- Devel - libvirt List Archives

[libvirt] [PATCH v2 0/9] Additional host name check for network storage pools
by John Ferlan 24 Apr '15

24 Apr '15

v1 here: http://www.redhat.com/archives/libvir-list/2015-April/msg00873.html Essentially a rewrite of v1 changing to use virsocketaddr.* rather than virutil.* Also making check for hostname duplication prior to startPool rather than during XML processing John Ferlan (9): socketaddr: Add isNumeric flag to virSocketAddrParseInternal socketaddr: Introduce virSocketAddrParseName iscsi: Check for validity of pool source hostname netfs: Check for validity of pool source hostname gluster: Check for validity of pool source hostname sheepdog: Check for validity of pool source hostname virsockaddr: Split up functionality of virSocketAddrFormatFull virsocketaddr: Introduce virSocketAddrIsSameTCPHost storage: Check for duplicate resolved host name at pool create src/conf/storage_conf.c | 80 ++++++++++++++- src/conf/storage_conf.h | 6 +- src/libvirt_private.syms | 3 + src/storage/storage_backend_fs.c | 5 +- src/storage/storage_backend_gluster.c | 6 +- src/storage/storage_backend_iscsi.c | 6 +- src/storage/storage_backend_sheepdog.c | 38 ++++--- src/storage/storage_driver.c | 9 +- src/util/virsocketaddr.c | 180 ++++++++++++++++++++++++++++----- src/util/virsocketaddr.h | 9 +- 10 files changed, 295 insertions(+), 47 deletions(-) -- 2.1.0

2 12

[libvirt] [PATCH] Fix memory leak in virNetSocketNewConnectUNIX
by Jiri Denemark 24 Apr '15

24 Apr '15

==26726== by 0x673CD67: __vasprintf_chk (vasprintf_chk.c:80) ==26726== by 0x5673605: UnknownInlinedFun (stdio2.h:210) ==26726== by 0x5673605: virVasprintfInternal (virstring.c:476) ==26726== by 0x56736EE: virAsprintfInternal (virstring.c:497) ==26726== by 0x5680C37: virGetUserRuntimeDirectory (virutil.c:866) ==26726== by 0x5783A89: virNetSocketNewConnectUNIX (virnetsocket.c:572) ==26726== by 0x57751AF: virNetClientNewUNIX (virnetclient.c:344) ==26726== by 0x57689B3: doRemoteOpen (remote_driver.c:895) ==26726== by 0x5769F8E: remoteConnectOpen (remote_driver.c:1195) ==26726== by 0x57092DF: do_open (libvirt.c:1189) ==26726== by 0x570A7BF: virConnectOpenAuth (libvirt.c:1341) https://bugzilla.redhat.com/show_bug.cgi?id=1215042 Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/rpc/virnetsocket.c | 44 +++++++++++++++++++++----------------------- 1 file changed, 21 insertions(+), 23 deletions(-) diff --git a/src/rpc/virnetsocket.c b/src/rpc/virnetsocket.c index a59e3e1..7ae2796 100644 --- a/src/rpc/virnetsocket.c +++ b/src/rpc/virnetsocket.c @@ -546,6 +546,7 @@ int virNetSocketNewConnectUNIX(const char *path, virSocketAddr localAddr; virSocketAddr remoteAddr; char *rundir = NULL; + int ret = -1; memset(&localAddr, 0, sizeof(localAddr)); memset(&remoteAddr, 0, sizeof(remoteAddr)); @@ -559,50 +560,50 @@ int virNetSocketNewConnectUNIX(const char *path, virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Auto-spawn of daemon requested, " "but no binary specified")); - goto error; + goto cleanup; } if (!(binname = last_component(binary)) || binname[0] == '\0') { virReportError(VIR_ERR_INTERNAL_ERROR, _("Cannot determine basename for binary '%s'"), binary); - goto error; + goto cleanup; } if (!(rundir = virGetUserRuntimeDirectory())) - goto error; + goto cleanup; if (virFileMakePathWithMode(rundir, 0700) < 0) { virReportSystemError(errno, _("Cannot create user runtime directory '%s'"), rundir); - goto error; + goto cleanup; } if (virAsprintf(&lockpath, "%s/%s.lock", rundir, binname) < 0) - goto error; + goto cleanup; if ((lockfd = open(lockpath, O_RDWR | O_CREAT, 0600)) < 0 || virSetCloseExec(lockfd) < 0) { virReportSystemError(errno, _("Unable to create lock '%s'"), lockpath); - goto error; + goto cleanup; } if (virFileLock(lockfd, false, 0, 1, true) < 0) { virReportSystemError(errno, _("Unable to lock '%s'"), lockpath); - goto error; + goto cleanup; } } if ((fd = socket(PF_UNIX, SOCK_STREAM, 0)) < 0) { virReportSystemError(errno, "%s", _("Failed to create socket")); - goto error; + goto cleanup; } remoteAddr.data.un.sun_family = AF_UNIX; if (virStrcpyStatic(remoteAddr.data.un.sun_path, path) == NULL) { virReportSystemError(ENOMEM, _("Path %s too long for unix socket"), path); - goto error; + goto cleanup; } if (remoteAddr.data.un.sun_path[0] == '@') remoteAddr.data.un.sun_path[0] = '\0'; @@ -612,42 +613,39 @@ int virNetSocketNewConnectUNIX(const char *path, if (!(spawnDaemon && errno == ENOENT)) { virReportSystemError(errno, _("Failed to connect socket to '%s'"), path); - goto error; + goto cleanup; } if (virNetSocketForkDaemon(binary) < 0) - goto error; + goto cleanup; retries--; usleep(5000); } - if (lockfd != -1) { - unlink(lockpath); - VIR_FORCE_CLOSE(lockfd); - VIR_FREE(lockpath); - } - localAddr.len = sizeof(localAddr.data); if (getsockname(fd, &localAddr.data.sa, &localAddr.len) < 0) { virReportSystemError(errno, "%s", _("Unable to get local socket name")); - goto error; + goto cleanup; } if (!(*retsock = virNetSocketNew(&localAddr, &remoteAddr, true, fd, -1, 0))) - goto error; + goto cleanup; - return 0; + ret = 0; - error: + cleanup: if (lockfd != -1) { unlink(lockpath); VIR_FORCE_CLOSE(lockfd); } VIR_FREE(lockpath); VIR_FREE(rundir); - VIR_FORCE_CLOSE(fd); - return -1; + + if (ret == -1) + VIR_FORCE_CLOSE(fd); + + return ret; } #else int virNetSocketNewConnectUNIX(const char *path ATTRIBUTE_UNUSED, -- 2.3.5

