graphics devices PATCH 2/2: contains corresponding changes in qemu driver. Regards, Pritesh

3 5

[libvirt] Authentication with virConnectAuthPtr
by Eduardo Otubo 07 May '09

07 May '09

Hello all, I'll start using virConnectAuthPtr to handle the authentication proccess at the phyp driver I'm wrinting <https://www.redhat.com/archives/libvir-list/2009-April/msg00493.html> and I have some doubts in my mind: 1) What are the types of authentication that 'int *credtype' can hold? And who fills it with the information I'll need? 2) Once known the credential type, I need to use the function pointer 'virConnectAuthCallbackPtr cb' to get the information whatever it is, right? I mean, it can be a password, a pubkey or anything else, right? Is there a callback able to handle password or pubkeys? 3) And finally I'll be able to use the 'void *cbdata' to manage the authentication in my way. In my case, using libssh. In fact, in the end of the process, I'll just need a password or a key to get things working. There is no driver authenticatin agaist ssh channel, that's why I got so confused with this topic. I don't know if my thoughts are pretty clear here. But I would like to confirm the information 2 and 3, and check how 1 works. Could anyone help me? Thanks in advance, []'s -- Eduardo Otubo Software Engineer Linux Technology Center IBM Systems & Technology Group Mobile: +55 19 8135 0885 otubo(a)linux.vnet.ibm.com

2 2

[libvirt] discrepancies in issuing virsh capabilities with different user-accounts
by Gerrit Slomma 06 May '09

06 May '09

Hello When doing a virsh capabilities with root i get a different output than when issuing the command with my own unprivileged account roadrunner. Furthermore the first issuing of the command as roadrunner throws a error-message. Restarting of the libvirt-daemon gives the correct output. Sample: Last login: Wed May 6 21:00:56 2009 from 192.168.1.120 [root@rr018 ~]# virsh capabilities|grep kvm [root@rr018 ~]# su - roadrunner [roadrunner@rr018 ~]$ virsh capabilities|grep kvm libvir: Remote Fehler : unable to connect to '@/home/roadrunner/.libvirt/libvirt-sock': Verbindungsaufbau abgelehnt Fehler: Verbindung zum Hypervisor scheiterte Failed to bind socket to '@/home/roadrunner/.libvirt/libvirt-sock': Address already in use Failed to bind socket to '@/home/roadrunner/.libvirt/libvirt-sock': Address already in use Failed to bind socket to '@/home/roadrunner/.libvirt/libvirt-sock': Address already in use Failed to bind socket to '@/home/roadrunner/.libvirt/libvirt-sock': Address already in use ^C [roadrunner@rr018 ~]$ virsh capabilities|grep kvm <domain type='kvm'> <emulator>/usr/bin/kvm</emulator> <domain type='kvm'> [roadrunner@rr018 ~]$ exit logout [root@rr018 ~]# virsh capabilities|grep kvm [root@rr018 ~]# /etc/init.d/libvirtd restart libvirtd-Daemon beenden: [ OK ] libvirtd-Daemon starten: [ OK ] [root@rr018 ~]# virsh capabilities|grep kvm <domain type='kvm'> <emulator>/usr/bin/kvm</emulator> <domain type='kvm'> [root@rr018 ~]#

1 1

Re: [libvirt] KVM processes -- should we be able to attach them to the libvirtd process?
by Daniel P. Berrange 06 May '09

06 May '09

On Wed, May 06, 2009 at 08:56:18PM +0200, Gerrit Slomma wrote: > Daniel P. Berrange schrieb: > >On Tue, May 05, 2009 at 11:38:13PM -0500, Matthew Farrellee wrote: > > > >>It doesn't appear to be the case that the libvirtd daemon can trivially > >>restart and continue with no interruptions. Right now it loses track of > >>VMs. > >> > > > >That a is a bug then, if you can reproduce it, please file a BZ ticket > >so we can track it down & fix it. > > > > > >>In a scenario where VMs are not deployed and locked to specific physical > >>nodes, it can be highly valuable to have ways to ensure a VM is no > >>longer running when a layer of its management stops functioning. > >> > > > >IMHO this is a problem to be solved by clustering software. If the > >clustering software detects a failure with the management service, > >then it should power fence the entire node. Relying on management > >service failure to kill the VMs will never be reliable enough. > > > I think he is pointing towards a VM that runs on a host where it isn't > defined at via a corresponding *.xml. > If you restart a libvirt i looses connection to this or these specific > VM(s). That is a bug that needs fixing. Even if there is no persistent config, we should not loose track of the running VM, because we always write out the 'live' XML config to /var/run/libvirt explicitly so that it is available at restart. Daniel -- |: 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 :|

1 0

[libvirt] PATCH: Raise log level for dlopen() problems
by Daniel P. Berrange 06 May '09

06 May '09

Some of the problems when dlopen'ing a module really should be reported to the user/admin more readily. So this raises the logging level for the important failure messages, so they're visible by default Daniel diff -r 15c4668d403b src/driver.c --- a/src/driver.c Wed Apr 29 15:29:17 2009 +0100 +++ b/src/driver.c Thu Apr 30 14:49:27 2009 +0100 @@ -54,13 +54,13 @@ virDriverLoadModule(const char *name) return NULL; if (access(modfile, R_OK) < 0) { - DEBUG("Moodule %s not accessible", modfile); + VIR_WARN("Module %s not accessible", modfile); goto cleanup; } handle = dlopen(modfile, RTLD_NOW | RTLD_LOCAL); if (!handle) { - DEBUG("failed to load module %s %s", modfile, dlerror()); + VIR_ERROR("failed to load module %s %s", modfile, dlerror()); goto cleanup; } @@ -70,12 +70,12 @@ virDriverLoadModule(const char *name) regsym = dlsym(handle, regfunc); if (!regsym) { - DEBUG("Missing module registration symbol %s", regfunc); + VIR_ERROR("Missing module registration symbol %s", regfunc); goto cleanup; } if ((*regsym)() < 0) { - DEBUG("Failed module registration %s", regfunc); + VIR_ERROR("Failed module registration %s", regfunc); goto cleanup; } -- |: 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] PATCH: More robust name & UUID uniqueness checking for QEMU
by Daniel P. Berrange 06 May '09

06 May '09

When defining a VM config, we need to apply the following logi - If existing VM has same UUID - If name also matches => allow - Else => raise error - Else - If name matches => raise error - Else => allow When creating a live VM, or restoring a VM we need to apply similar, but slightly different logic - If existing VM has same UUID - If name also matches - If existing VM is running => raise error - Else => allow - Else => raise error - Else - If name matches => raise error - Else => allow This patch applies those checks for the QEMU driver Daniel diff -r 4e6a98395da5 src/qemu_driver.c --- a/src/qemu_driver.c Thu Apr 30 14:50:14 2009 +0100 +++ b/src/qemu_driver.c Thu Apr 30 15:03:03 2009 +0100 @@ -2145,22 +2145,37 @@ static virDomainPtr qemudDomainCreate(vi if (virSecurityDriverVerify(conn, def) < 0) goto cleanup; - vm = virDomainFindByName(&driver->domains, def->name); - if (vm) { - qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, - _("domain '%s' is already defined"), - def->name); - goto cleanup; - } + /* See if a VM with matching UUID already exists */ vm = virDomainFindByUUID(&driver->domains, def->uuid); if (vm) { - char uuidstr[VIR_UUID_STRING_BUFLEN]; - - virUUIDFormat(def->uuid, uuidstr); - qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, - _("domain with uuid '%s' is already defined"), - uuidstr); - goto cleanup; + /* UUID matches, but if names don't match, refuse it */ + if (STRNEQ(vm->def->name, def->name)) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(vm->def->uuid, uuidstr); + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain '%s' is already defined with uuid %s"), + vm->def->name, uuidstr); + goto cleanup; + } + + /* UUID & name match, but if VM is already active, refuse it */ + if (virDomainIsActive(vm)) { + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain is already active as '%s'"), vm->def->name); + goto cleanup; + } + virDomainObjUnlock(vm); + } else { + /* UUID does not match, but if a name matches, refuse it */ + vm = virDomainFindByName(&driver->domains, def->name); + if (vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(vm->def->uuid, uuidstr); + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain '%s' is already defined with uuid %s"), + def->name, uuidstr); + goto cleanup; + } } if (!(vm = virDomainAssignDef(conn, @@ -2348,6 +2363,11 @@ static int qemudDomainDestroy(virDomainP _("no domain with matching uuid '%s'"), uuidstr); goto cleanup; } + if (!virDomainIsActive(vm)) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + "%s", _("domain is not running")); + goto cleanup; + } qemudShutdownVMDaemon(dom->conn, driver, vm); event = virDomainEventNewFromObj(vm, @@ -3258,17 +3278,36 @@ static int qemudDomainRestore(virConnect goto cleanup; } - /* Ensure the name and UUID don't already exist in an active VM */ + /* See if a VM with matching UUID already exists */ vm = virDomainFindByUUID(&driver->domains, def->uuid); - if (!vm) - vm = virDomainFindByName(&driver->domains, def->name); if (vm) { + /* UUID matches, but if names don't match, refuse it */ + if (STRNEQ(vm->def->name, def->name)) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(vm->def->uuid, uuidstr); + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain '%s' is already defined with uuid %s"), + vm->def->name, uuidstr); + goto cleanup; + } + + /* UUID & name match, but if VM is already active, refuse it */ if (virDomainIsActive(vm)) { qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_INVALID, _("domain is already active as '%s'"), vm->def->name); goto cleanup; - } else { - virDomainObjUnlock(vm); + } + virDomainObjUnlock(vm); + } else { + /* UUID does not match, but if a name matches, refuse it */ + vm = virDomainFindByName(&driver->domains, def->name); + if (vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(vm->def->uuid, uuidstr); + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain '%s' is already defined with uuid %s"), + def->name, uuidstr); + goto cleanup; } } @@ -3603,18 +3642,41 @@ static virDomainPtr qemudDomainDefine(vi if (virSecurityDriverVerify(conn, def) < 0) goto cleanup; - vm = virDomainFindByName(&driver->domains, def->name); + /* See if a VM with matching UUID already exists */ + vm = virDomainFindByUUID(&driver->domains, def->uuid); if (vm) { + /* UUID matches, but if names don't match, refuse it */ + if (STRNEQ(vm->def->name, def->name)) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(vm->def->uuid, uuidstr); + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain '%s' is already defined with uuid %s"), + vm->def->name, uuidstr); + goto cleanup; + } + + /* UUID & name match */ virDomainObjUnlock(vm); newVM = 0; + } else { + /* UUID does not match, but if a name matches, refuse it */ + vm = virDomainFindByName(&driver->domains, def->name); + if (vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(vm->def->uuid, uuidstr); + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + _("domain '%s' is already defined with uuid %s"), + def->name, uuidstr); + goto cleanup; + } } if (!(vm = virDomainAssignDef(conn, &driver->domains, def))) { - virDomainDefFree(def); - goto cleanup; - } + goto cleanup; + } + def = NULL; vm->persistent = 1; if (virDomainSaveConfig(conn, @@ -3636,6 +3698,7 @@ static virDomainPtr qemudDomainDefine(vi if (dom) dom->id = vm->def->id; cleanup: + virDomainDefFree(def); if (vm) virDomainObjUnlock(vm); if (event) -- |: 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] PATCH: Don't reset / detach host PCI devices in test scripts !
by Daniel P. Berrange 06 May '09

