- Devel - libvirt List Archives

[libvirt] [PATCH v2] qemu: Fix using guest architecture as lookup key
by Andrea Bolognani 16 Sep '15

16 Sep '15

When looking for a QEMU binary suitable for running ppc64le guests we have to take into account the fact that we use the QEMU target as key for the hash, so direct comparison is not good enough. Factor out the logic from virQEMUCapsFindBinaryForArch() to a new virQEMUCapsFindTarget() function and use that both when looking for QEMU binaries available on the system and when looking up QEMU capabilities later. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1260753 --- src/qemu/qemu_capabilities.c | 100 ++++++++++++++++++++++++++++--------------- 1 file changed, 65 insertions(+), 35 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 4ad1bdb..6757907 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -386,6 +386,28 @@ static const char *virQEMUCapsArchToString(virArch arch) return virArchToString(arch); } +/* Given a host and guest architectures, find a suitable QEMU target. + * + * This is meant to be used as a second attempt if qemu-system-$guestarch + * can't be found, eg. on a x86_64 host you want to use qemu-system-i386, + * if available, instead of qemu-system-x86_64 to run i686 guests */ +static virArch +virQEMUCapsFindTarget(virArch hostarch, + virArch guestarch) +{ + /* Both ppc64 and ppc64le guests can use the ppc64 target */ + if (ARCH_IS_PPC64(guestarch)) + guestarch = VIR_ARCH_PPC64; + + /* armv7l guests on aarch64 hosts can use the aarch64 target + * i686 guests on x86_64 hosts can use the x86_64 target */ + if ((guestarch == VIR_ARCH_ARMV7L && hostarch == VIR_ARCH_AARCH64) || + (guestarch == VIR_ARCH_I686 && hostarch == VIR_ARCH_X86_64)) { + return hostarch; + } + + return guestarch; +} static virCommandPtr virQEMUCapsProbeCommand(const char *qemu, @@ -690,51 +712,52 @@ virQEMUCapsProbeCPUModels(virQEMUCapsPtr qemuCaps, uid_t runUid, gid_t runGid) return ret; } - static char * -virQEMUCapsFindBinaryForArch(virArch hostarch, - virArch guestarch) +virQEMUCapsFindBinary(const char *format, + const char *archstr) { - char *ret; - const char *archstr; - char *binary; - - if (ARCH_IS_PPC64(guestarch)) - archstr = virQEMUCapsArchToString(VIR_ARCH_PPC64); - else - archstr = virQEMUCapsArchToString(guestarch); + char *ret = NULL; + char *binary = NULL; - if (virAsprintf(&binary, "qemu-system-%s", archstr) < 0) - return NULL; + if (virAsprintf(&binary, format, archstr) < 0) + goto out; ret = virFindFileInPath(binary); VIR_FREE(binary); - if (ret && !virFileIsExecutable(ret)) - VIR_FREE(ret); + if (ret && virFileIsExecutable(ret)) + goto out; - if (guestarch == VIR_ARCH_ARMV7L && - !ret && - hostarch == VIR_ARCH_AARCH64) { - ret = virFindFileInPath("qemu-system-aarch64"); - if (ret && !virFileIsExecutable(ret)) - VIR_FREE(ret); - } + VIR_FREE(ret); - if (guestarch == VIR_ARCH_I686 && - !ret && - hostarch == VIR_ARCH_X86_64) { - ret = virFindFileInPath("qemu-system-x86_64"); - if (ret && !virFileIsExecutable(ret)) - VIR_FREE(ret); - } + out: + return ret; +} + +static char * +virQEMUCapsFindBinaryForArch(virArch hostarch, + virArch guestarch) +{ + char *ret = NULL; + const char *archstr; + + /* First attempt: try the guest architecture as it is */ + archstr = virQEMUCapsArchToString(guestarch); + if ((ret = virQEMUCapsFindBinary("qemu-system-%s", archstr)) != NULL) + goto out; - if (guestarch == VIR_ARCH_I686 && - !ret) { - ret = virFindFileInPath("qemu"); - if (ret && !virFileIsExecutable(ret)) - VIR_FREE(ret); + /* Second attempt: use some arch-specific rules */ + archstr = virQEMUCapsArchToString(virQEMUCapsFindTarget(hostarch, + guestarch)); + if ((ret = virQEMUCapsFindBinary("qemu-system-%s", archstr)) != NULL) + goto out; + + /* Third attempt, i686 only: try 'qemu' */ + if (guestarch == VIR_ARCH_I686) { + if ((ret = virQEMUCapsFindBinary("%s", "qemu")) != NULL) + goto out; } + out: return ret; } @@ -3793,10 +3816,17 @@ virQEMUCapsCacheLookupByArch(virQEMUCapsCachePtr cache, virMutexLock(&cache->lock); ret = virHashSearch(cache->binaries, virQEMUCapsCompareArch, &data); - VIR_DEBUG("Returning caps %p for arch %s", ret, virArchToString(arch)); + if (!ret) { + /* If the first attempt at findind capabilities has failed, try + * again using the QEMU target as lookup key instead */ + data.arch = virQEMUCapsFindTarget(virArchFromHost(), data.arch); + ret = virHashSearch(cache->binaries, virQEMUCapsCompareArch, &data); + } virObjectRef(ret); virMutexUnlock(&cache->lock); + VIR_DEBUG("Returning caps %p for arch %s", ret, virArchToString(arch)); + return ret; } -- 2.4.3