3 3

[libvirt] [PATCH] build: add pragma directive to fix build on some gcc
by Pavel Hrdina 24 Apr '15

24 Apr '15

Commit 1268820a removed obsolete index() function and replaced it by strchr. Few versions of gcc has a bug and reports a warning about strchr: ../../src/util/virstring.c:1006: error: logical '&&' with non-zero constant will always evaluate as true [-Wlogical-op] Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- Pushed as build-breaker fix. src/util/virstring.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/src/util/virstring.c b/src/util/virstring.c index 7b0cad7..5794f96 100644 --- a/src/util/virstring.c +++ b/src/util/virstring.c @@ -986,6 +986,16 @@ virStringHasControlChars(const char *str) } +/* Work around spurious strchr() diagnostics given by -Wlogical-op + * for gcc < 4.6. Doing it via a local pragma keeps the damage + * smaller than disabling it on the package level. Unfortunately, the + * affected GCCs don't allow diagnostic push/pop which would have + * further reduced the impact. */ +#if BROKEN_GCC_WLOGICALOP +# pragma GCC diagnostic ignored "-Wlogical-op" +#endif + + /** * virStringStripControlChars: * @str: the string to strip -- 2.0.5

1 0

[libvirt] [PATCH 0/5] Make virDomainObjListFindByName run in O(1)
by Michal Privoznik 24 Apr '15

24 Apr '15

As discussed here: https://www.redhat.com/archives/libvir-list/2015-April/msg01135.html Michal Privoznik (5): virDomainObjListAddLocked: s/false/NULL/ for @oldDef virDomainObjListNew: Use virObjectFreeHashData Introduce virDomainObjEndAPI virDomainObjListFindByName: Return referenced object virDomainObjList: Introduce yet another hash table src/bhyve/bhyve_driver.c | 3 +- src/conf/domain_conf.c | 83 +++++++++++++++++++++++------------ src/conf/domain_conf.h | 2 + src/libvirt_private.syms | 1 + src/libxl/libxl_driver.c | 10 ++--- src/lxc/lxc_driver.c | 3 +- src/openvz/openvz_driver.c | 11 +++-- src/parallels/parallels_driver.c | 3 +- src/qemu/qemu_domain.c | 7 +-- src/qemu/qemu_driver.c | 14 ++---- src/test/test_driver.c | 93 +++++++++++++--------------------------- src/uml/uml_driver.c | 15 +++---- 12 files changed, 110 insertions(+), 135 deletions(-) -- 2.0.5

4 15

[libvirt] [PATCH 0/4] tests: Cleanups and improvements
by Cole Robinson 23 Apr '15

23 Apr '15

First 3 patches are uninteresting cleanups, removing a bunch of boiler plate code. The last patch adds a way to easily regenerate test output, like when adding a new test case (plop in the qemu XML, use the env var and the test suite will create foo.args for you), or when a code changes causes test suite output to expectedly change. Cole Robinson (4): tests: Add VIR_TEST_DEBUG and VIR_TEST_VERBOSE tests: Use *DefParseFile more tests: Add virtTestCompareToFile tests: Add VIR_TEST_REGENERATE_OUTPUT HACKING | 7 ++ docs/hacking.html.in | 12 +++ tests/bhyvexml2argvtest.c | 41 ++-------- tests/bhyvexml2xmltest.c | 18 +---- tests/cputest.c | 45 ++++------- tests/domaincapstest.c | 9 +-- tests/domainconftest.c | 7 +- tests/jsontest.c | 48 +++++------- tests/lxcxml2xmltest.c | 17 +---- tests/networkxml2conftest.c | 16 +--- tests/networkxml2firewalltest.c | 9 +-- tests/networkxml2xmltest.c | 15 +--- tests/networkxml2xmlupdatetest.c | 15 +--- tests/nodeinfotest.c | 20 +---- tests/nwfilterxml2firewalltest.c | 11 +-- tests/nwfilterxml2xmltest.c | 15 +--- tests/qemuagenttest.c | 9 +-- tests/qemuargv2xmltest.c | 15 ++-- tests/qemucaps2xmltest.c | 27 +------ tests/qemuhelptest.c | 6 +- tests/qemuhotplugtest.c | 19 ++--- tests/qemumonitorjsontest.c | 9 +-- tests/qemumonitortest.c | 24 +++--- tests/qemuxml2argvtest.c | 27 ++----- tests/qemuxmlnstest.c | 19 +---- tests/secretxml2xmltest.c | 15 +--- tests/securityselinuxlabeltest.c | 13 +--- tests/sexpr2xmltest.c | 9 +-- tests/statstest.c | 3 +- tests/storagebackendsheepdogtest.c | 20 +---- tests/storagepoolxml2xmltest.c | 15 +--- tests/storagevolxml2argvtest.c | 40 ++-------- tests/storagevolxml2xmltest.c | 21 +---- tests/sysinfotest.c | 10 +-- tests/testutils.c | 74 ++++++++++++++---- tests/testutils.h | 14 ++++ tests/testutilslxc.c | 2 +- tests/testutilsqemu.c | 2 +- tests/utiltest.c | 30 +++----- tests/virbuftest.c | 60 +++++++-------- tests/vircaps2xmltest.c | 10 +-- tests/vircgrouptest.c | 9 +-- tests/virhashtest.c | 152 ++++++++++++------------------------- tests/virpcitest.c | 4 +- tests/virportallocatortest.c | 35 +++------ tests/vmx2xmltest.c | 9 +-- tests/xencapstest.c | 9 +-- tests/xlconfigtest.c | 27 +------ tests/xmconfigtest.c | 27 +------ tests/xml2sexprtest.c | 18 +---- tests/xml2vmxtest.c | 19 +---- 51 files changed, 340 insertions(+), 767 deletions(-) -- 2.3.5