06 May '09

The PCI passthrough patches made it so that qemudBuildCommandLine() would actually try to detach your host devices & reset them. Most definitely not what you want when running this via a test case! This patch moves the host device management out into a separate method, so that qemudBuildCommandLine() doesn't do anything except safely build the command line. Daniel qemu_conf.c | 46 ----------------------------------- qemu_driver.c | 76 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 76 insertions(+), 46 deletions(-) Index: src/qemu_conf.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_conf.c,v retrieving revision 1.133 diff -u -p -u -p -r1.133 qemu_conf.c --- src/qemu_conf.c 2 Mar 2009 20:22:35 -0000 1.133 +++ src/qemu_conf.c 2 Mar 2009 20:47:36 -0000 @@ -47,7 +47,6 @@ #include "datatypes.h" #include "xml.h" #include "nodeinfo.h" -#include "pci.h" #define VIR_FROM_THIS VIR_FROM_QEMU @@ -1395,52 +1394,7 @@ int qemudBuildCommandLine(virConnectPtr ADD_ARG_LIT("-pcidevice"); ADD_ARG_LIT(pcidev); VIR_FREE(pcidev); - - if (hostdev->managed) { - pciDevice *dev = pciGetDevice(conn, - hostdev->source.subsys.u.pci.domain, - hostdev->source.subsys.u.pci.bus, - hostdev->source.subsys.u.pci.slot, - hostdev->source.subsys.u.pci.function); - if (!dev) - goto error; - - if (pciDettachDevice(conn, dev) < 0) { - pciFreeDevice(conn, dev); - goto error; - } - - pciFreeDevice(conn, dev); - } /* else { - XXX validate that non-managed device isn't in use, eg - by checking that device is either un-bound, or bound - to pci-stub.ko - } */ } - - } - - /* Now that all the PCI hostdevs have be dettached, we can reset them */ - for (i = 0 ; i < vm->def->nhostdevs ; i++) { - virDomainHostdevDefPtr hostdev = vm->def->hostdevs[i]; - pciDevice *dev; - - if (hostdev->mode != VIR_DOMAIN_HOSTDEV_MODE_SUBSYS || - hostdev->source.subsys.type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) - continue; - - dev = pciGetDevice(conn, - hostdev->source.subsys.u.pci.domain, - hostdev->source.subsys.u.pci.bus, - hostdev->source.subsys.u.pci.slot, - hostdev->source.subsys.u.pci.function); - if (!dev) - goto error; - - if (pciResetDevice(conn, dev) < 0) - goto error; - - pciFreeDevice(conn, dev); } if (migrateFrom) { Index: src/qemu_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_driver.c,v retrieving revision 1.208 diff -u -p -u -p -r1.208 qemu_driver.c --- src/qemu_driver.c 2 Mar 2009 17:39:43 -0000 1.208 +++ src/qemu_driver.c 2 Mar 2009 20:47:36 -0000 @@ -1133,6 +1133,79 @@ static int qemudNextFreeVNCPort(struct q return -1; } +static int qemuPrepareHostDevices(virConnectPtr conn, + virDomainDefPtr def) { + int i; + + /* We have to use 2 loops here. *All* devices must + * be detached before we reset any of them, because + * in some cases you have to reset the whole PCI bus, + * which impacts all devices on it + */ + + for (i = 0 ; i < def->nhostdevs ; i++) { + virDomainHostdevDefPtr hostdev = def->hostdevs[i]; + + if (hostdev->mode != VIR_DOMAIN_HOSTDEV_MODE_SUBSYS) + continue; + if (hostdev->source.subsys.type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) + continue; + + if (!hostdev->managed) { + pciDevice *dev = pciGetDevice(conn, + hostdev->source.subsys.u.pci.domain, + hostdev->source.subsys.u.pci.bus, + hostdev->source.subsys.u.pci.slot, + hostdev->source.subsys.u.pci.function); + if (!dev) + goto error; + + if (pciDettachDevice(conn, dev) < 0) { + pciFreeDevice(conn, dev); + goto error; + } + + pciFreeDevice(conn, dev); + } /* else { + XXX validate that non-managed device isn't in use, eg + by checking that device is either un-bound, or bound + to pci-stub.ko + } */ + } + + /* Now that all the PCI hostdevs have be dettached, we can safely + * reset them */ + for (i = 0 ; i < def->nhostdevs ; i++) { + virDomainHostdevDefPtr hostdev = def->hostdevs[i]; + pciDevice *dev; + + if (hostdev->mode != VIR_DOMAIN_HOSTDEV_MODE_SUBSYS) + continue; + if (hostdev->source.subsys.type != VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) + continue; + + dev = pciGetDevice(conn, + hostdev->source.subsys.u.pci.domain, + hostdev->source.subsys.u.pci.bus, + hostdev->source.subsys.u.pci.slot, + hostdev->source.subsys.u.pci.function); + if (!dev) + goto error; + + if (pciResetDevice(conn, dev) < 0) { + pciFreeDevice(conn, dev); + goto error; + } + + pciFreeDevice(conn, dev); + } + + return 0; + +error: + return -1; +} + static virDomainPtr qemudDomainLookupByName(virConnectPtr conn, const char *name); @@ -1210,6 +1283,9 @@ static int qemudStartVMDaemon(virConnect return -1; } + if (qemuPrepareHostDevices(conn, vm->def) < 0) + return -1; + vm->def->id = driver->nextvmid++; if (qemudBuildCommandLine(conn, driver, vm, qemuCmdFlags, &argv, &progenv, -- |: 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 :|

4 5

[libvirt] [PATCH] Refresh QEMU driver caps in getCapabilities
by Cole Robinson 06 May '09

06 May '09

Hi all, The attached patch fixes QEMU getCapabilities to refresh caps before returning them to the user. Currently the capabilities are only refreshed once (at daemon startup), which means libvirtd needs to be restarted to pick up changes if QEMU or KVM are installed while the daemon is running. See: https://bugzilla.redhat.com/show_bug.cgi?id=460649 There are several things 'wrong' with this change: - We reset/rescan fields that won't change (host arch, mac address prefix). This should be fixed at some point, but isn't a big deal since total performance impact is negligible (see below). - We only refresh the capabilities when the user calls getCapabilities, which means we are still carrying around stale caps prior to that, which is what libvirt validates against. In practice, virt-manager and virt-install both call getCapabilities often so this isn't an issue, though the caps internal API should probably address this at some point. To test the performance impact, I used a simple python script: import libvirt conn = libvirt.open("qemu:///system") for i in range(0, 30): conn.getCapabilities() The time difference was on average .02 seconds slower, which I think is negligible. If at somepoint in the future, capabilities generation becomes smarter (searching PATH emulators, scraping device list output, etc.) it might be worth re-checking the time impact. But for now it doesn't seem to be an issue. Thanks, Cole

4 7

[libvirt] [PATCH 1/1] Changes in version detection and moving a global variable in the global struct
by Pritesh Kothari 06 May '09

06 May '09

Hi All, PATCH 1/1: contains changes in version detection and moving a global variable in the global struct, so that no more global vairables exists. Regards, Pritesh

3 4

[libvirt] [PATCH 1/1] Making path detection more robust in VirtualBox
by Pritesh Kothari 06 May '09

06 May '09

Hi All, PATCH 1/1: contains changes in the glue code for making path detection more robust in VirtualBox Regards, Pritesh

3 2

[libvirt] participate in irc chat
by jens 06 May '09

06 May '09

Hello, on http://libvirt.org/contact.html I read about the #virt irc channel. How can I participate in this channel, it seems to be locked because I have to specify a channel key. Greetings Jens

2 2

[libvirt] libvirt / virt-manager issue with NFS share as storage plus a philosophy question - resend w/subject
by mike.hinz＠yr20.com 06 May '09