2 2

[libvirt] [PATCH] libxl: fix compiler error introduced by commit ba25c214
by Jim Fehlig 16 Sep '15

16 Sep '15

libxl/libxl_conf.c: In function 'libxlDriverConfigNew': libxl/libxl_conf.c:1560:30: error: 'log_level' may be used uninitialized in this function [-Werror=maybe-uninitialized] Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- Pushing under the build-breaker rule. src/libxl/libxl_conf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/libxl/libxl_conf.c b/src/libxl/libxl_conf.c index 40fa4b5..35fd495 100644 --- a/src/libxl/libxl_conf.c +++ b/src/libxl/libxl_conf.c @@ -1496,7 +1496,7 @@ libxlDriverConfigNew(void) { libxlDriverConfigPtr cfg; char *log_file = NULL; - xentoollog_level log_level; + xentoollog_level log_level = XTL_DEBUG; char ebuf[1024]; unsigned int free_mem; -- 2.5.0

1 0

[libvirt] [PATCH] qemu: The ppc64 target can run both ppc64 and ppc64le guests
by Andrea Bolognani 15 Sep '15

15 Sep '15

When looking for a QEMU binary suitable for running ppc64le guests we have to take into account the fact that we use the QEMU target as key for the hash, so direct comparison is not good enough: normalize everything that can run on the ppc64 target to VIR_ARCH_PPC64 instead. This approach was already used in virQEMUCapsFindBinaryForArch(), and I've added a comment there to clarify why it's needed. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1260753 --- src/qemu/qemu_capabilities.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 4ad1bdb..a06141e 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -699,6 +699,7 @@ virQEMUCapsFindBinaryForArch(virArch hostarch, const char *archstr; char *binary; + /* qemu-system-ppc64 can run both ppc64 and ppc64le guests */ if (ARCH_IS_PPC64(guestarch)) archstr = virQEMUCapsArchToString(VIR_ARCH_PPC64); else @@ -3791,6 +3792,10 @@ virQEMUCapsCacheLookupByArch(virQEMUCapsCachePtr cache, virQEMUCapsPtr ret = NULL; struct virQEMUCapsSearchData data = { .arch = arch }; + /* QEMU's ppc64 target can run both ppc64 and ppc64le guests */ + if (ARCH_IS_PPC64(data.arch)) + data.arch = VIR_ARCH_PPC64; + virMutexLock(&cache->lock); ret = virHashSearch(cache->binaries, virQEMUCapsCompareArch, &data); VIR_DEBUG("Returning caps %p for arch %s", ret, virArchToString(arch)); -- 2.4.3

2 2

[libvirt] [PATCH v3] Ignore virtio-mmio disks in qemuAssignDevicePCISlots()
by Pavel Fedin 15 Sep '15

15 Sep '15

Fixes the following error when attempting to add a disk with bus='virtio' to a machine which actually supports virtio-mmio (caught with ARM virt): virtio disk cannot have an address of type 'virtio-mmio' The problem has been likely introduced by e8d55172544c1fafe31a9e09346bdebca4f0d6f9. Before that qemuAssignDevicePCISlots() was never called for ARM "virt" machine. Signed-off-by: Pavel Fedin <p.fedin(a)samsung.com> --- v2 => v3 - Bring back qemuCaps to qemuAssignDevicePCISlots(), was lost in a3ecd63e928ff39d73c1c14b0fb3be8addbc977b - Swap conditions so as not to call virQEMUCapsGet() every time v1 => v2 - Added check for QEMU_CAPS_DEVICE_VIRTIO_MMIO, this leaves the error message for machines which actually do not support virtio-mmio. In this case the user may still attempt to add such a disk manually. --- src/qemu/qemu_command.c | 11 +++++++++-- src/qemu/qemu_command.h | 1 + 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ec5e3d4..ea1bb28 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -2241,7 +2241,7 @@ qemuDomainAssignPCIAddresses(virDomainDefPtr def, virDomainPCIAddressReserveNextSlot(addrs, &info, flags) < 0) goto cleanup; - if (qemuAssignDevicePCISlots(def, addrs) < 0) + if (qemuAssignDevicePCISlots(def, qemuCaps, addrs) < 0) goto cleanup; for (i = 1; i < addrs->nbuses; i++) { @@ -2274,7 +2274,7 @@ qemuDomainAssignPCIAddresses(virDomainDefPtr def, if (qemuValidateDevicePCISlotsChipsets(def, qemuCaps, addrs) < 0) goto cleanup; - if (qemuAssignDevicePCISlots(def, addrs) < 0) + if (qemuAssignDevicePCISlots(def, qemuCaps, addrs) < 0) goto cleanup; for (i = 0; i < def->ncontrollers; i++) { @@ -2406,6 +2406,7 @@ qemuDomainAssignPCIAddresses(virDomainDefPtr def, */ int qemuAssignDevicePCISlots(virDomainDefPtr def, + virQEMUCapsPtr qemuCaps, virDomainPCIAddressSetPtr addrs) { size_t i, j; @@ -2598,6 +2599,12 @@ qemuAssignDevicePCISlots(virDomainDefPtr def, VIR_DOMAIN_DEVICE_ADDRESS_TYPE_CCW) continue; + /* Also ignore virtio-mmio disks if our machine allows them */ + if (def->disks[i]->info.type == + VIR_DOMAIN_DEVICE_ADDRESS_TYPE_VIRTIO_MMIO && + virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE_VIRTIO_MMIO)) + continue; + if (def->disks[i]->info.type != VIR_DOMAIN_DEVICE_ADDRESS_TYPE_NONE) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("virtio disk cannot have an address of type '%s'"), diff --git a/src/qemu/qemu_command.h b/src/qemu/qemu_command.h index 767d31f..4aa7f2d 100644 --- a/src/qemu/qemu_command.h +++ b/src/qemu/qemu_command.h @@ -284,6 +284,7 @@ virDomainPCIAddressSetPtr qemuDomainPCIAddressSetCreate(virDomainDefPtr def, bool dryRun); int qemuAssignDevicePCISlots(virDomainDefPtr def, + virQEMUCapsPtr qemuCaps, virDomainPCIAddressSetPtr addrs); int qemuAssignDeviceAliases(virDomainDefPtr def, virQEMUCapsPtr qemuCaps); -- 1.9.5.msysgit.0