2 9

[libvirt] [PATCH] qemu: cgroup: Fix priorities when setting emulatorpin
by Peter Krempa 23 Apr '15

23 Apr '15

Use the custom emulator pin setting with the highest priority same as with vcpupin. --- src/qemu/qemu_cgroup.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_cgroup.c b/src/qemu/qemu_cgroup.c index e83342d..bf0621f 100644 --- a/src/qemu/qemu_cgroup.c +++ b/src/qemu/qemu_cgroup.c @@ -1110,10 +1110,10 @@ qemuSetupCgroupForEmulator(virDomainObjPtr vm) if (virCgroupMoveTask(priv->cgroup, cgroup_emulator) < 0) goto cleanup; - if (def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO) - cpumask = priv->autoCpuset; - else if (def->cputune.emulatorpin) + if (def->cputune.emulatorpin) cpumask = def->cputune.emulatorpin->cpumask; + else if (def->placement_mode == VIR_DOMAIN_CPU_PLACEMENT_MODE_AUTO) + cpumask = priv->autoCpuset; else if (def->cpumask) cpumask = def->cpumask; -- 2.3.5

2 1

[libvirt] [PATCH v4 0/9] Implement Add/Del IOThreads
by John Ferlan 23 Apr '15

23 Apr '15

v3 here: http://www.redhat.com/archives/libvir-list/2015-April/msg00621.html Changes since v3 Patch 1: * Add back in and use virDomainIOThreadIDDefFree * Change _virDomainIOThreadIDDef 'undefined' to 'autofill' * Adjust virDomainDefParseXML to only allocate what's defined allowing virDomainDefPostParseInternal to Add in IOThreads which weren't defined and set the 'autofill' boolean * Have virDomainIOThreadIDAdd return a virDomainIOThreadIDDefPtr Patch 2: No changes Patch 3: (NEW) * Discussed in prior review regarding making virDomainPinIsDuplicate a static function (needed to move it) Patch 4: (NEW) * Move the iothreadspin data (e.g., the cpumask) into _virDomainIOThreadIDDef * Kept the 'niothreadspin' and manipulated as necessary since there was code to write out <cputune> data that I didn't want to reinvent/rototill just to search through iothreadids for a cpumask * Adjusted virDomainIOThreadPinDefParseXML to handle storing the cpumask in the right iothreadsid[] entry. If not found (it may not be since the virDomainDefPostParseInternal hasn't run), then create an autofill version of an iothreadids entry. * Remove/adjust a lot of code that used to handle iothreadspin Patch 5: (NEW) * Slight adjustment for iothreadsched to allow for "any" id value. This code stores iothread id's as bitmap entries, so it didn't have the same issues as the iothreadspin code Patch 6-7: Unchanged Patch 8: * Adjusted the search for the new thread code to use existing alias * Use the virDomainIOThreadIDAdd returned pointer * Comment adjustments from code review * Removal of erroneously cut-n-pasted code in Delete path * Use the ->dst for the message * Changes based on having cpumask in the iothrid data Patch 9: No changes John Ferlan (9): conf: Add new domain XML element 'iothreadids' qemu: Use domain iothreadids to IOThread's 'thread_id' conf: Move virDomainPinIsDuplicate and make static Move iothreadspin information into iothreadids conf: Adjust the iothreadsched expectations Implement virDomainAddIOThread and virDomainDelIOThread remote: Add support for AddIOThread and DelIOThread qemu: Add support to Add/Delete IOThreads virsh: Add iothreadadd and iothreaddel commands docs/formatdomain.html.in | 46 +- docs/schemas/domaincommon.rng | 12 + include/libvirt/libvirt-domain.h | 6 + src/conf/domain_audit.c | 9 + src/conf/domain_audit.h | 6 + src/conf/domain_conf.c | 340 ++++++++++---- src/conf/domain_conf.h | 25 +- src/driver-hypervisor.h | 12 + src/libvirt-domain.c | 118 +++++ src/libvirt_private.syms | 6 +- src/libvirt_public.syms | 6 + src/qemu/qemu_cgroup.c | 31 +- src/qemu/qemu_command.c | 38 +- src/qemu/qemu_command.h | 3 + src/qemu/qemu_domain.c | 36 -- src/qemu/qemu_domain.h | 3 - src/qemu/qemu_driver.c | 511 ++++++++++++++++++--- src/qemu/qemu_process.c | 40 +- src/remote/remote_driver.c | 2 + src/remote/remote_protocol.x | 30 +- src/remote_protocol-structs | 12 + .../qemuxml2argv-cputune-iothreadsched-toomuch.xml | 1 + .../qemuxml2argv-iothreads-ids-partial.args | 10 + .../qemuxml2argv-iothreads-ids-partial.xml | 33 ++ .../qemuxml2argv-iothreads-ids.args | 8 + .../qemuxml2argv-iothreads-ids.xml | 33 ++ tests/qemuxml2argvtest.c | 2 + tests/qemuxml2xmltest.c | 2 + tools/virsh-domain.c | 164 +++++++ tools/virsh.pod | 31 ++ 30 files changed, 1320 insertions(+), 256 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-iothreads-ids-partial.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-iothreads-ids-partial.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-iothreads-ids.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-iothreads-ids.xml -- 2.1.0