06 May '09

I had originally posted a plea for help on the virt-manager related list, but the same issue exists with virsh and this list seems to have higher traffic. I'd been chatting on the irc channel with various people about this, but wanted to put this on the mail list for the record and to also raise a philosophical question. First, we've confirmed that Fedora 11 preview (fully updated) has an issue with properly mounting NFS shares without the noacl option. Fedora 10 works fine with regard to mounting NFS shares. That issue causes problems when attempting to write files to an NFS share and more importantly, for this group, causes problems when attempting to create an NFS based storage pool and then subsequently creating volumes on said NFS share. That bug is 499178. This was all figured out with Openfiler support and with the support of the fine folks working on libvirt and virt-manager. There is a somewhat ugly workaround for this and I can describe it in case anyone is interested. Now comes the philosophy piece: It seems like that the way that libvirt and virt-manager want to handle storage is to be able to fully control the mounting process. That seems to me to be a very nice thing as you have a single way to manage storage along with your VMs. However, at least with respect to NFS, there are a number of parameters that people might need to give to the mount command for things like performance optimization and who knows what else. Note that this is a consideration completely aside from the bug issue above. It seems to me that either you have to just use the user created mount points, either manually or by fstab OR you have to allow all the possible options to be passed if you want to fully control the process. Right now, it's tricky to do, if you want to get paras passed for tuning or whatever and requires some tricky handwork to get it done. One way to accomplish this from the user input perspective would be to add an 'options' input field that could take multiple parameters on the step 2 of 2 page of the 'add storage pool' function. So, is my thinking all screwed up about this or is there merit to this concept? I tend to think, at least right now, that the logical thing is to either have the user create the mount points through the normal mechanisms that have in place roughly forever and then you just ask for what these are with respect to the NFS pools or you have to allow for user to pass the required parameters via your mechanism, although there are certainly many, many options when it comes to NFS. Thanks for all the help to date! I hope the above is helpful! Regards, Mike Hinz President YR20 1718 Fry Road, Suite 440 Houston, TX 77084 mike.hinz(a)yr20.com 832-225-1293 (o) 713-594-3095 (m) 832-550-2657 (f) Regards, Mike Hinz President YR20 1718 Fry Road, Suite 440 Houston, TX 77084 mike.hinz(a)yr20.com 832-225-1293 (o) 713-594-3095 (m) 832-550-2657 (f)

2 1

[libvirt] Cannot determine QEMU argv syntax with KVM 85
by Frederik Himpe 06 May '09

06 May '09

With KVM 85, it is impossible to start domains with libvirtd anymore (I tried both libvirtd 0.6.1 and 0.6.3): # virsh start Cooker error: Failed to start domain Cooker error: internal error Cannot determine QEMU argv syntax /usr/bin/qemu-kvm When I downgrade to KMV 84, everything is working again. KVM 85's output of qemu --help: # LC_ALL=C /usr/bin/qemu-kvm --help QEMU PC emulator version 0.10.0 (kvm-85), Copyright (c) 2003-2008 Fabrice Bellard usage: qemu [options] [disk_image] 'disk_image' is a raw hard image image for IDE hard disk 0 Standard options: -h or -help display this help and exit -M machine select emulated machine (-M ? for list) -cpu cpu select CPU (-cpu ? for list) -smp n set the number of CPUs to 'n' [default=1] -fda/-fdb file use 'file' as floppy disk 0/1 image -hda/-hdb file use 'file' as IDE hard disk 0/1 image -hdc/-hdd file use 'file' as IDE hard disk 2/3 image -cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master) -drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i] [,cyls=c,heads=h,secs=s[,trans=t]][,snapshot=on|off] [,cache=writethrough|writeback|none][,format=f][,serial=s] [,boot=on|off] use 'file' as a drive image -mtdblock file use 'file' as on-board Flash memory image -sd file use 'file' as SecureDigital card image -pflash file use 'file' as a parallel flash image -boot [a|c|d|n] boot on floppy (a), hard disk (c), CD-ROM (d), or network (n) -snapshot write to temporary files instead of disk image files -m megs set virtual RAM size to megs MB [default=128] -k language use keyboard layout (for example 'fr' for French) -audio-help print list of audio drivers and their options -soundhw c1,... enable audio support and only specified sound cards (comma separated list) use -soundhw ? to get the list of supported cards use -soundhw all to enable all of them -usb enable the USB driver (will be the default soon) -usbdevice name add the host or guest USB device 'name' -name string set the name of the guest -uuid %08x-%04x-%04x-%04x-%012x specify machine UUID Display options: -nographic disable graphical output and redirect serial I/Os to console -curses use a curses/ncurses interface instead of SDL -no-frame open SDL window without a frame and window decorations -alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt) -no-quit disable SDL window close capability -sdl enable SDL -portrait rotate graphical output 90 deg left (only PXA LCD) -vga [std|cirrus|vmware|none] select video card type -full-screen start in full screen -vnc display start a VNC server on display i386 target only: -win2k-hack use it when installing Windows 2000 to avoid a disk full bug -rtc-td-hack use it to fix time drift in Windows ACPI HAL -no-fd-bootchk disable boot signature checking for floppy disks -no-acpi disable ACPI -no-hpet disable HPET -acpitable [sig=str][,rev=n][,oem_id=str][,oem_table_id=str][,oem_rev=n][,asl_compiler_id=str][,asl_compiler_rev=n][,data=file1[:file2]...] ACPI table description Network options: -net nic[,vlan=n][,macaddr=addr][,model=type][,name=str] create a new Network Interface Card and connect it to VLAN 'n' -net user[,vlan=n][,name=str][,hostname=host] connect the user mode network stack to VLAN 'n' and send hostname 'host' to DHCP clients -net tap[,vlan=n][,name=str][,fd=h][,ifname=name][,script=file][,downscript=dfile] connect the host TAP network interface to VLAN 'n' and use the network scripts 'file' (default=/etc/qemu-ifup) and 'dfile' (default=/etc/qemu-ifdown); use '[down]script=no' to disable script execution; use 'fd=h' to connect to an already opened TAP interface -net socket[,vlan=n][,name=str][,fd=h][,listen=[host]:port][,connect=host:port] connect the vlan 'n' to another VLAN using a socket connection -net socket[,vlan=n][,name=str][,fd=h][,mcast=maddr:port] connect the vlan 'n' to multicast maddr and port -net vde[,vlan=n][,name=str][,sock=socketpath][,port=n][,group=groupname][,mode=octalmode] connect the vlan 'n' to port 'n' of a vde switch running on host and listening for incoming connections on 'socketpath'. Use group 'groupname' and mode 'octalmode' to change default ownership and permissions for communication port. -net none use it alone to have zero network devices; if no -net option is provided, the default is '-net nic -net user' -tftp dir allow tftp access to files in dir [-net user] -bootp file advertise file in BOOTP replies -smb dir allow SMB access to files in 'dir' [-net user] -redir [tcp|udp]:host-port:[guest-host]:guest-port redirect TCP or UDP connections from host to guest [-net user] -bt hci,null dumb bluetooth HCI - doesn't respond to commands -bt hci,host[:id] use host's HCI with the given name -bt hci[,vlan=n] emulate a standard HCI in virtual scatternet 'n' -bt vhci[,vlan=n] add host computer to virtual scatternet 'n' using VHCI -bt device:dev[,vlan=n] emulate a bluetooth device 'dev' in scatternet 'n' Linux boot specific: -kernel bzImage use 'bzImage' as kernel image -append cmdline use 'cmdline' as kernel command line -initrd file use 'file' as initial ram disk Debug/Expert options: -serial dev redirect the serial port to char device 'dev' -parallel dev redirect the parallel port to char device 'dev' -monitor dev redirect the monitor to char device 'dev' -pidfile file write PID to 'file' -singlestep always run in singlestep mode -S freeze CPU at startup (use 'c' to start execution) -gdb dev wait for gdb connection on 'dev' -s shorthand for -gdb tcp::1234 -d item1,... output log to /tmp/qemu.log (use -d ? for a list of log items) -hdachs c,h,s[,t] force hard disk 0 physical geometry and the optional BIOS translation (t=none or lba) (usually qemu can guess them) -L path set the directory for the BIOS, VGA BIOS and keymaps -bios file set the filename for the BIOS -no-reboot exit instead of rebooting -no-shutdown stop before shutdown -loadvm [tag|id] start right away with a saved state (loadvm in monitor) -daemonize daemonize QEMU after initializing -option-rom rom load a file, rom, into the option ROM space -clock force the use of the given methods for timer alarm. To see what timers are available use -clock ? -localtime set the real time clock to local time [default=utc] -startdate select initial date of the clock -icount [N|auto] enable virtual instruction counter with 2^N clock ticks per instruction -echr chr set terminal escape character instead of ctrl-a -virtioconsole c set virtio console -show-cursor show cursor -tb-size n set TB size -incoming p prepare for incoming migration, listen on port p -chroot dir Chroot to dir just before starting the VM. -runas user Change to user id user just before starting the VM. -no-kvm disable KVM hardware virtualization -no-kvm-irqchip disable KVM kernel mode PIC/IOAPIC/LAPIC -no-kvm-pit disable KVM kernel mode PIT -no-kvm-pit-reinjection disable KVM kernel mode PIT interrupt reinjection -pcidevice host=bus:dev.func[,dma=none][,name=string] expose a PCI device to the guest OS. dma=none: don't perform any dma translations (default is to use an iommu) 'string' is used in log output. -enable-nesting enable support for running a VM inside the VM (AMD only) -cpu-vendor STRING override the cpuid vendor string -nvram FILE provide ia64 nvram contents -tdf enable guest time drift compensation -kvm-shadow-memory MEGABYTES allocate MEGABYTES for kvm mmu shadowing -mempath FILE provide backing storage for guest RAM -mem-prealloc preallocate guest memory (use with -mempath) During emulation, the following keys are useful: ctrl-alt-f toggle full screen ctrl-alt-n switch to virtual console 'n' ctrl-alt toggle mouse and keyboard grab When using -nographic, press 'ctrl-a h' to get some help. The working KVM 84 gives this: # LC_ALL=C /usr/bin/qemu-kvm --help QEMU PC emulator version 0.9.1 (kvm-84), Copyright (c) 2003-2008 Fabrice Bellard usage: qemu [options] [disk_image] 'disk_image' is a raw hard image image for IDE hard disk 0 Standard options: -h or -help display this help and exit -M machine select emulated machine (-M ? for list) -cpu cpu select CPU (-cpu ? for list) -smp n set the number of CPUs to 'n' [default=1] -fda/-fdb file use 'file' as floppy disk 0/1 image -hda/-hdb file use 'file' as IDE hard disk 0/1 image -hdc/-hdd file use 'file' as IDE hard disk 2/3 image -cdrom file use 'file' as IDE cdrom image (cdrom is ide1 master) -drive [file=file][,if=type][,bus=n][,unit=m][,media=d][,index=i] [,cyls=c,heads=h,secs=s[,trans=t]][,snapshot=on|off] [,cache=writethrough|writeback|none][,format=f][,serial=s] [,boot=on|off] use 'file' as a drive image -mtdblock file use 'file' as on-board Flash memory image -sd file use 'file' as SecureDigital card image -pflash file use 'file' as a parallel flash image -boot [a|c|d|n] boot on floppy (a), hard disk (c), CD-ROM (d), or network (n) -snapshot write to temporary files instead of disk image files -m megs set virtual RAM size to megs MB [default=128] -k language use keyboard layout (for example "fr" for French) -audio-help print list of audio drivers and their options -soundhw c1,... enable audio support and only specified sound cards (comma separated list) use -soundhw ? to get the list of supported cards use -soundhw all to enable all of them -usb enable the USB driver (will be the default soon) -usbdevice name add the host or guest USB device 'name' -name string set the name of the guest -uuid %08x-%04x-%04x-%04x-%012x specify machine UUID Display options: -nographic disable graphical output and redirect serial I/Os to console -curses use a curses/ncurses interface instead of SDL -no-frame open SDL window without a frame and window decorations -alt-grab use Ctrl-Alt-Shift to grab mouse (instead of Ctrl-Alt) -no-quit disable SDL window close capability -sdl enable SDL -portrait rotate graphical output 90 deg left (only PXA LCD) -vga [std|cirrus|vmware|none] select video card type -full-screen start in full screen -vnc display start a VNC server on display -name string set the name of the guest -uuid %08x-%04x-%04x-%04x-%012x specify machine UUID Network options: -net nic[,vlan=n][,macaddr=addr][,model=type][,name=str] create a new Network Interface Card and connect it to VLAN 'n' -net user[,vlan=n][,name=str][,hostname=host] connect the user mode network stack to VLAN 'n' and send hostname 'host' to DHCP clients -net tap[,vlan=n][,name=str][,fd=h][,ifname=name][,script=file][,downscript=dfile] connect the host TAP network interface to VLAN 'n' and use the network scripts 'file' (default=/etc/qemu-ifup) and 'dfile' (default=/etc/qemu-ifdown); use '[down]script=no' to disable script execution; use 'fd=h' to connect to an already opened TAP interface -net socket[,vlan=n][,name=str][,fd=h][,listen=[host]:port][,connect=host:port] connect the vlan 'n' to another VLAN using a socket connection -net socket[,vlan=n][,name=str][,fd=h][,mcast=maddr:port] connect the vlan 'n' to multicast maddr and port -net vde[,vlan=n][,name=str][,sock=socketpath][,port=n][,group=groupname][,mode=octalmode] connect the vlan 'n' to port 'n' of a vde switch running on host and listening for incoming connections on 'socketpath'. Use group 'groupname' and mode 'octalmode' to change default ownership and permissions for communication port. -net none use it alone to have zero network devices; if no -net option is provided, the default is '-net nic -net user' -tftp dir allow tftp access to files in dir [-net user] -bootp file advertise file in BOOTP replies -smb dir allow SMB access to files in 'dir' [-net user] -redir [tcp|udp]:host-port:[guest-host]:guest-port redirect TCP or UDP connections from host to guest [-net user] -bt hci,null dumb bluetooth HCI - doesn't respond to commands -bt hci,host[:id] use host's HCI with the given name -bt hci[,vlan=n] emulate a standard HCI in virtual scatternet 'n' -bt vhci[,vlan=n] add host computer to virtual scatternet 'n' using VHCI -bt device:dev[,vlan=n] emulate a bluetooth device 'dev' in scatternet 'n' i386 target only: -win2k-hack use it when installing Windows 2000 to avoid a disk full bug -rtc-td-hack use it to fix time drift in Windows ACPI HAL -no-fd-bootchk disable boot signature checking for floppy disks -no-acpi disable ACPI -no-hpet disable HPET Linux boot specific: -kernel bzImage use 'bzImage' as kernel image -append cmdline use 'cmdline' as kernel command line -initrd file use 'file' as initial ram disk Debug/Expert options: -serial dev redirect the serial port to char device 'dev' -parallel dev redirect the parallel port to char device 'dev' -monitor dev redirect the monitor to char device 'dev' -pidfile file write PID to 'file' -S freeze CPU at startup (use 'c' to start execution) -s wait gdb connection to port -p port set gdb connection port [default=1234] -d item1,... output log to /tmp/qemu.log (use -d ? for a list of log items) -hdachs c,h,s[,t] force hard disk 0 physical geometry and the optional BIOS translation (t=none or lba) (usually qemu can guess them) -L path set the directory for the BIOS, VGA BIOS and keymaps -bios file set the filename for the BIOS -no-kvm disable KVM hardware virtualization -no-kvm-irqchip disable KVM kernel mode PIC/IOAPIC/LAPIC -no-kvm-pit disable KVM kernel mode PIT -no-kvm-pit-reinjection disable KVM kernel mode PIT interrupt reinjection -enable-nesting enable support for running a VM inside the VM (AMD only) -pcidevice host=bus:dev.func[,dma=none][,name=string] expose a PCI device to the guest OS. dma=none: don't perform any dma translations (default is to use an iommu) 'string' is used in log output. -no-reboot exit instead of rebooting -no-shutdown stop before shutdown -loadvm [tag|id] start right away with a saved state (loadvm in monitor) -daemonize daemonize QEMU after initializing -tdf inject timer interrupts that got lost -kvm-shadow-memory megs set the amount of shadow pages to be allocated -mem-path set the path to hugetlbfs/tmpfs mounted directory, also enables allocation of guest memory with huge pages -mem-prealloc toggles preallocation of -mem-path backed physical memory at startup. Default is enabled. -option-rom rom load a file, rom, into the option ROM space -clock force the use of the given methods for timer alarm. To see what timers are available use -clock ? -localtime set the real time clock to local time [default=utc] -startdate select initial date of the clock -icount [N|auto] enable virtual instruction counter with 2^N clock ticks per instruction -echr chr set terminal escape character instead of ctrl-a -virtioconsole c set virtio console -show-cursor show cursor -tb-size n set TB size -incoming p prepare for incoming migration, listen on port p During emulation, the following keys are useful: ctrl-alt-f toggle full screen ctrl-alt-n switch to virtual console 'n' ctrl-alt toggle mouse and keyboard grab When using -nographic, press 'ctrl-a h' to get some help. Packages installed: libvirt-debug-0.6.3-1mdv2009.1 python-virtinst-0.400.3-1mdv2009.1 libvirt-utils-0.6.3-1mdv2009.1 virt-manager-0.7.0-2mdv2009.1 virt-viewer-0.0.3-2mdv2009.0 python-libvirt-0.6.3-1mdv2009.1 lib64virt0-0.6.3-1mdv2009.1 What could be wrong here? -- Frederik Himpe