2 1

[libvirt] [PATCH] Minor typo fixes in documentation
by Christian Loehle 15 Sep '15

15 Sep '15

Signed-off-by: Christian Loehle <cloehle(a)linutronix.de>

2 1

[libvirt] [PATCH 0/3] Implement mockup capabilities cache in QEMU tests
by Pavel Fedin 15 Sep '15

15 Sep '15

Since commit e8d55172544c1fafe31a9e09346bdebca4f0d6f9 qemu driver checks emulator capabilities during domain XML post-parse. However, test suite does not initialize it, therefore a condition to skip all checks if there is no cache supplied was added. This is actually a hack, whose sole purpose is to make existing test suite working. Additionally, it prevents from writing new tests for this particular functionality. This series solves this problem by implementing proper cache mockup in test suite. The main idea is to create a cache in standard way and put there pre-defined capabilities sets (which tests already have). In this implementation we introduce "test capability sets", which are named entities. virQEMUCapsCacheLookup() now has a test mode, which is enabled by setting qemuTestCapsName variable to the name of capability set which we want to use in current test. This is based on Martin Kletzander's idea to implement test mode with assertions. Currently, since name of capability set is different from /usr/bin/something, we will get errors if we try to do something with the host filesystem. This is achieved by setting cacheDir and libDir of our mockup cache to "/dev/null". The concept of using named capability sets allows us to further improve the test suite. Now it is possibe to get rid of hardcoded capability sets and replace them with XML files, readable by virQEMUCapsInitCached(). This will allow us to better exercise our capabilities parser. Taking this into account, qemuTestCapsCacheInsert() is actually temporary. It will not be needed any more when capabilities are moved to XML files. Additionally, some small refactoring is done, and common driver initialization and cleanup sequence is put into two functions: qemuTestDriverInit() and qemuTestDriverFree(). I don't include changelist because this is a major rework of the RFC, almost everything has been changed. Pavel Fedin (3): Implement infrastracture for mocking up QEMU capabilities cache Use mockup cache Removed unneeded check src/qemu/qemu_capabilities.c | 16 +++++------- src/qemu/qemu_capspriv.h | 36 +++++++++++++++++++++++++ src/qemu/qemu_domain.c | 5 +--- tests/qemuagenttest.c | 18 +++++++++---- tests/qemuargv2xmltest.c | 19 +++++++------- tests/qemuhotplugtest.c | 32 ++++++++++++----------- tests/qemuxml2argvtest.c | 17 ++++++------ tests/qemuxml2xmltest.c | 16 ++++++++---- tests/qemuxmlnstest.c | 17 ++++++------ tests/testutilsqemu.c | 62 ++++++++++++++++++++++++++++++++++++++++++++ tests/testutilsqemu.h | 9 +++++++ 11 files changed, 182 insertions(+), 65 deletions(-) mode change 100644 => 100755 src/qemu/qemu_capabilities.c create mode 100644 src/qemu/qemu_capspriv.h mode change 100644 => 100755 tests/qemuagenttest.c mode change 100644 => 100755 tests/qemuargv2xmltest.c mode change 100644 => 100755 tests/qemuhotplugtest.c mode change 100644 => 100755 tests/qemuxml2argvtest.c mode change 100644 => 100755 tests/qemuxml2xmltest.c mode change 100644 => 100755 tests/qemuxmlnstest.c mode change 100644 => 100755 tests/testutilsqemu.c mode change 100644 => 100755 tests/testutilsqemu.h -- 2.1.4

2 7

[libvirt] [PATCH] conf/qemu: enforce NUMA nodes only for x86 memory hotplug
by Nikunj A Dadhania 15 Sep '15