2 26

[libvirt] [PATCH] build: avoid obsolete index()
by Eric Blake 23 Apr '15

23 Apr '15

Commit 2a530a3e5 is not portable to mingw, which intentionally avoids declaring the obsolete index(). See also: https://bugzilla.redhat.com/show_bug.cgi?id=1214605 * src/util/virstring.c (virStringStripControlChars): Use strchr. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the build-breaker rule. src/util/virstring.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/virstring.c b/src/util/virstring.c index 1cd4987..7b0cad7 100644 --- a/src/util/virstring.c +++ b/src/util/virstring.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2012-2014 Red Hat, Inc. + * Copyright (C) 2012-2015 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 @@ -1003,7 +1003,7 @@ virStringStripControlChars(char *str) len = strlen(str); for (i = 0, j = 0; i < len; i++) { - if (index(control_chars, str[i])) + if (strchr(control_chars, str[i])) continue; str[j++] = str[i]; -- 2.1.0

1 0

[libvirt] [PATCH] parallels: implement .domainGetMaxMemory
by Dmitry Guryanov 23 Apr '15

23 Apr '15

Since we haven't implemented balloon parameters tuning we can just return amount of memory in this function. Signed-off-by: Dmitry Guryanov <dguryanov(a)parallels.com> --- src/parallels/parallels_driver.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/src/parallels/parallels_driver.c b/src/parallels/parallels_driver.c index 391e927..d13a4e2 100644 --- a/src/parallels/parallels_driver.c +++ b/src/parallels/parallels_driver.c @@ -1089,6 +1089,24 @@ parallelsDomainManagedSaveRemove(virDomainPtr domain, unsigned int flags) return ret; } +static unsigned long long +parallelsDomainGetMaxMemory(virDomainPtr domain) +{ + parallelsConnPtr privconn = domain->conn->privateData; + virDomainObjPtr dom = NULL; + int ret = -1; + + dom = virDomainObjListFindByUUID(privconn->domains, domain->uuid); + if (dom == NULL) { + parallelsDomNotFoundError(domain); + return -1; + } + + ret = dom->def->mem.max_balloon; + virObjectUnlock(dom); + return ret; +} + static virHypervisorDriver parallelsDriver = { .name = "Parallels", .connectOpen = parallelsConnectOpen, /* 0.10.0 */ @@ -1133,6 +1151,7 @@ static virHypervisorDriver parallelsDriver = { .domainHasManagedSaveImage = parallelsDomainHasManagedSaveImage, /* 1.2.13 */ .domainManagedSave = parallelsDomainManagedSave, /* 1.2.14 */ .domainManagedSaveRemove = parallelsDomainManagedSaveRemove, /* 1.2.14 */ + .domainGetMaxMemory = parallelsDomainGetMaxMemory, /* 1.2.14 */ }; static virConnectDriver parallelsConnectDriver = { -- 2.1.0

2 3

[libvirt] [PATCH 0/2] parallels: some fixes
by Maxim Nestratov 23 Apr '15

23 Apr '15

Maxim Nestratov (2): parallels: fix crash in prlsdkAddNet in case of CT definition parallels: fix IS_CT macro

2 5

[libvirt] [PATCH 0/4] Introduce parallelsDomObjFromDomain()
by Michal Privoznik 23 Apr '15

23 Apr '15

We have such wrapper in other drivers too: qemu, lxc, network, ... Michal Privoznik (4): struct _parallelsConn: Mark @domains as immutable pointer parallels: Introduce parallelsDomObjFromDomain() parallels_driver: Utilize parallelsDomObjFromDomain() parallels_sdk: Utilize parallelsDomObjFromDomain() src/parallels/parallels_driver.c | 93 ++++++---------------------------------- src/parallels/parallels_sdk.c | 5 +-- src/parallels/parallels_utils.c | 32 ++++++++++++++ src/parallels/parallels_utils.h | 5 +++ 4 files changed, 51 insertions(+), 84 deletions(-) -- 2.0.5

3 7

[libvirt] [PATCH] qemu: lock: unlock vm during qemuBuildCommandLine
by zhang bo 23 Apr '15

23 Apr '15

The function qemuBuildCommandLine() may take a long time, for example if we configure tens of vifs for the guest, each may cost hundrands of milliseconds to create tap dev, senconds in total. Thus, unlock vm before calling it. Signed-off-by: Zhang Bo <oscar.zhangbo(a)huawei.com> Signed-off-by: Zhou Yimin <zhouyimin(a)huawei.com> --- src/qemu/qemu_process.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 753afe8..d1aaaec 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -4628,14 +4628,18 @@ int qemuProcessStart(virConnectPtr conn, } VIR_DEBUG("Building emulator command line"); + virObjectUnlock(vm); if (!(cmd = qemuBuildCommandLine(conn, driver, vm->def, priv->monConfig, priv->monJSON, priv->qemuCaps, migrateFrom, stdin_fd, snapshot, vmop, &buildCommandLineCallbacks, false, qemuCheckFips(), priv->autoNodeset, - &nnicindexes, &nicindexes))) + &nnicindexes, &nicindexes))) { + virObjectLock(vm); goto cleanup; + } + virObjectLock(vm); /* now that we know it is about to start call the hook if present */ if (virHookPresent(VIR_HOOK_DRIVER_QEMU)) { -- 1.7.12.4