2 1

[libvirt] [PATCH 0/1] Added Networking API to change/create/etc. hostonly/internal network in VirtualBox
by Pritesh Kothari 06 May '09

06 May '09

Hi All: [PATCH 0/1]: Contains sample xml file showing features supported so far. [PATCH 1/1]: Contains the patch for adding Networking API to change/create/etc. hostonly/internal network in VirtualBox Regards, Pritesh

2 2

[libvirt] [PATCH 1/2] Support for 'hostonly' and 'internal' network in VirtualBox
by Pritesh Kothari 06 May '09

06 May '09

Hi All, I have added the support for hostonly and internal network as suggested on irc and list and am attaching the patch for same. PATCH 1/2: contains changes in the xml parsing/formatting on libvirt side namely adding a mode attribute to <source> tag in domain/interface xml PATCH 2/2: contains changes in the vbox driver itself. Regards, Pritesh

2 1

[libvirt] [PATCH 1/1] Added Networking API to change/create/etc. hostonly/internal network in VirtualBox
by Pritesh Kothari 06 May '09

06 May '09

Hi All: [PATCH 0/1]: Contains sample xml file showing features supported so far. [PATCH 1/1]: Contains the patch for adding Networking API to change/create/etc. hostonly/internal network in VirtualBox Regards, Pritesh

2 1

[libvirt] USB pass-through only works after many disconnects at the beginning
by Felix Schwarz 06 May '09

06 May '09

Hi, I want to pass through a USB device to a KVM guest (Windows 2000). My host OS is a standard Fedora 10, 64 bit. However, the device is only passed to the guest when I use qemu monitor directly and only after several tries... The device has no Linux drivers ('AVM GmbH ISDN-Controller FRITZ!Card v2.0') so I think I don't have to tell the kernel not to use it. I have KVM 74 and tried initially with libvirt 0.5.1 so from what I read this should be ok. At first I added this to my domain xml via virsh: <hostdev mode='subsystem' type='usb'> <source> <vendor id='0x....'/> <product id='0x....'/> </source> </hostdev> However, the device did not turn up in the guest. Therefore I updated libvirt/-python 0.6.1 and python-virtinst from updates-testing and recompiled virt-manager 0.7 from F11 because I read that virt-manager 0.7 had some support for USB pass-through. Now I had some new UI, but it turned out that using that did not produce better results... After that I called qemu-kvm directly with '-usb -usbdevice host:...:...'. qemu produced this output on my terminal: husb: open device 2.9 husb: config #1 need -1 husb: 1 interfaces claimed for configuration -1 husb: grabbed usb device 2.9 husb: config #1 need 1 USBDEVFS_DISCONNECT: No such device husb: device 2.9 disconnected When I used qemu monitor to re-attach the device, something liek that happened (re-attached multiple times): husb: open device 2.11 husb: config #1 need -1 husb: 1 interfaces claimed for configuration -1 husb: grabbed usb device 2.11 husb: config #1 need 1 husb: 1 interfaces claimed for configuration 1 husb: config #1 need 1 USBDEVFS_DISCONNECT: No such device husb: config #1 need 1 USBDEVFS_DISCONNECT: No such device husb: config #1 need 1 USBDEVFS_DISCONNECT: No such device husb: device 2.11 disconnected ... Eventually (after several tries), the device was attached and no more disconnects happened. After that I could use the device as expected. However plain qemu is not what I would like to have. I read about 'virsh attach-device' so I tried to use my usb hostdev snippet but to no avail (even after 10s of tries). So how can I use USB pass through? Could this be Windows-related? Judging from all the other threads (like [1]) it should be really easy to use. fs [1] https://www.redhat.com/archives/libvir-list/2009-February/msg00387.html