15 Sep '15

libvirt enforces at least one NUMA node for memory hotplug support on all architectures. While it might be required for some x86 guest, PowerPC can hotplug memory on non-NUMA system. The generic checks are replaced with arch specific check and xml validation too does not enforce "node" for non-x86 arch. CC: Peter Krempa <pkrempa(a)redhat.com> Signed-off-by: Nikunj A Dadhania <nikunj(a)linux.vnet.ibm.com> --- src/conf/domain_conf.c | 9 ++++++--- src/qemu/qemu_command.c | 28 +++++++++++++++++----------- 2 files changed, 23 insertions(+), 14 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index fd0450f..4cb2d4a 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -12430,6 +12430,7 @@ virDomainMemorySourceDefParseXML(xmlNodePtr node, static int virDomainMemoryTargetDefParseXML(xmlNodePtr node, + const virDomainDef *domDef, xmlXPathContextPtr ctxt, virDomainMemoryDefPtr def) { @@ -12437,7 +12438,7 @@ virDomainMemoryTargetDefParseXML(xmlNodePtr node, xmlNodePtr save = ctxt->node; ctxt->node = node; - if (virXPathUInt("string(./node)", ctxt, &def->targetNode) < 0) { + if (virXPathUInt("string(./node)", ctxt, &def->targetNode) < 0 && ARCH_IS_X86(domDef->os.arch)) { virReportError(VIR_ERR_XML_ERROR, "%s", _("invalid or missing value of memory device node")); goto cleanup; @@ -12457,6 +12458,7 @@ virDomainMemoryTargetDefParseXML(xmlNodePtr node, static virDomainMemoryDefPtr virDomainMemoryDefParseXML(xmlNodePtr memdevNode, + const virDomainDef *domDef, xmlXPathContextPtr ctxt, unsigned int flags) { @@ -12495,7 +12497,7 @@ virDomainMemoryDefParseXML(xmlNodePtr memdevNode, goto error; } - if (virDomainMemoryTargetDefParseXML(node, ctxt, def) < 0) + if (virDomainMemoryTargetDefParseXML(node, domDef, ctxt, def) < 0) goto error; if (virDomainDeviceInfoParseXML(memdevNode, NULL, &def->info, flags) < 0) @@ -12647,7 +12649,7 @@ virDomainDeviceDefParse(const char *xmlStr, goto error; break; case VIR_DOMAIN_DEVICE_MEMORY: - if (!(dev->data.memory = virDomainMemoryDefParseXML(node, ctxt, flags))) + if (!(dev->data.memory = virDomainMemoryDefParseXML(node, def, ctxt, flags))) goto error; break; case VIR_DOMAIN_DEVICE_NONE: @@ -16328,6 +16330,7 @@ virDomainDefParseXML(xmlDocPtr xml, for (i = 0; i < n; i++) { virDomainMemoryDefPtr mem = virDomainMemoryDefParseXML(nodes[i], + def, ctxt, flags); if (!mem) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ae03618..51160e7 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -4979,8 +4979,12 @@ qemuBuildMemoryBackendStr(unsigned long long size, *backendProps = NULL; *backendType = NULL; - /* memory devices could provide a invalid guest node */ - if (guestNode >= virDomainNumaGetNodeCount(def->numa)) { + /* memory devices could provide a invalid guest node. Moreover, + * x86 guests needs at least one numa node to support memory + * hotplug + */ + if ((virDomainNumaGetNodeCount(def->numa) == 0 && ARCH_IS_X86(def->os.arch)) || + guestNode > virDomainNumaGetNodeCount(def->numa)) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("can't add memory backend for guest node '%d' as " "the guest has only '%zu' NUMA nodes configured"), @@ -4991,10 +4995,12 @@ qemuBuildMemoryBackendStr(unsigned long long size, if (!(props = virJSONValueNewObject())) return -1; - memAccess = virDomainNumaGetNodeMemoryAccessMode(def->numa, guestNode); - if (virDomainNumatuneGetMode(def->numa, guestNode, &mode) < 0 && - virDomainNumatuneGetMode(def->numa, -1, &mode) < 0) - mode = VIR_DOMAIN_NUMATUNE_MEM_STRICT; + if (virDomainNumaGetNodeCount(def->numa)) { + memAccess = virDomainNumaGetNodeMemoryAccessMode(def->numa, guestNode); + if (virDomainNumatuneGetMode(def->numa, guestNode, &mode) < 0 && + virDomainNumatuneGetMode(def->numa, -1, &mode) < 0) + mode = VIR_DOMAIN_NUMATUNE_MEM_STRICT; + } if (pagesize == 0) { /* Find the huge page size we want to use */ @@ -9238,11 +9244,11 @@ qemuBuildCommandLine(virConnectPtr conn, goto error; } - /* due to guest support, qemu would silently enable NUMA with one node - * once the memory hotplug backend is enabled. To avoid possible - * confusion we will enforce user originated numa configuration along - * with memory hotplug. */ - if (virDomainNumaGetNodeCount(def->numa) == 0) { + /* x86 windows guest needs at least one numa node to be + * present. While its not possible to detect what guest os is + * running, enforce this limitation only to x86 architecture. + */ + if (ARCH_IS_X86(def->os.arch) && virDomainNumaGetNodeCount(def->numa) == 0) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("At least one numa node has to be configured when " "enabling memory hotplug")); -- 2.4.3

3 5

[libvirt] [PATCH] rbd: Remove snapshots if the FORCED flag has been provided
by Wido den Hollander 14 Sep '15