4 11

[libvirt] [RFCv2 PoCv1 PATCH 00/15] Just Admin API
by Martin Kletzander 23 Apr '15

23 Apr '15

*** BLURB *** ... Just kidding :) Sooo... After very very VERY long time, here's a draft of an admin interface that is supposed to open up new possibilities to be done on a live daemon. The aim here is to create some first inches of that API in order to open up the possibility of new API function creation to more people. I already spent so much time explaining so much of that that I don't know what else to point at in here. Maybe the fact that last three patches are just an example on how this might work. Of course there won't be any functions like listClientIDs with the need of getting each client info by another API. There's going to be e.g. virAdmClientInfo and virAdmGetClients() will return the list of them, etc. Long story short, let's not repeat past mistakes ;) With all that said, feel free to hate, love, comment or just try out compiling this series and let me know how many things I've missed and screwed up (hopefully zero). Martin Kletzander (15): util: add virJSONValueCopy util: Add virabstracts file for keeping abstract classes datatypes: Use abstract connect class in virConnect Break virNetServer into virNetSubServers Teach gendispatch how to handle admin dispatching files Add admin protocol Build client headers for admin protocol Add admin error domain Add libvirt-admin library Add XML files with admin API specification Add configuration options for permissions on daemon's admin socket Add support for admin API in libvirt daemon rpc: Add virNetSubServerGetNClients admin: Add virAdmHello function Example virt-admin .gitignore | 5 + Makefile.am | 4 +- cfg.mk | 11 +- configure.ac | 19 +- daemon/Makefile.am | 33 +- daemon/admin_server.c | 119 +++++ daemon/admin_server.h | 36 ++ daemon/libvirtd-config.c | 5 +- daemon/libvirtd-config.h | 1 + daemon/libvirtd.aug | 1 + daemon/libvirtd.c | 138 ++++-- daemon/libvirtd.conf | 8 + daemon/libvirtd.h | 14 +- daemon/remote.c | 194 ++++----- daemon/test_libvirtd.aug.in | 1 + docs/Makefile.am | 23 +- docs/apibuild.py | 10 +- include/libvirt/Makefile.am | 6 +- include/libvirt/libvirt-admin.h | 63 +++ include/libvirt/virterror.h | 3 +- libvirt-admin.pc.in | 13 + libvirt.spec.in | 10 + po/POTFILES.in | 3 + src/Makefile.am | 134 +++++- src/admin/admin_protocol.x | 73 ++++ src/admin_protocol-structs | 17 + src/datatypes.c | 39 +- src/datatypes.h | 45 +- src/internal.h | 1 + src/libvirt-admin.c | 363 ++++++++++++++++ src/libvirt-host.c | 3 +- src/libvirt_admin.syms | 19 + src/libvirt_private.syms | 6 + src/libvirt_remote.syms | 17 +- src/locking/lock_daemon.c | 40 +- src/locking/lock_daemon_dispatch.c | 18 +- src/lxc/lxc_controller.c | 18 +- src/rpc/gendispatch.pl | 141 +++--- src/rpc/virnetserver.c | 859 +++++++++++++------------------------ src/rpc/virnetserver.h | 61 +-- src/rpc/virnetserverprogram.c | 12 +- src/rpc/virnetserverprogram.h | 9 +- src/rpc/virnetsubserver.c | 672 +++++++++++++++++++++++++++++ src/rpc/virnetsubserver.h | 78 ++++ src/util/virabstracts.c | 100 +++++ src/util/virabstracts.h | 57 +++ src/util/virerror.c | 95 ++-- src/util/virerror.h | 4 +- src/util/virjson.c | 65 ++- src/util/virjson.h | 4 +- tests/confdata/libvirtd.conf | 6 + tests/confdata/libvirtd.out | 5 + tests/jsontest.c | 111 +++++ tools/virt-admin/Makefile.am | 70 +++ tools/virt-admin/virt-admin.c | 68 +++ tools/virt-admin/virt-admin.pod | 43 ++ 56 files changed, 3092 insertions(+), 881 deletions(-) create mode 100644 daemon/admin_server.c create mode 100644 daemon/admin_server.h create mode 100644 include/libvirt/libvirt-admin.h create mode 100644 libvirt-admin.pc.in create mode 100644 src/admin/admin_protocol.x create mode 100644 src/admin_protocol-structs create mode 100644 src/libvirt-admin.c create mode 100644 src/libvirt_admin.syms create mode 100644 src/rpc/virnetsubserver.c create mode 100644 src/rpc/virnetsubserver.h create mode 100644 src/util/virabstracts.c create mode 100644 src/util/virabstracts.h create mode 100644 tools/virt-admin/Makefile.am create mode 100644 tools/virt-admin/virt-admin.c create mode 100644 tools/virt-admin/virt-admin.pod -- 2.3.5

4 29

[libvirt] [PATCH] rng: fix port number range validation
by Daniel P. Berrange 23 Apr '15