3 4

[libvirt] (no subject)
by mike.hinz＠yr20.com 05 May '09

05 May '09

I had originally posted a plea for help on the virt-manager related list, but the same issue exists with virsh and this list seems to have higher traffic. I'd been chatting on the irc channel with various people about this, but wanted to put this on the mail list for the record and to also raise a philosophical question. First, we've confirmed that Fedora 11 preview (fully updated) has an issue with properly mounting NFS shares without the noacl option. Fedora 10 works fine with regard to mounting NFS shares. That issue causes problems when attempting to write files to an NFS share and more importantly, for this group, causes problems when attempting to create an NFS based storage pool and then subsequently creating volumes on said NFS share. That bug is 499178. This was all figured out with Openfiler support and with the support of the fine folks working on libvirt and virt-manager. There is a somewhat ugly workaround for this and I can describe it in case anyone is interested. Now comes the philosophy piece: It seems like that the way that libvirt and virt-manager want to handle storage is to be able to fully control the mounting process. That seems to me to be a very nice thing as you have a single way to manage storage along with your VMs. However, at least with respect to NFS, there are a number of parameters that people might need to give to the mount command for things like performance optimization and who knows what else. Note that this is a consideration completely aside from the bug issue above. It seems to me that either you have to just use the user created mount points, either manually or by fstab OR you have to allow all the possible options to be passed if you want to fully control the process. Right now, it's tricky to do, if you want to get paras passed for tuning or whatever and requires some tricky handwork to get it done. One way to accomplish this from the user input perspective would be to add an 'options' input field that could take multiple parameters on the step 2 of 2 page of the 'add storage pool' function. So, is my thinking all screwed up about this or is there merit to this concept? I tend to think, at least right now, that the logical thing is to either have the user create the mount points through the normal mechanisms that have in place roughly forever and then you just ask for what these are with respect to the NFS pools or you have to allow for user to pass the required parameters via your mechanism, although there are certainly many, many options when it comes to NFS. Thanks for all the help to date! I hope the above is helpful! Regards, Mike Hinz President YR20 1718 Fry Road, Suite 440 Houston, TX 77084 mike.hinz(a)yr20.com 832-225-1293 (o) 713-594-3095 (m) 832-550-2657 (f)

1 0

[libvirt] pool default not starting on f10
by Jason Guiditta 05 May '09

05 May '09

Hello, hopefully this is something simple I just don't understand how to debug, but right now, it is keeping me from creating any new VMs. As stated in the subject, I am running F10, on x86_64 arch. Libvirt version is 0.6.1-5.fc10. Using virt-manager to create a new vm, when I get to the step to create a disk image, I get the error: 'Storage parameter error. internal error storage pool is not active'. Running virsh pool-start default gets me: error: Failed to start pool default error: out of memory Note there is no reason my machine would be out of memory, is has 6GB and nothing else running (and tried this right after booting) virsh pool-list --all shows default is inactive and 'yes' for autostart I would be happy to file a bug if needed, but thought I would throw this out to the list first to see if I was just unaware of some simple fix. Let me know if there are other details I can provide to make this easier to solve. Thanks, -j

2 6

[libvirt] [PATCH v2] Don't throw away QEMU startup errors when migrating.
by Cole Robinson 05 May '09

05 May '09

The attached patch removes the error message squashing that can take place in the qemu driver when launching a new VM via restore or migrate. Thanks, Cole

2 2

[libvirt] [PATCH] Test driver VolCreate cleanups
by Cole Robinson 05 May '09

05 May '09

The attached patch fixes some small issues with VolCreateXML implementation in the test driver. Thanks, Cole

2 2

[libvirt] [PATCH] Remove some old debugging code
by Cole Robinson 05 May '09

05 May '09

The attached patch removes some debugging code accidentally applied with my async vol creation patches. Thanks, Cole

2 2

[libvirt] [PATCH] Use correct error code in public VolCreateXML
by Cole Robinson 05 May '09

05 May '09

The attached patch fixes a use of an incorrect error code in libvirt.c VolCreateXML. Thanks, Cole

2 2

[libvirt] [PATCH 0/1] NPIV node device backend
by David Allan 05 May '09

05 May '09

Here is an initial implementation of the NPIV support backend using the node device API. It only implements create. Destroy should be fairly straightforward, but I want to get the patch out to the list for comment while I continue to work. The code currently requires the device name and host number to be specified in the XML, but it does not use those values. I will clean that up. The user will not be able to specify either of those values, as they are created by the system when it discovers the newly created device. <device> <name>foo</name> <parent>pci_10df_fe00_scsi_host_0</parent> <capability type='scsi_host'> <host>99999</host> <capability type='fc_host'> <wwnn>c0001b0000000018</wwnn> <wwpn>c0001b1000000018</wwpn> </capability> </capability> </device>

2 2

[libvirt] Patch: Fix build with -Werror=format-security cflag
by Frederik Himpe 05 May '09

05 May '09

This patch is needed to build libvirt with -Werror=format-security gcc cflag, which is used by default on Mandriva diff -ur libvirt-0.6.3.orig/src/virsh.c libvirt-0.6.3/src/virsh.c --- libvirt-0.6.3.orig/src/virsh.c 2009-04-15 21:43:26.000000000 +0200 +++ libvirt-0.6.3/src/virsh.c 2009-05-02 13:56:11.000000000 +0200 @@ -4463,7 +4463,7 @@ } /* Print this device */ - vshPrint(ctl, indentBuf); + vshPrint(ctl, "%s", indentBuf); vshPrint(ctl, "%s\n", devices[devid]); @@ -4487,7 +4487,7 @@ /* If there is a child device, then print another blank line */ if (nextlastdev != -1) { - vshPrint(ctl, indentBuf); + vshPrint(ctl, "%s", indentBuf); vshPrint(ctl, " |\n"); } @@ -4511,7 +4511,7 @@ /* If there was no child device, and we're the last in * a list of devices, then print another blank line */ if (nextlastdev == -1 && devid == lastdev) { - vshPrint(ctl, indentBuf); + vshPrint(ctl, "%s", indentBuf); vshPrint(ctl, "\n"); } }

2 1

[libvirt] [PATCH 0/7] Improve QEMU process startup error reporting
by Cole Robinson 04 May '09

04 May '09

Error reporting for failed QEMU processes isn't too great at the moment. A couple cases I have hit: - Trying to start a VM with a cdrom pointing to a nonexistent file. The svirt hook function choked on this and indicated as much in the domain log, but I still had to wait for the pidfile timeout (10 seconds), and the only error reported was 'Domain did not show up'. - QEMU dies quickly at startup. Happened with the pulse audio issue, or if we build cli args incorrectly. The reporting here varies: if QEMU dies before writing out the pidfile, the above case applies. If the pidfile is written, we timeout in ReadLogOutput, and it's hit or miss whether we even report the logfile output to the user. The following patches improve the above cases by ensuring we never have to wait for a timeout, and that the logfile output is always reported in the error message. Cole Robinson (7): Add documentation for __virExec and virExec virExec: Delay daemonizing as long as possible. Add helper function virExecDaemonize Report qemu log data if we fail to daemonize the process. Add pidfile argument to __virExec Check that QEMU is still alive while reading startup output. Don't throw away QEMU startup errors when migrating. src/libvirt_private.syms | 2 +- src/proxy_internal.c | 16 +-- src/qemu_conf.c | 7 - src/qemu_driver.c | 105 +++++++---- src/remote_internal.c | 15 +-- src/uml_driver.c | 11 +- src/util.c | 186 +++++++++++++++----- src/util.h | 15 ++- .../qemuxml2argvdata/qemuxml2argv-boot-cdrom.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-boot-floppy.args | 2 +- .../qemuxml2argv-boot-network.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-bootloader.args | 2 +- .../qemuxml2argv-clock-localtime.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-clock-utc.args | 2 +- .../qemuxml2argv-console-compat.args | 2 +- .../qemuxml2argv-disk-cdrom-empty.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-cdrom.args | 2 +- .../qemuxml2argv-disk-drive-boot-cdrom.args | 2 +- .../qemuxml2argv-disk-drive-boot-disk.args | 2 +- .../qemuxml2argv-disk-drive-cache-v1-none.args | 2 +- .../qemuxml2argv-disk-drive-cache-v1-wb.args | 2 +- .../qemuxml2argv-disk-drive-cache-v1-wt.args | 2 +- .../qemuxml2argv-disk-drive-cache-v2-none.args | 2 +- .../qemuxml2argv-disk-drive-cache-v2-wb.args | 2 +- .../qemuxml2argv-disk-drive-cache-v2-wt.args | 2 +- .../qemuxml2argv-disk-drive-fmt-qcow.args | 2 +- .../qemuxml2argv-disk-drive-shared.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-floppy.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-disk-many.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-disk-usb.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-virtio.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-disk-xenvbd.args | 2 +- .../qemuxml2argv-graphics-sdl-fullscreen.args | 2 +- .../qemuxml2argv-graphics-sdl.args | 2 +- .../qemuxml2argv-graphics-vnc-sasl.args | 2 +- .../qemuxml2argv-graphics-vnc-tls.args | 2 +- .../qemuxml2argv-graphics-vnc.args | 2 +- .../qemuxml2argv-hostdev-pci-address.args | 2 +- .../qemuxml2argv-hostdev-usb-address.args | 2 +- .../qemuxml2argv-hostdev-usb-product.args | 2 +- .../qemuxml2argv-input-usbmouse.args | 2 +- .../qemuxml2argv-input-usbtablet.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-input-xen.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-migrate.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-minimal.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-misc-acpi.args | 2 +- .../qemuxml2argv-misc-no-reboot.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-misc-uuid.args | 2 +- .../qemuxml2argv-net-eth-ifname.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-net-eth.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-net-user.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-net-virtio.args | 2 +- .../qemuxml2argv-parallel-tcp.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-restore-v1.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-restore-v2.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-dev.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-file.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-many.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-pty.args | 2 +- .../qemuxml2argv-serial-tcp-telnet.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-tcp.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-udp.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-serial-unix.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-serial-vc.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-sound.args | 2 +- 65 files changed, 288 insertions(+), 183 deletions(-)