14 Sep '15

When a RBD volume has snapshots it can not be removed. This patch introduces a new flag to force volume removal, VIR_STORAGE_VOL_DELETE_FORCED. With this flag any existing snapshots will be removed prior to removing the volume. No existing mechanism in libvirt allowed us to pass such information, so that's why a new flag was introduced. Signed-off-by: Wido den Hollander <wido(a)widodh.nl> --- include/libvirt/libvirt-storage.h | 1 + src/storage/storage_backend_rbd.c | 58 +++++++++++++++++++++++++++++++++++++++ 2 files changed, 59 insertions(+) diff --git a/include/libvirt/libvirt-storage.h b/include/libvirt/libvirt-storage.h index 453089e..36ff979 100644 --- a/include/libvirt/libvirt-storage.h +++ b/include/libvirt/libvirt-storage.h @@ -115,6 +115,7 @@ typedef enum { typedef enum { VIR_STORAGE_VOL_DELETE_NORMAL = 0, /* Delete metadata only (fast) */ VIR_STORAGE_VOL_DELETE_ZEROED = 1 << 0, /* Clear all data to zeros (slow) */ + VIR_STORAGE_VOL_DELETE_FORCED = 2, /* Force removal of volume, even if in use */ } virStorageVolDeleteFlags; typedef enum { diff --git a/src/storage/storage_backend_rbd.c b/src/storage/storage_backend_rbd.c index ac5085a..ca5e802 100644 --- a/src/storage/storage_backend_rbd.c +++ b/src/storage/storage_backend_rbd.c @@ -428,9 +428,13 @@ static int virStorageBackendRBDDeleteVol(virConnectPtr conn, { int ret = -1; int r = 0; + int max_snaps = 128; + int i, snap_count, protected; virStorageBackendRBDState ptr; ptr.cluster = NULL; ptr.ioctx = NULL; + rbd_snap_info_t *snaps; + rbd_image_t image = NULL; VIR_DEBUG("Removing RBD image %s/%s", pool->def->source.name, vol->name); @@ -443,6 +447,59 @@ static int virStorageBackendRBDDeleteVol(virConnectPtr conn, if (virStorageBackendRBDOpenIoCTX(&ptr, pool) < 0) goto cleanup; + r = rbd_open(ptr.ioctx, vol->name, &image, NULL); + if (r < 0) { + virReportSystemError(-r, _("failed to open the RBD image '%s'"), + vol->name); + goto cleanup; + } + + do { + if (VIR_ALLOC_N(snaps, max_snaps)) + goto cleanup; + + snap_count = rbd_snap_list(image, snaps, &max_snaps); + if (snap_count <= 0) { + VIR_FREE(snaps); + } + } while (snap_count == -ERANGE); + + VIR_DEBUG("Found %d snapshots for volume %s/%s", snap_count, + pool->def->source.name, vol->name); + + if (snap_count > 0 && (flags & VIR_STORAGE_VOL_DELETE_FORCED)) { + for (i = 0; i < snap_count; i++) { + if (rbd_snap_is_protected(image, snaps[i].name, &protected)) + goto cleanup; + + if (protected == 1) { + VIR_DEBUG("Snapshot %s/%s@%s is protected needs to be " + "unprotected", pool->def->source.name, vol->name, + snaps[i].name); + + if (rbd_snap_unprotect(image, snaps[i].name) < 0) + goto cleanup; + } + + VIR_DEBUG("Removing snapshot %s/%s@%s", pool->def->source.name, + vol->name, snaps[i].name); + + if (rbd_snap_remove(image, snaps[i].name) < 0) { + virReportSystemError(-r, _("failed to remove snapshot '%s/%s@%s'"), + pool->def->source.name, vol->name, + snaps[i].name); + goto cleanup; + } + } + + rbd_snap_list_end(snaps); + } + + if (rbd_close(image) < 0) + goto cleanup; + + VIR_DEBUG("Removing volume %s/%s", pool->def->source.name, vol->name); + r = rbd_remove(ptr.ioctx, vol->name); if (r < 0 && (-r) != ENOENT) { virReportSystemError(-r, _("failed to remove volume '%s/%s'"), @@ -453,6 +510,7 @@ static int virStorageBackendRBDDeleteVol(virConnectPtr conn, ret = 0; cleanup: + VIR_FREE(snaps); virStorageBackendRBDCloseRADOSConn(&ptr); return ret; } -- 1.9.1

1 0

[libvirt] [PATCH] qemuDomainUpdateDeviceConfig: Allow startupPolicy update, yet again
by Michal Privoznik 14 Sep '15

14 Sep '15

https://bugzilla.redhat.com/show_bug.cgi?id=1159219 So, in 11e058ca589808bd I've tried to make UpdateDevice update startupPolicy too. And it worked well until somebody came around and pushed d0dc6c036914da which accidentally removed my contribution. Redo my commit. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 91eb661..03ef972 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -8353,6 +8353,7 @@ qemuDomainUpdateDeviceConfig(virQEMUCapsPtr qemuCaps, * We allow updating src/type//driverType/cachemode/ */ orig->cachemode = disk->cachemode; + orig->startupPolicy = disk->startupPolicy; virStorageSourceFree(orig->src); orig->src = disk->src; -- 2.4.6

2 3

[libvirt] [PATCH] util: Add space before comment end marker
by Andrea Bolognani 14 Sep '15