23 Apr '15

The PortNumber data type is declared to derive from 'short'. Unfortunately this is an signed type, so validates the range [-32,768, 32,767] which excludes valid port numbers between 32767 and 65535. We can't use 'unsignedShort', since we need -1 to be a valid port number too. This change is to use 'int' and set an explicit max boundary instead of relying on the data types' built-in max. One of the existing tests is changed to use a high port number to validate the schema. https://bugzilla.redhat.com/show_bug.cgi?id=1214664 Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- docs/schemas/basictypes.rng | 3 ++- tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.xml | 2 +- 3 files changed, 4 insertions(+), 3 deletions(-) diff --git a/docs/schemas/basictypes.rng b/docs/schemas/basictypes.rng index f086ad2..a83063a 100644 --- a/docs/schemas/basictypes.rng +++ b/docs/schemas/basictypes.rng @@ -349,8 +349,9 @@ </define> <define name="PortNumber"> - <data type="short"> + <data type="int"> <param name="minInclusive">-1</param> + <param name="maxInclusive">65535</param> </data> </define> diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.args b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.args index 1b61fcf..34dc767 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.args @@ -1,4 +1,4 @@ LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \ /usr/bin/qemu -S -M pc -m 214 -smp 1 -monitor unix:/tmp/test-monitor,server,nowait \ -no-acpi -boot c -usb -hda /dev/HostVG/QEMUGuest1 \ --net none -serial none -parallel none -vnc 127.0.0.1:0,share=allow-exclusive +-net none -serial none -parallel none -vnc 127.0.0.1:59630,share=allow-exclusive diff --git a/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.xml b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.xml index f78ff48..4b88e49 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-policy.xml @@ -25,7 +25,7 @@ <controller type='pci' index='0' model='pci-root'/> <input type='mouse' bus='ps2'/> <input type='keyboard' bus='ps2'/> - <graphics type='vnc' port='5900' autoport='no' listen='127.0.0.1' sharePolicy='allow-exclusive'> + <graphics type='vnc' port='65530' autoport='no' listen='127.0.0.1' sharePolicy='allow-exclusive'> <listen type='address' address='127.0.0.1'/> </graphics> <video> -- 2.1.0

2 1

[libvirt] [PATCH 0/2] Apparmor PCI passthrough fixes
by Cédric Bosdonnat 23 Apr '15

23 Apr '15

Hi all, These two simple patches intend to fix apparmor profiles for PCI passthrough with the qemu driver. Cédric Bosdonnat (2): Allow access to vendor and device file for PCI device passthrough Apparmor: allow reading block-rbd.so examples/apparmor/libvirt-qemu | 1 + src/util/virpci.c | 4 +++- 2 files changed, 4 insertions(+), 1 deletion(-) -- 2.1.4

2 3

[libvirt] [PATCH] util: string: Avoid use of deprecated function index()
by Peter Krempa 23 Apr '15

23 Apr '15

--- src/util/virstring.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/src/util/virstring.c b/src/util/virstring.c index 1cd4987..e48c29a 100644 --- a/src/util/virstring.c +++ b/src/util/virstring.c @@ -976,6 +976,21 @@ static const char control_chars[] = "\x10\x11\x12\x13\x14\x15\x16\x17" "\x18\x19\x1A\x1B\x1C\x1D\x1E\x1F"; + +static bool +virStringIsControlChar(char c) +{ + const char *cch; + + for (cch = control_chars; *cch; cch++) { + if (*cch == c) + return true; + } + + return false; +} + + bool virStringHasControlChars(const char *str) { @@ -1003,7 +1018,7 @@ virStringStripControlChars(char *str) len = strlen(str); for (i = 0, j = 0; i < len; i++) { - if (index(control_chars, str[i])) + if (virStringIsControlChar(str[i])) continue; str[j++] = str[i]; -- 2.3.5

2 1

[libvirt] [PATCH] qemu: Don't fail to reboot domains with unresponsive agent
by zhang bo 23 Apr '15

23 Apr '15

just as what b8e25c35d7f80a2fadc0e51e95318e39db3d1687 did, we fall back to the ACPI method when the guest agent is unresponsive in qemuDomainReboot. Signed-off-by: YueWenyuan <yuewenyuan(a)huawei.com> Signed-off-by: Zhang Bo <oscar.zhangbo(a)huawei.com> --- src/qemu/qemu_driver.c | 67 +++++++++++++++++++++++++------------------------- 1 file changed, 34 insertions(+), 33 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 6fc9696..964a9c5 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2002,21 +2002,14 @@ qemuDomainReboot(virDomainPtr dom, unsigned int flags) virDomainObjPtr vm; int ret = -1; qemuDomainObjPrivatePtr priv; - bool useAgent = false; + bool useAgent = false, agentRequested, acpiRequested; bool isReboot = true; + bool agentForced; int agentFlag = QEMU_AGENT_SHUTDOWN_REBOOT; virCheckFlags(VIR_DOMAIN_REBOOT_ACPI_POWER_BTN | VIR_DOMAIN_REBOOT_GUEST_AGENT, -1); - /* At most one of these two flags should be set. */ - if ((flags & VIR_DOMAIN_REBOOT_ACPI_POWER_BTN) && - (flags & VIR_DOMAIN_REBOOT_GUEST_AGENT)) { - virReportInvalidArg(flags, "%s", - _("flags for acpi power button and guest agent are mutually exclusive")); - return -1; - } - if (!(vm = qemuDomObjFromDomain(dom))) goto cleanup; @@ -2028,38 +2021,25 @@ qemuDomainReboot(virDomainPtr dom, unsigned int flags) } priv = vm->privateData; + agentRequested = flags & VIR_DOMAIN_REBOOT_GUEST_AGENT; + acpiRequested = flags & VIR_DOMAIN_REBOOT_ACPI_POWER_BTN; if (virDomainRebootEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; - if ((flags & VIR_DOMAIN_REBOOT_GUEST_AGENT) || - (!(flags & VIR_DOMAIN_REBOOT_ACPI_POWER_BTN) && - priv->agent)) + /* Prefer agent unless we were requested to not to. */ + if (agentRequested || (!flags && priv->agent)) useAgent = true; - if (!useAgent) { -#if WITH_YAJL - if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_MONITOR_JSON)) { - if (!virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_NO_SHUTDOWN)) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("Reboot is not supported with this QEMU binary")); - goto cleanup; - } - } else { -#endif - virReportError(VIR_ERR_OPERATION_INVALID, "%s", - _("Reboot is not supported without the JSON monitor")); - goto cleanup; -#if WITH_YAJL - } -#endif - } - if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; - if (useAgent && !qemuDomainAgentAvailable(vm, true)) - goto endjob; + agentForced = agentRequested && !acpiRequested; + if (useAgent && !qemuDomainAgentAvailable(vm, true)) { + if (agentForced) + goto endjob; + useAgent = false; + } if (!virDomainObjIsActive(vm)) { virReportError(VIR_ERR_OPERATION_INVALID, @@ -2071,7 +2051,28 @@ qemuDomainReboot(virDomainPtr dom, unsigned int flags) qemuDomainObjEnterAgent(vm); ret = qemuAgentShutdown(priv->agent, agentFlag); qemuDomainObjExitAgent(vm); - } else { + } + + /* If we are not enforced to use just an agent, try ACPI + * shutdown as well in case agent did not succeed. + */ + if ((!useAgent) || + (ret < 0 && (acpiRequested || !flags))) { +#if WITH_YAJL + if (virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_MONITOR_JSON)) { + if (!virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_NO_SHUTDOWN)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("ACPI reboot is not supported with this QEMU binary")); + goto endjob; + } + } else { +#endif + virReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("ACPI reboot is not supported without the JSON monitor")); + goto endjob; +#if WITH_YAJL + } +#endif qemuDomainObjEnterMonitor(driver, vm); ret = qemuMonitorSystemPowerdown(priv->mon); if (qemuDomainObjExitMonitor(driver, vm) < 0) -- 1.7.12.4