2 18

[libvirt] Virtio Disk Drivers and Windows
by Aaron Clausen 04 May '09

04 May '09

I've imported a Windows Server 2003 as a guest using the IDE drivers. When I try to use the virtio drivers instead, Windows just blue screens. Am I stuck with the IDE drivers under Windows? -- Aaron Clausen mightymartianca(a)gmail.com

1 1

[libvirt] [PATCH] Fix a comment typo in storage_backend_fs.c
by Cole Robinson 04 May '09

04 May '09

This comment hid from my attempt to 'grep FILE_BOCHS', which is the actual name of the constant. Thanks, Cole

2 2

[libvirt] VM image running on 2 Intels
by Zvi Dubitzky 30 Apr '09

30 Apr '09

I have 2 types of hosts as below ( each is 64 bits processor . number of processor is not equal . First has 1G memory . the second 3 G memor): IBM blade HS21 with Intel Xeon L5320 (1,86Ghz) . 8 processors (blade) IBM desktop with Intel(R) Core(TM)2 Duo CPU E6550 @ 2.33GHz I have generated an Ubuntu 8.10 client image on the first . It runs smooth on the first via virsh . It runs ok on the second but not well via virsh ( define + start <xml> ). just by a direct : sudo qemu-system-x86_64 call . If I generate a similar image on the second host (using same .iso Ubuntu CD image) then it runs on the first and second host smoothly via virsh (define + start <xml>) Is there an explanation for that ? Also : please comment if the below qemu-system-x86_64 explicit image start of a VM and the same done by a virsh command by : define + start <.xml> are equivalent and if not what change should I make to the .xml file (below) 1. sudo qemu-system-x86_64 -hda <path/to/image/file> -m 512 -net nic, model=e1000 -nic tap 2. The xml file thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] [PATCH 1/2] support for multiple graphics devices
by Pritesh Kothari 30 Apr '09

30 Apr '09

Hi All, I have added support for multiple graphics devices, the patches are as below. I have checked them against current cvs head and works fine PATCH 1/2: contains changes in libvirt for multiple graphics devices PATCH 2/2: contains corresponding changes in qemu driver. Regards, Pritesh

2 1

[libvirt] libvirt will create two guest with the same name, but with different UUIDs
by Kaitlin Rupert 30 Apr '09

30 Apr '09

Hi, libvirt allows you to create two guests with same name, but with different UUIDs. This causes some problems in the reporting though. virsh will report that a guest is running (when it isn't). This also causes some confusion for the dominfo call - virsh will only return info for one of the guests. I've seen similar issues when using the virDomainLookupByName() API. I suspect all the issues are the same bug, but I haven't had a chance to track down where. Thanks! # virsh version Compiled against library: libvir 0.6.3 Using library: libvir 0.6.3 Using API: QEMU 0.6.3 Running hypervisor: QEMU 0.9.1 # virsh list --all Id Name State ---------------------------------- 109 demo2 running - domU1 shut off - test_domain shut off # virsh dumpxml demo2 <domain type='kvm' id='109'> <name>demo2</name> <uuid>4dea24b3-1d52-d8f3-2516-782e98a23fa0</uuid> <memory>131072</memory> <currentMemory>131072</currentMemory> <vcpu>1</vcpu> <os> <type arch='i686' machine='pc'>hvm</type> <boot dev='hd'/> </os> <clock offset='utc'/> <on_poweroff>restart</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/bin/qemu-kvm</emulator> <disk type='file' device='disk'> <source file='/tmp/default-kvm-dimage'/> <target dev='hda' bus='ide'/> </disk> <interface type='user'> <mac address='24:42:53:21:52:45'/> </interface> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='5900' autoport='yes'/> </devices> </domain> # cat demo2 <domain type='kvm' id='55'> <name>demo2</name> <uuid>4dea24b3-1d52-d8f3-2516-782e98a23fa3</uuid> <memory>131072</memory> <currentMemory>131072</currentMemory> <vcpu>1</vcpu> <os> <type arch='i686' machine='pc'>hvm</type> <boot dev='hd'/> </os> <clock offset='utc'/> <on_poweroff>restart</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/bin/qemu-kvm</emulator> <disk type='file' device='disk'> <source file='/tmp/default-kvm-dimage'/> <target dev='hda' bus='ide'/> </disk> <interface type='user'> <mac address='24:42:53:21:52:45'/> </interface> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='5902' autoport='yes'/> </devices> </domain> # virsh define demo2 Domain demo2 defined from demo2 # virsh list --all Id Name State ---------------------------------- 109 demo2 running - demo2 running - domU1 shut off - test_domain shut off # ps -ef | grep qemu root 11464 1 96 09:04 ? 00:02:12 /usr/bin/qemu-kvm -S -M pc -cpu qemu32 -m 128 -smp 1 -name demo2 -uuid 4dea24b3-1d52-d8f3-2516-782e98a23fa0 -monitor pty -pidfile /var/run/libvirt/qemu//demo2.pid -no-acpi -boot c -drive file=/tmp/default-kvm-dimage,if=ide,index=0,boot=on -net nic,macaddr=24:42:53:21:52:45,vlan=0 -net user,vlan=0 -serial none -parallel none -usb -vnc 127.0.0.1:0 root 11483 27407 0 09:06 pts/1 00:00:00 grep qemu # -- Kaitlin Rupert IBM Linux Technology Center kaitlin(a)linux.vnet.ibm.com

2 1

[libvirt] PATCH: Add a new error code for invalid operations
by Daniel P. Berrange 29 Apr '09

29 Apr '09

A number of operations are only valid when the corresponding object is in a particular state. eg, you can only suspend a domain, if it is currently running. We don't have a very consistent way of reporting this problem across drivers, so this patch introduces a new error code for this situation called VIR_ERR_INVALID_OPERATION. This code is intended to be used if the operation is supported, but not applicable for the object at this time. It updates the Xen and QEMU drivers to use this error code. Daniel diff -r 2e97361b7242 include/libvirt/virterror.h --- a/include/libvirt/virterror.h Mon Apr 27 11:44:32 2009 +0100 +++ b/include/libvirt/virterror.h Mon Apr 27 12:12:23 2009 +0100 @@ -157,6 +157,7 @@ typedef enum { VIR_ERR_INVALID_NODE_DEVICE,/* invalid node device object */ VIR_ERR_NO_NODE_DEVICE,/* node device not found */ VIR_ERR_NO_SECURITY_MODEL, /* security model not found */ + VIR_ERR_OPERATION_INVALID, /* operation is not applicable at this time */ } virErrorNumber; /** diff -r 2e97361b7242 src/qemu_driver.c --- a/src/qemu_driver.c Mon Apr 27 11:44:32 2009 +0100 +++ b/src/qemu_driver.c Mon Apr 27 12:12:23 2009 +0100 @@ -1325,7 +1325,7 @@ static int qemudStartVMDaemon(virConnect FD_ZERO(&keepfd); if (virDomainIsActive(vm)) { - qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("VM is already active")); return -1; } @@ -2214,7 +2214,7 @@ static int qemudDomainSuspend(virDomainP goto cleanup; } if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } @@ -2267,7 +2267,7 @@ static int qemudDomainResume(virDomainPt goto cleanup; } if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } @@ -2755,7 +2755,7 @@ static int qemudDomainSave(virDomainPtr } if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } @@ -2890,7 +2890,7 @@ static int qemudDomainSetVcpus(virDomain } if (virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_NO_SUPPORT, "%s", + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("cannot change vcpu count of an active domain")); goto cleanup; } @@ -2942,8 +2942,16 @@ qemudDomainPinVcpu(virDomainPtr dom, vm = virDomainFindByUUID(&driver->domains, dom->uuid); qemuDriverUnlock(driver); - if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_ARG, + if (!vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(dom->uuid, uuidstr); + qemudReportError(dom->conn, dom, NULL, VIR_ERR_NO_DOMAIN, + _("no domain with matching uuid '%s'"), uuidstr); + goto cleanup; + } + + if (!virDomainIsActive(vm)) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s",_("cannot pin vcpus on an inactive domain")); goto cleanup; } @@ -3003,8 +3011,16 @@ qemudDomainGetVcpus(virDomainPtr dom, vm = virDomainFindByUUID(&driver->domains, dom->uuid); qemuDriverUnlock(driver); - if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_ARG, + if (!vm) { + char uuidstr[VIR_UUID_STRING_BUFLEN]; + virUUIDFormat(dom->uuid, uuidstr); + qemudReportError(dom->conn, dom, NULL, VIR_ERR_NO_DOMAIN, + _("no domain with matching uuid '%s'"), uuidstr); + goto cleanup; + } + + if (!virDomainIsActive(vm)) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s",_("cannot pin vcpus on an inactive domain")); goto cleanup; } @@ -3248,7 +3264,7 @@ static int qemudDomainRestore(virConnect vm = virDomainFindByName(&driver->domains, def->name); if (vm) { if (virDomainIsActive(vm)) { - qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError(conn, NULL, NULL, VIR_ERR_OPERATION_INVALID, _("domain is already active as '%s'"), vm->def->name); goto cleanup; } else { @@ -3503,7 +3519,7 @@ static int qemudDomainUndefine(virDomain } if (virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("cannot delete active domain")); goto cleanup; } @@ -3900,7 +3916,7 @@ static int qemudDomainAttachDevice(virDo if (!virDomainIsActive(vm)) { qemuDriverUnlock(driver); - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("cannot attach device on inactive domain")); goto cleanup; } @@ -4052,7 +4068,7 @@ static int qemudDomainDetachDevice(virDo if (!virDomainIsActive(vm)) { qemuDriverUnlock(driver); - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("cannot detach device on inactive domain")); goto cleanup; } @@ -4212,7 +4228,7 @@ qemudDomainBlockStats (virDomainPtr dom, goto cleanup; } if (!virDomainIsActive (vm)) { - qemudReportError (dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError (dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } @@ -4351,7 +4367,7 @@ qemudDomainInterfaceStats (virDomainPtr } if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } @@ -4497,7 +4513,7 @@ qemudDomainMemoryPeek (virDomainPtr dom, } if (!virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError(dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } @@ -4824,7 +4840,7 @@ qemudDomainMigratePerform (virDomainPtr } if (!virDomainIsActive(vm)) { - qemudReportError (dom->conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + qemudReportError (dom->conn, dom, NULL, VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); goto cleanup; } diff -r 2e97361b7242 src/virterror.c --- a/src/virterror.c Mon Apr 27 11:44:32 2009 +0100 +++ b/src/virterror.c Mon Apr 27 12:12:23 2009 +0100 @@ -1018,6 +1018,12 @@ virErrorMsg(virErrorNumber error, const else errmsg = _("Security model not found: %s"); break; + case VIR_ERR_OPERATION_INVALID: + if (info == NULL) + errmsg = _("Requested operation is not valid"); + else + errmsg = _("Requested operation is not valid: %s"); + break; } return (errmsg); } diff -r 2e97361b7242 src/xen_internal.c --- a/src/xen_internal.c Mon Apr 27 11:44:32 2009 +0100 +++ b/src/xen_internal.c Mon Apr 27 12:12:23 2009 +0100 @@ -1068,9 +1068,14 @@ xenHypervisorGetSchedulerType(virDomainP } priv = (xenUnifiedPrivatePtr) domain->conn->privateData; - if (priv->handle < 0 || domain->id < 0) { + if (priv->handle < 0) { virXenErrorFunc(domain->conn, VIR_ERR_INTERNAL_ERROR, __FUNCTION__, - "priv->handle or domain->id invalid", 0); + "priv->handle invalid", 0); + return NULL; + } + if (domain->id < 0) { + virXenError(domain->conn, VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); return NULL; } @@ -1143,9 +1148,14 @@ xenHypervisorGetSchedulerParameters(virD } priv = (xenUnifiedPrivatePtr) domain->conn->privateData; - if (priv->handle < 0 || domain->id < 0) { + if (priv->handle < 0) { virXenErrorFunc(domain->conn, VIR_ERR_INTERNAL_ERROR, __FUNCTION__, - "priv->handle or domain->id invalid", 0); + "priv->handle invalid", 0); + return -1; + } + if (domain->id < 0) { + virXenError(domain->conn, VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); return -1; } @@ -1241,9 +1251,14 @@ xenHypervisorSetSchedulerParameters(virD } priv = (xenUnifiedPrivatePtr) domain->conn->privateData; - if (priv->handle < 0 || domain->id < 0) { - virXenErrorFunc (domain->conn, VIR_ERR_INTERNAL_ERROR, __FUNCTION__, - "priv->handle or domain->id invalid", 0); + if (priv->handle < 0) { + virXenErrorFunc(domain->conn, VIR_ERR_INTERNAL_ERROR, __FUNCTION__, + "priv->handle invalid", 0); + return -1; + } + if (domain->id < 0) { + virXenError(domain->conn, VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); return -1; } diff -r 2e97361b7242 src/xend_internal.c --- a/src/xend_internal.c Mon Apr 27 11:44:32 2009 +0100 +++ b/src/xend_internal.c Mon Apr 27 12:12:23 2009 +0100 @@ -2970,7 +2970,7 @@ xenDaemonDomainSuspend(virDomainPtr doma } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } @@ -2997,7 +2997,7 @@ xenDaemonDomainResume(virDomainPtr domai } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } @@ -3025,7 +3025,7 @@ xenDaemonDomainShutdown(virDomainPtr dom } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } @@ -3054,7 +3054,7 @@ xenDaemonDomainReboot(virDomainPtr domai } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } @@ -3085,7 +3085,7 @@ xenDaemonDomainDestroy(virDomainPtr doma } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } @@ -3160,7 +3160,7 @@ xenDaemonDomainSave(virDomainPtr domain, } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } @@ -3197,7 +3197,7 @@ xenDaemonDomainCoreDump(virDomainPtr dom } if (domain->id < 0) { - virXendError(domain->conn, VIR_ERR_INVALID_ARG, + virXendError(domain->conn, VIR_ERR_OPERATION_INVALID, _("Domain %s isn't running."), domain->name); return(-1); } -- |: 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] [PATCH] Remove unused file volume format 'dir'
by Cole Robinson 29 Apr '09