14 Sep '15

This allows the Wikipedia link to be recognized correctly by eg. gnome-terminal's Open Link and Copy Link Address features. --- Pushed as trivial src/util/virarch.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/util/virarch.h b/src/util/virarch.h index 3206ce2..af5ff83 100644 --- a/src/util/virarch.h +++ b/src/util/virarch.h @@ -45,7 +45,7 @@ typedef enum { VIR_ARCH_MIPS64, /* MIPS 64 BE http://en.wikipedia.org/wiki/MIPS_architecture */ VIR_ARCH_MIPS64EL, /* MIPS 64 LE http://en.wikipedia.org/wiki/MIPS_architecture */ - VIR_ARCH_OR32, /* OpenRisc 32 BE http://en.wikipedia.org/wiki/OpenRISC#QEMU_support*/ + VIR_ARCH_OR32, /* OpenRisc 32 BE http://en.wikipedia.org/wiki/OpenRISC#QEMU_support */ VIR_ARCH_PARISC, /* PA-Risc 32 BE http://en.wikipedia.org/wiki/PA-RISC */ VIR_ARCH_PARISC64, /* PA-Risc 64 BE http://en.wikipedia.org/wiki/PA-RISC */ -- 2.4.3

1 0

[libvirt] [PATCH] virsh: Fix job status indicator for 0 length block jobs
by Peter Krempa 14 Sep '15

14 Sep '15

Although 0 length block jobs aren't entirely useful, the output of virsh blockjob is empty due to the condition that suppresses the output for migration jobs that did not start. Add a flag for virshPrintJobProgress that specifies that the job was started so that the function can be used both in migration where the job starts delayed and for block jobs where we already know that the job was started. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1196711 --- tools/virsh-domain.c | 24 +++++++++++++++++------- 1 file changed, 17 insertions(+), 7 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index b029b65..7837974 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -1683,13 +1683,23 @@ cmdBlkiotune(vshControl * ctl, const vshCmd * cmd) } +/** + * virshPrintJobProgress: + * @label: name of the block job + * @remaining: amount of remaining work + * @total: amount of total work + * @started: The block job was started already + * + * Prints a human friendly percentage string decribing the state of the given + * job. The output is omitted if the job was not started. + */ static void virshPrintJobProgress(const char *label, unsigned long long remaining, - unsigned long long total) + unsigned long long total, bool started) { int progress; - if (total == 0) + if (total == 0 && !started) /* migration has not been started */ return; @@ -1921,7 +1931,7 @@ virshBlockJobWait(virshBlockJobWaitDataPtr data) if (data->verbose) virshPrintJobProgress(data->job_name, info.end - info.cur, - info.end); + info.end, true); if (data->timeout && virTimeMillisNow(&curr) < 0) { vshSaveLibvirtError(); @@ -1949,7 +1959,7 @@ virshBlockJobWait(virshBlockJobWaitDataPtr data) if (data->verbose && (ret == VIR_DOMAIN_BLOCK_JOB_COMPLETED || ret == VIR_DOMAIN_BLOCK_JOB_READY)) - virshPrintJobProgress(data->job_name, 0, 1); + virshPrintJobProgress(data->job_name, 0, 1, true); sigaction(SIGINT, &old_sig_action, NULL); return ret; @@ -2621,7 +2631,7 @@ virshBlockJobInfo(vshControl *ctl, info.bandwidth, info.cur, info.end); } else { virshPrintJobProgress(virshDomainBlockJobToString(info.type), - info.end - info.cur, info.end); + info.end - info.cur, info.end, true); if (speed) { const char *unit; double val = vshPrettyCapacity(speed, &unit); @@ -4356,7 +4366,7 @@ virshWatchJob(vshControl *ctl, retchar == '0') { if (verbose) { /* print [100 %] */ - virshPrintJobProgress(label, 0, 1); + virshPrintJobProgress(label, 0, 1, true); } break; } @@ -4392,7 +4402,7 @@ virshWatchJob(vshControl *ctl, if (ret == 0) { if (verbose) virshPrintJobProgress(label, jobinfo.dataRemaining, - jobinfo.dataTotal); + jobinfo.dataTotal, false); if (!jobStarted && (jobinfo.type == VIR_DOMAIN_JOB_BOUNDED || -- 2.4.5

2 1

[libvirt] [PATCH 0/2] Fix permissions for per-VM and qemu directories
by Martin Kletzander 14 Sep '15

14 Sep '15

This is basically v2 of: https://www.redhat.com/archives/libvir-list/2015-September/msg00295.html with some changes: - permissions for per-VM dirs are fixed as well - permissions for the /var/lib/libvirt/qemu directory are fixed in the install phase of Makefile.am in addition to the spec file We could also fix the permissions in qemuStateInitialize, but I chose to do it this way. Bear in mind, that both approaches have their pros and cons, of course. Martin Kletzander (2): qemu: Do not allow others into per-VM subdirectories qemu: Allow others to browse /var/lib/libvirt/qemu libvirt.spec.in | 4 ++-- src/Makefile.am | 2 +- src/qemu/qemu_process.c | 4 ++-- 3 files changed, 5 insertions(+), 5 deletions(-) -- 2.5.2

2 3

[libvirt] [PATCH] qemu: hotplug: Properly clean up drive backend if frontend hotplug fails
by Peter Krempa 14 Sep '15