2 1

[libvirt] [PATCH] bhyve: fix build in tests
by Roman Bogorodskiy 23 Apr '15

23 Apr '15

Commit 835cf84 dropped expectedVirtTypes argument for virDomainDefParse*() functions, however bhyve tests still try to pass that to virDomainDefParseFile(), therefore build fails. Fix build by fixing virDomainDefParseFile() usage. --- tests/bhyvexml2argvtest.c | 1 - tests/bhyvexml2xmltest.c | 1 - 2 files changed, 2 deletions(-) diff --git a/tests/bhyvexml2argvtest.c b/tests/bhyvexml2argvtest.c index c5b0681..26d0075 100644 --- a/tests/bhyvexml2argvtest.c +++ b/tests/bhyvexml2argvtest.c @@ -32,7 +32,6 @@ static int testCompareXMLToArgvFiles(const char *xml, goto out; if (!(vmdef = virDomainDefParseFile(xml, driver.caps, driver.xmlopt, - 1 << VIR_DOMAIN_VIRT_BHYVE, VIR_DOMAIN_DEF_PARSE_INACTIVE))) goto out; diff --git a/tests/bhyvexml2xmltest.c b/tests/bhyvexml2xmltest.c index a1a3701..2e742cf 100644 --- a/tests/bhyvexml2xmltest.c +++ b/tests/bhyvexml2xmltest.c @@ -27,7 +27,6 @@ testCompareXMLToXMLFiles(const char *inxml, const char *outxml) goto fail; if (!(def = virDomainDefParseString(inXmlData, driver.caps, driver.xmlopt, - 1 << VIR_DOMAIN_VIRT_BHYVE, VIR_DOMAIN_DEF_PARSE_INACTIVE))) goto fail; -- 2.3.5

2 2

[libvirt] [PATCH] configure: Report --with-loader-nvram value in summary
by Cole Robinson 22 Apr '15

22 Apr '15

--- configure.ac | 1 + 1 file changed, 1 insertion(+) diff --git a/configure.ac b/configure.ac index 7f6fd7c..d5fb71d 100644 --- a/configure.ac +++ b/configure.ac @@ -2980,6 +2980,7 @@ AC_MSG_NOTICE([ XML Catalog: $XML_CATALOG_FILE]) AC_MSG_NOTICE([ Init script: $with_init_script]) AC_MSG_NOTICE([Char device locks: $with_chrdev_lock_files]) AC_MSG_NOTICE([ Default Editor: $DEFAULT_EDITOR]) +AC_MSG_NOTICE([ Loader/NVRAM: $with_loader_nvram]) AC_MSG_NOTICE([]) AC_MSG_NOTICE([Developer Tools]) AC_MSG_NOTICE([]) -- 2.3.5

2 4

[libvirt] [PATCH V2 0/2] Support libvirt's lock manager in the libxl driver
by Jim Fehlig 22 Apr '15

22 Apr '15

V2 of https://www.redhat.com/archives/libvir-list/2015-April/msg00845.html Patch 2 has been ACKed and pushed since it is independent of the lock manager integration. Patch 1 has been changed to address Daniel's comments. Patch 2 (was patch 3 in V1) is unchanged for the time being. Jim Fehlig (2): locking: relax PID requirement libxl: provide integration with lock manager src/Makefile.am | 12 +++++++++ src/libxl/libvirtd_libxl.aug | 2 ++ src/libxl/libxl.conf | 10 ++++++++ src/libxl/libxl_conf.c | 14 +++++++++++ src/libxl/libxl_conf.h | 6 +++++ src/libxl/libxl_domain.c | 47 ++++++++++++++++++++++++++++++++++-- src/libxl/libxl_domain.h | 1 + src/libxl/libxl_driver.c | 25 +++++++++++++++++++ src/libxl/libxl_migration.c | 6 +++++ src/libxl/test_libvirtd_libxl.aug.in | 1 + src/locking/lock_daemon.c | 2 +- src/locking/lock_daemon_dispatch.c | 16 ++++++------ src/locking/lock_driver_lockd.c | 7 ++---- 13 files changed, 133 insertions(+), 16 deletions(-) -- 1.8.4.5