29 Apr '09

It isn't documented, detected, and barely has any code accomodating it. Appears to be left over from an earlier iteration of the storage APIs. Thanks, Cole

2 2

[libvirt] [PATCH 2/2] Support for 'hostonly' and 'internal' network in VirtualBox
by Pritesh Kothari 29 Apr '09

29 Apr '09

Hi All, I have added the support for hostonly and internal network as suggested on irc and list and am attaching the patch for same. PATCH 1/2: contains changes in the xml parsing/formatting on libvirt side namely adding a mode attribute to <source> tag in domain/interface xml PATCH 2/2: contains changes in the vbox driver itself. Regards, Pritesh

1 0

[libvirt] [PATCH] fix soem missing directories and ownership in spec file
by Daniel Veillard 28 Apr '09

28 Apr '09

The spec file in CVs was missing the creation fo some directories and forgot to take ownership of the cache subdir used for memory dumps with QEmu/KVM 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: Allow for quiet build output
by Daniel P. Berrange 28 Apr '09

28 Apr '09

The following patch is a slightly gross hack that puts a wrapper around libtool to make it STFU, and only print the name of the output file, and associated input files. Since we can't change the automake generated Makefile.in to add '@' in front of all libtool commands, this sill won't be quiet by default. To make it really quiet you need to run 'make -s' eg, a short transcript with this applied.... $ make -s Making all in gnulib/lib (CC) -o c-ctype.lo c-ctype.c (CC) -o asnprintf.lo asnprintf.c (CC) -o physmem.lo physmem.c (CC) -o printf-args.lo printf-args.c (CC) -o printf-parse.lo printf-parse.c (CC) -o tempname.lo tempname.c (CC) -o vasnprintf.lo vasnprintf.c (LD) -o libgnu.la c-ctype.lo asnprintf.lo physmem.lo printf-args.lo printf-parse.lo tempname.lo vasnprintf.lo Making all in include Making all in libvirt Making all in src (CC) -o bridge.lo bridge.c (CC) -o buf.lo buf.c (CC) -o conf.lo conf.c (CC) -o event.lo event.c (CC) -o hash.lo hash.c (CC) -o iptables.lo iptables.c (CC) -o logging.lo logging.c (CC) -o memory.lo memory.c (CC) -o pci.lo pci.c (CC) -o qparams.lo qparams.c (CC) -o threads.lo threads.c (CC) -o uuid.lo uuid.c (CC) -o util.lo util.c (CC) -o virterror.lo virterror.c (CC) -o xml.lo xml.c (LD) -o libvirt_util.la bridge.lo buf.lo conf.lo event.lo hash.lo iptables.lo logging.lo memory.lo pci.lo qparams.lo threads.lo uuid.lo util.lo virterror.lo xml.lo (CC) -o libvirt_driver_la-driver.lo driver.c (CC) -o libvirt_driver_la-datatypes.lo datatypes.c (CC) -o libvirt_driver_la-domain_event.lo domain_event.c (CC) -o libvirt_driver_la-stats_linux.lo stats_linux.c (CC) -o libvirt_driver_la-libvirt.lo libvirt.c (CC) -o libvirt_driver_la-capabilities.lo capabilities.c (CC) -o libvirt_driver_la-domain_conf.lo domain_conf.c (CC) -o libvirt_driver_la-nodeinfo.lo nodeinfo.c (CC) -o libvirt_driver_la-network_conf.lo network_conf.c (CC) -o libvirt_driver_la-storage_conf.lo storage_conf.c (CC) -o libvirt_driver_la-node_device_conf.lo node_device_conf.c (LD) -o libvirt_driver.la libvirt_driver_la-driver.lo libvirt_driver_la-datatypes.lo libvirt_driver_la-domain_event.lo libvirt_driver_la-stats_linux.lo libvirt_driver_la-libvirt.lo libvirt_driver_la-capabilities.lo libvirt_driver_la-domain_conf.lo libvirt_driver_la-nodeinfo.lo libvirt_driver_la-network_conf.lo libvirt_driver_la-storage_conf.lo libvirt_driver_la-node_device_conf.lo (CC) -o test.lo test.c (LD) -o libvirt_driver_test.la test.lo (CC) -o libvirt_driver_remote_la-remote_internal.lo remote_internal.c (CC) -o libvirt_driver_remote_la-remote_protocol.lo ../qemud/remote_protocol.c (LD) -o libvirt_driver_remote.la libvirt_driver_remote_la-remote_internal.lo libvirt_driver_remote_la-remote_protocol.lo (CC) -o libvirt_driver_xen_la-proxy_internal.lo proxy_internal.c Aside from changing libtool and automake themselves, I don't see any easier way to do this better. b/mylibtool | 62 ++++++++++++++++++++++++++++ examples/domain-events/events-c/Makefile.am | 3 + examples/hellolibvirt/Makefile.am | 3 + gnulib/lib/Makefile.am | 3 + proxy/Makefile.am | 3 + qemud/Makefile.am | 3 + src/Makefile.am | 3 + tests/Makefile.am | 3 + tools/Makefile.am | 3 + 9 files changed, 86 insertions(+) Daniel diff -r f60b121571f2 examples/domain-events/events-c/Makefile.am --- a/examples/domain-events/events-c/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/examples/domain-events/events-c/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,3 +1,6 @@ +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + INCLUDES = -I@top_srcdir@/include noinst_PROGRAMS = event-test event_test_CFLAGS = $(WARN_CFLAGS) diff -r f60b121571f2 examples/hellolibvirt/Makefile.am --- a/examples/hellolibvirt/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/examples/hellolibvirt/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,3 +1,6 @@ +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + INCLUDES = -I@top_srcdir@/include noinst_PROGRAMS = hellolibvirt hellolibvirt_CFLAGS = $(WARN_CFLAGS) diff -r f60b121571f2 gnulib/lib/Makefile.am --- a/gnulib/lib/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/gnulib/lib/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -11,6 +11,9 @@ # Generated by gnulib-tool. # Reproduce by: gnulib-tool --import --dir=. --lib=libgnu --source-base=gnulib/lib --m4-base=gnulib/m4 --doc-base=doc --tests-base=gnulib/tests --aux-dir=build-aux --with-tests --lgpl=2 --libtool --macro-prefix=gl --no-vc-files c-ctype close connect getaddrinfo gethostname getpass gettext inet_pton ioctl mkstemp mktempd perror physmem poll posix-shell random_r recv send setsockopt socket stpcpy strerror strndup strsep sys_stat time_r useless-if-before-free vasprintf vc-list-files verify +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + AUTOMAKE_OPTIONS = 1.5 gnits SUBDIRS = diff -r f60b121571f2 mylibtool --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/mylibtool Tue Apr 28 12:26:44 2009 +0100 @@ -0,0 +1,62 @@ +#!/bin/sh + +mode=libtool +cfiles="" +ofiles="" +afiles="" + +wantnext=0 +for v in "$@" +do + case $v + in + --mode=compile) + mode=CC + ;; + --mode=link) + mode=LD + ;; + esac + + case $v + in + *.c) + cfiles="$cfiles $v" + ;; + *.o) + if [ "$mode" = "LD" -o "$wantnext" = "1" ]; then + ofiles="$ofiles $v" + fi + ;; + *.lo) + if [ "$mode" = "LD" -o "$wantnext" = "1" ]; then + ofiles="$ofiles $v" + fi + ;; + esac + + if [ "$mode" = "LD" -a "$wantnext" = "1" ]; then + afiles="$afiles $v" + fi + + if [ "$v" = "-o" ]; then + wantnext=1 + else + wantnext=0 + fi +done + +#afiles=`echo $afiles | sed -e 's/ /\n/g' | sort | uniq | tr '\n' ' '` +#cfiles=`echo $cfiles | sed -e 's/ /\n/g' | sort | uniq | tr '\n' ' '` +#ofiles=`echo $ofiles | sed -e 's/ /\n/g' | sort | uniq | tr '\n' ' '` + +args="" +test -n "$afiles" && args="$args -o$afiles" +test -n "$ofiles" -a "$mode" = "CC" && args="$args -o" +test -n "$ofiles" && args="$args$ofiles" +test -n "$cfiles" && args="$args$cfiles" + +echo "($mode)$args" + +here=`dirname $0` +exec $here/libtool "$@" diff -r f60b121571f2 proxy/Makefile.am --- a/proxy/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/proxy/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,5 +1,8 @@ ## Process this file with automake to produce Makefile.in +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + if WITH_PROXY INCLUDES = -I$(top_srcdir)/gnulib/lib -I../gnulib/lib \ -I$(top_builddir)/include -I@top_srcdir@/include \ diff -r f60b121571f2 qemud/Makefile.am --- a/qemud/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/qemud/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,5 +1,8 @@ ## Process this file with automake to produce Makefile.in +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + DAEMON_SOURCES = \ event.c event.h \ qemud.c qemud.h \ diff -r f60b121571f2 src/Makefile.am --- a/src/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/src/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,5 +1,8 @@ ## Process this file with automake to produce Makefile.in +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + INCLUDES = \ -I$(top_srcdir)/gnulib/lib -I../gnulib/lib \ -I../include \ diff -r f60b121571f2 tests/Makefile.am --- a/tests/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/tests/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,5 +1,8 @@ ## Process this file with automake to produce Makefile.in +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool + SHELL = $(PREFERABLY_POSIX_SHELL) SUBDIRS = confdata sexpr2xmldata \ diff -r f60b121571f2 tools/Makefile.am --- a/tools/Makefile.am Mon Apr 27 12:12:23 2009 +0100 +++ b/tools/Makefile.am Tue Apr 28 12:26:44 2009 +0100 @@ -1,3 +1,6 @@ + +AM_LIBTOOLFLAGS = --silent +LIBTOOL = $(top_srcdir)/mylibtool bin_SCRIPTS = \ virt-xml-validate -- |: 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 :|