14 Sep '15

Commit 8125113c added code that should remove the disk backend if the fronted hotplug failed for any reason. The code had a bug though as it used the disk string for unplug rather than the backend alias. Fix the code by pre-creating an alias string and using it instead of the disk string. In cases where qemu does not support QEMU_CAPS_DEVICE, we ignore the unplug of the backend since we can't really create an alias in that case. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1262399 --- src/qemu/qemu_hotplug.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 63fafa6..c956e8c 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -321,6 +321,7 @@ qemuDomainAttachVirtioDiskDevice(virConnectPtr conn, qemuDomainObjPrivatePtr priv = vm->privateData; char *devstr = NULL; char *drivestr = NULL; + char *drivealias = NULL; bool releaseaddr = false; virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); const char *src = virDomainDiskGetSource(disk); @@ -365,6 +366,10 @@ qemuDomainAttachVirtioDiskDevice(virConnectPtr conn, if (!(drivestr = qemuBuildDriveStr(conn, disk, false, priv->qemuCaps))) goto error; + if (virAsprintf(&drivealias, "%s%s", QEMU_DRIVE_HOST_PREFIX, + disk->info.alias) < 0) + goto error; + if (!(devstr = qemuBuildDriveDevStr(vm->def, disk, 0, priv->qemuCaps))) goto error; } @@ -379,7 +384,8 @@ qemuDomainAttachVirtioDiskDevice(virConnectPtr conn, ret = qemuMonitorAddDevice(priv->mon, devstr); if (ret < 0) { virErrorPtr orig_err = virSaveLastError(); - if (qemuMonitorDriveDel(priv->mon, drivestr) < 0) { + if (!drivealias || + qemuMonitorDriveDel(priv->mon, drivealias) < 0) { VIR_WARN("Unable to remove drive %s (%s) after failed " "qemuMonitorAddDevice", drivestr, devstr); @@ -415,6 +421,7 @@ qemuDomainAttachVirtioDiskDevice(virConnectPtr conn, cleanup: VIR_FREE(devstr); VIR_FREE(drivestr); + VIR_FREE(drivealias); virObjectUnref(cfg); return ret; -- 2.4.5

2 2

[libvirt] LSN-2015-0003: CVE-2015-5247 denial of service through root-squash NFS storage pools
by Eric Blake 12 Sep '15

12 Sep '15

Libvirt Security Notice: LSN-2015-0003 ====================================== Summary: denial of service through root-squash NFS storage pools Reported on: 20150814 Published on: 20150903 Fixed on: 20150903 Reported by: Han Han <hhan(a)redhat.com> Patched by: John Ferlan <jferlan(a)redhat.com> See also: CVE-2015-5247 Description ----------- The virStorageVolCreateXML API had a bug where it could create a volume on a root-squash NFS mount, but then fail to remove that volume if later steps during the API encountered problems. This was further compounded by code which used a wrong conditional on whether the new volume needed to have permissions changed, making it more likely to trigger the failed unlink attempt. Poor error handling after a failed unlink left libvirt with an inconsistent view of the storage volume that could then result in a libvirtd crash. While the libvirtd crash might be delayed until by subsequent actions from a read-only connection, the conditions that set up the crash can only be triggered by a client with a read-write connection. Impact ------ When using fine-grained Access Control Lists (ACL), the virStorageVolCreateXML API only requires the storage_vol:create permission. A client with this privilege but lacking the more-powerful domain:write permission could exploit the API bugs to cause a denial-of-service attack by taking down libvirtd through a crash. It can also be argued that the ability to cause libvirt to create files which it cannot delete can be used as a denial-of-service attack on storage resources. Workaround ---------- The problems with libvirt creating a file which it does not then clean up on error is specific to root-squash NFS, so one mitigation is avoiding the use of the root-squash option when exporting NFS volumes for use by libvirt storage pools. Note that in general, the use of root-squash NFS does not add any real security (it makes certain tasks harder for a root user, but the root user can trivially change ids to another user to still perform those tasks). Furthermore, it is possible to prevent the denial of service attacks by stopping the use of the fine grained access control mechanism (while this does not prevent a crash, such a crash is no longer a security problem as there is no longer a privilege boundary between a user creating a volume and a user with full system access). Affected product ---------------- Name: libvirt Repository: git://libvirt.org/git/libvirt.git http://libvirt.org/git/?p=libvirt.git Branch: master Broken in: v1.2.14 Broken in: v1.2.15 Broken in: v1.2.16 Broken in: v1.2.17 Broken in: v1.2.18 Broken in: v1.2.19 Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Broken by: 7c2d65dde2595c07d56aad1e043f7b1836592d89 Fixed by: db9277a39bc364806e8d3e08a08fc128d59b7094 Fixed by: 691dd388aee99f8b06177540303b690586d5f5b3 Fixed by: 35847860f65f92e444db9730e00cdaef45198e0c Branch: v1.2.14-maint Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Fixed by: 605b12068392d29beb44a8ab7d6ec176d6b05237 Fixed by: 454cb7c40dbcff84192094963d71369ac7d94546 Branch: v1.2.15-maint Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Fixed by: 3c41b3ea5e68f391b8ff901082608bda5f7f3fbc Fixed by: fe2cf73800e3be87d1d4d811facb3f2be48126e5 Branch: v1.2.16-maint Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Broken by: 7c2d65dde2595c07d56aad1e043f7b1836592d89 Fixed by: 9e48400f4606bac16b7e4db195f610928c3d5a04 Fixed by: 2f4b41861c1729ff4b754986782d7428ccdca455 Fixed by: 7f0505705c70f7eb1e435a2e7732d1a74abfadfd Branch: v1.2.17-maint Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Broken by: 7c2d65dde2595c07d56aad1e043f7b1836592d89 Fixed by: d055989083df4bf68eb1388d327ebffb3501bb83 Fixed by: 98242f94cd181f0257535479369054f07f951b21 Fixed by: a3ee6885d95a2ce6fb7e58bb0737cfb1612e0fb7 Branch: v1.2.18-maint Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Broken by: 7c2d65dde2595c07d56aad1e043f7b1836592d89 Fixed by: e63b32e22dafd99547f82f5383fdbf58b5f651a1 Fixed by: 075eb526c9817d9d8e3a759e3fbe180d8d326dcf Fixed by: 966cc922221be2b8cc6a9842ed0dc4cf1568a7b3 Branch: v1.2.19-maint Broken by: 155ca616eb231181f6978efc9e3a1eb0eb60af8a Broken by: 7c2d65dde2595c07d56aad1e043f7b1836592d89 Fixed by: e0025d2967bbe3f283937216c9e2c12b6e9d1010 Fixed by: 8b1d84e640f1a6e6ebb47caf23a664e2f651b32d Fixed by: 3468542f06f6f5dc94defa1603c6a6adea3e2da8 -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org