2 4

[libvirt] [PATCH] spec: Fix libxl build with driver modules
by Cole Robinson 22 Apr '15

22 Apr '15

--- Pushed as build breaker libvirt.spec.in | 3 +++ 1 file changed, 3 insertions(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index 5d8a9d4..c5c6608 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -2127,6 +2127,9 @@ exit 0 %if %{with_libxl} %files daemon-driver-libxl %defattr(-, root, root) +%config(noreplace) %{_sysconfdir}/libvirt/libxl.conf +%{_datadir}/augeas/lenses/libvirtd_libxl.aug +%{_datadir}/augeas/lenses/tests/test_libvirtd_libxl.aug %dir %attr(0700, root, root) %{_localstatedir}/log/libvirt/libxl/ %ghost %dir %{_localstatedir}/run/libvirt/libxl/ %dir %attr(0700, root, root) %{_localstatedir}/lib/libvirt/libxl/ -- 2.3.5

1 0

[libvirt] [PATCH] configure: Fix --loader-nvram typo
by Cole Robinson 22 Apr '15

22 Apr '15

--- Pushed as trivial configure.ac | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/configure.ac b/configure.ac index aed0934..7f6fd7c 100644 --- a/configure.ac +++ b/configure.ac @@ -2776,7 +2776,7 @@ AC_ARG_WITH([loader-nvram], fi AC_DEFINE_UNQUOTED([DEFAULT_LOADER_NVRAM], ["$withval"], - [List of laoder:nvram pairs])]) + [List of loader:nvram pairs])]) # Some GNULIB base64 symbols clash with a kerberos library AC_DEFINE_UNQUOTED([isbase64],[libvirt_gl_isbase64],[Hack to avoid symbol clash]) -- 2.3.5

1 0

[libvirt] [PATCH] parallels: don't forget to unlock domain after successful virDomainObjListFindByUUID call
by Maxim Nestratov 22 Apr '15

22 Apr '15

Also a typo is fixed (s/detached/attached/) Signed-off-by: Maxim Nestratov mnestratov(a)parallels.com --- src/parallels/parallels_driver.c | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/src/parallels/parallels_driver.c b/src/parallels/parallels_driver.c index 46e6c3e..41908a5 100644 --- a/src/parallels/parallels_driver.c +++ b/src/parallels/parallels_driver.c @@ -1081,7 +1081,7 @@ static int parallelsDomainAttachDeviceFlags(virDomainPtr dom, const char *xml, privdom = virDomainObjListFindByUUID(privconn->domains, dom->uuid); if (privdom == NULL) { parallelsDomNotFoundError(dom); - goto cleanup; + return -1; } if (!(flags & VIR_DOMAIN_AFFECT_CONFIG)) { @@ -1120,13 +1120,14 @@ static int parallelsDomainAttachDeviceFlags(virDomainPtr dom, const char *xml, break; default: virReportError(VIR_ERR_OPERATION_UNSUPPORTED, - _("device type '%s' cannot be detached"), + _("device type '%s' cannot be attached"), virDomainDeviceTypeToString(dev->type)); break; } ret = 0; cleanup: + virObjectUnlock(privdom); return ret; } -- 1.7.1

2 2

[libvirt] [PATCH 0/3] Support libvirt's lock manager in the libxl driver
by Jim Fehlig 22 Apr '15

22 Apr '15

Patch 1 relaxes the requirement of PID != 0 in the locking code. danpb and I discussed this on IRC: danpb: jimfehlig: it wouldn't be the end of the world if you leave pid as 0 danpb: jimfehlig: it'd just loose some safety sanity check in that scenario Patch 2 introduces a config file for the libxl driver, with the lone config entry 'autoballoon'. I thought this would make review easier, vs adding all the config file support in patch 3, which is fairly large as is. Jim Fehlig (3): locking: relax PID requirement libxl: Introduce configuration file for libxl driver libxl: provide integration with lock manager libvirt.spec.in | 8 ++++ src/Makefile.am | 36 ++++++++++++++++- src/libxl/libvirtd_libxl.aug | 44 +++++++++++++++++++++ src/libxl/libxl.conf | 22 +++++++++++ src/libxl/libxl_conf.c | 75 +++++++++++++++++++++++++++++++++--- src/libxl/libxl_conf.h | 11 ++++++ src/libxl/libxl_domain.c | 47 +++++++++++++++++++++- src/libxl/libxl_domain.h | 1 + src/libxl/libxl_driver.c | 34 ++++++++++++++++ src/libxl/libxl_migration.c | 6 +++ src/libxl/test_libvirtd_libxl.aug.in | 6 +++ src/locking/lock_daemon.c | 2 +- src/locking/lock_daemon_dispatch.c | 49 +++++++---------------- src/locking/lock_driver_lockd.c | 7 +--- 14 files changed, 298 insertions(+), 50 deletions(-) create mode 100644 src/libxl/libvirtd_libxl.aug create mode 100644 src/libxl/libxl.conf create mode 100644 src/libxl/test_libvirtd_libxl.aug.in -- 1.8.4.5

2 10

[libvirt] [PATCH 0/2] Add tox.ini to run tests with tox
by Victor Stinner 22 Apr '15

22 Apr '15

Tox is a convinient tool creating virtual environment to install dependencies and run tests. To run all tests, just type "tox". To only test a specific Python version, use "tox -e py34" (Python 3.4). Victor Stinner (2): sanitytest.py parameters are now optional Add tox.ini to run tests on Python 2.6, 2.7 & 3.4 .gitignore | 1 + sanitytest.py | 23 ++++++++++++++++++----- tox.ini | 10 ++++++++++ 3 files changed, 29 insertions(+), 5 deletions(-) create mode 100644 tox.ini -- 2.1.0

4 8