1 0

[libvirt] qemu+tls server certificate validation failure (The certificate is not trusted)
by Scott Beardsley 28 Apr '09

28 Apr '09

I'm having a problem with remote TLS libvirt connections from an Ubuntu Jaunty client. I've reported the bug here[1] but haven't had any hits yet so I thought I'd come to the source. Let me know if ya'll have any ideas or know of any bugs in the versions I'm using (see below). I just upgraded my client to Jaunty from Intrepid and I can no longer connect to Hardy or Intrepid libvirt servers that have TLS enabled. I get the following errors: $ virt-viewer -c qemu+tls://example.com/system virt.example.com libvir: Remote error : server certificate failed validation: The certificate is not trusted. libvir: Remote error : unable to connect to 'example.com': Invalid argument unable to connect to libvirt qemu+tls://example.com/system $ In the past (ie hardy, intrepid) I was able to use the following command. Now I get an error: $ virt-viewer -c qemu://example.com/system virt.example.com libvir: error : could not connect to qemu://example.com/system unable to connect to libvirt qemu://example.com/system $ The server's config has not changed (I've tested against libvirt-bin versions 0.4.4-3ubuntu3.1 and 0.4.0-2ubuntu8.1 on the server side). I have the CA certificate installed on both server and client (in /etc/pki/CA/cacert.pem). That cert signed both my x509 client cert and the server cert. Here is some proof that it *should* work: $ openssl s_client -CAfile /etc/pki/CA/cacert.pem -cert /etc/pki/libvirt/clientcert.pem -key /etc/pki/libvirt/private/clientkey.pem -connect example.com:16514 2>/dev/null|sed -ne '/-BEGIN CERTIFICATE-/,/-END CERTIFICATE-/p' >servercert.pem $ openssl verify -CAfile /etc/pki/CA/cacert.pem servercert.pem servercert.pem: OK $ openssl verify -CAfile /etc/pki/CA/cacert.pem /etc/pki/libvirt/clientcert.pem /etc/pki/libvirt/clientcert.pem: OK $ When I run strace against virt-viewer I can see that it is accessing and (successfully opening) the correct certs/keys: $ grep /etc/pki /tmp/out stat64("/etc/pki/CA/cacert.pem", {st_mode=S_IFREG|0644, st_size=1716, ...}) = 0 stat64("/etc/pki/libvirt/private/clientkey.pem", {st_mode=S_IFREG|0644, st_size=887, ...}) = 0 stat64("/etc/pki/libvirt/clientcert.pem", {st_mode=S_IFREG|0644, st_size=1172, ...}) = 0 open("/etc/pki/CA/cacert.pem", O_RDONLY) = 5 open("/etc/pki/libvirt/private/clientkey.pem", O_RDONLY) = 5 open("/etc/pki/libvirt/clientcert.pem", O_RDONLY) = 5 $ Thanks in advance, Scott ------------ [1] https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/366455

2 3

[libvirt] PATCH: Fix compiler flag check for -Wformat-security
by Daniel P. Berrange 27 Apr '09

27 Apr '09

Certain versions of GCC will ignore the -Wformat-security flag unless you have also set the -Wformat flag. Unfortunately when checking flags, we only check one at a time in isolation, so it was thinking -Wformat-security is not supported in Fedora. The fix is to include all previously verified flags during the check process, but not for -Werror because for added fun, the autoconf test program itself causes compile warnings with some flags we have. The patch also fixes 3 harmless format problems in virsh Daniel Index: m4/compiler-flags.m4 =================================================================== RCS file: /data/cvs/libvirt/m4/compiler-flags.m4,v retrieving revision 1.2 diff -u -p -r1.2 compiler-flags.m4 --- m4/compiler-flags.m4 14 Jan 2009 15:23:26 -0000 1.2 +++ m4/compiler-flags.m4 27 Apr 2009 10:34:43 -0000 @@ -24,7 +24,15 @@ AC_DEFUN([gl_COMPILER_FLAGS], [AC_MSG_CHECKING(whether compiler accepts $1) AC_SUBST(COMPILER_FLAGS) ac_save_CFLAGS="$CFLAGS" - CFLAGS="$CFLAGS $1" + dnl Some flags are dependant, so we set all previously checked + dnl flags when testing. Except for -Werror which we have to + dnl check on its own, because some of our compiler flags cause + dnl warnings from the autoconf test program! + if test "$1" = "-Werror" ; then + CFLAGS="$CFLAGS $1" + else + CFLAGS="$CFLAGS $COMPILER_FLAGS $1" + fi AC_TRY_LINK([], [], has_option=yes, has_option=no,) echo 'int x;' >conftest.c $CC $CFLAGS -c conftest.c 2>conftest.err Index: src/virsh.c =================================================================== RCS file: /data/cvs/libvirt/src/virsh.c,v retrieving revision 1.201 diff -u -p -r1.201 virsh.c --- src/virsh.c 15 Apr 2009 20:09:09 -0000 1.201 +++ src/virsh.c 27 Apr 2009 10:34:45 -0000 @@ -4463,7 +4463,7 @@ cmdNodeListDevicesPrint(vshControl *ctl, } /* Print this device */ - vshPrint(ctl, indentBuf); + vshPrint(ctl, "%s", indentBuf); vshPrint(ctl, "%s\n", devices[devid]); @@ -4487,7 +4487,7 @@ cmdNodeListDevicesPrint(vshControl *ctl, /* If there is a child device, then print another blank line */ if (nextlastdev != -1) { - vshPrint(ctl, indentBuf); + vshPrint(ctl, "%s", indentBuf); vshPrint(ctl, " |\n"); } @@ -4511,7 +4511,7 @@ cmdNodeListDevicesPrint(vshControl *ctl, /* If there was no child device, and we're the last in * a list of devices, then print another blank line */ if (nextlastdev == -1 && devid == lastdev) { - vshPrint(ctl, indentBuf); + vshPrint(ctl, "%s", indentBuf); vshPrint(ctl, "\n"); } } -- |: 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 :|

3 4

[libvirt] libvirt Driver support Matrix.
by Abhishek K Jha 27 Apr '09

27 Apr '09

Hi, I needed to know if libvirt supports migration via the QEMU/KVM driver. ( virDomainMigrate) Could someone point me to the latest driver support matrix for libvirt. It seems the one at http://libvirt.org/hvsupport.html was last updated on 2008-06-05. Appreciate your help! Thanks much, Abhishek Jha. IBM India Software Labs, Pune 91-20-66592139(Direct) 91-20-26982424 Extn: 2139

5 6