1 0

[libvirt] [PATCH] qemu: command: Report stderr from qemu-bridge-helper
by Cole Robinson 11 Sep '15

11 Sep '15

There's a couple reports of things failing in this area (bug 1259070), but it's tough to tell what's going wrong without stderr from qemu-bridge-helper. So let's report stderr in the error message Couple new examples: virbr0 is inactive: internal error: /usr/libexec/qemu-bridge-helper --use-vnet --br=virbr0 --fd=21: failed to retrieve file descriptor: Transport endpoint is not connected stderr=failed to get mtu of bridge `virbr0': No such device bridge isn't on the ACL: internal error: /usr/libexec/qemu-bridge-helper --use-vnet --br=br0 --fd=21: failed to retrieve file descriptor: Transport endpoint is not connected stderr=access denied by acl file --- src/qemu/qemu_command.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index ec5e3d4..fc22f36 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -285,6 +285,7 @@ static int qemuCreateInBridgePortWithHelper(virQEMUDriverConfigPtr cfg, unsigned int flags) { virCommandPtr cmd; + char *errbuf = NULL, *cmdstr = NULL; int pair[2] = { -1, -1 }; if ((flags & ~VIR_NETDEV_TAP_CREATE_VNET_HDR) != VIR_NETDEV_TAP_CREATE_IFUP) @@ -306,6 +307,8 @@ static int qemuCreateInBridgePortWithHelper(virQEMUDriverConfigPtr cfg, virCommandAddArgFormat(cmd, "--use-vnet"); virCommandAddArgFormat(cmd, "--br=%s", brname); virCommandAddArgFormat(cmd, "--fd=%d", pair[1]); + virCommandSetErrorBuffer(cmd, &errbuf); + virCommandDoAsyncIO(cmd); virCommandPassFD(cmd, pair[1], VIR_COMMAND_PASS_FD_CLOSE_PARENT); virCommandClearCaps(cmd); @@ -320,9 +323,24 @@ static int qemuCreateInBridgePortWithHelper(virQEMUDriverConfigPtr cfg, do { *tapfd = recvfd(pair[0], 0); } while (*tapfd < 0 && errno == EINTR); + if (*tapfd < 0) { - virReportSystemError(errno, "%s", - _("failed to retrieve file descriptor for interface")); + char ebuf[1024]; + char *errstr = NULL; + + if (!(cmdstr = virCommandToString(cmd))) + goto cleanup; + virCommandAbort(cmd); + + if (errbuf && *errbuf && + virAsprintf(&errstr, "\nstderr=%s", errbuf) < 0) + goto cleanup; + + virReportError(VIR_ERR_INTERNAL_ERROR, + _("%s: failed to retrieve file descriptor: %s%s"), + cmdstr, virStrerror(errno, ebuf, sizeof(ebuf)), + errstr ? errstr : ""); + VIR_FREE(errstr); goto cleanup; } @@ -333,6 +351,8 @@ static int qemuCreateInBridgePortWithHelper(virQEMUDriverConfigPtr cfg, } cleanup: + VIR_FREE(cmdstr); + VIR_FREE(errbuf); virCommandFree(cmd); VIR_FORCE_CLOSE(pair[0]); return *tapfd < 0 ? -1 : 0; -- 2.5.0

2 2

[libvirt] [PATCH LIBVIRT] libxl: don't end job for ephemeal domain on start failure
by Ian Campbell 11 Sep '15

11 Sep '15

commit 4b53d0d4ac9c "libxl: don't remove persistent domain on start failure" cleans up the vm object and sets it to NULL if the vm is not persistent, however at end job vm (now NULL) is dereferenced via the call to libxlDomainObjEndJob. Avoid this by skipping "endjob" and going straight to "cleanup" in this case. Signed-off-by: Ian Campbell <ian.campbell(a)citrix.com> --- src/libxl/libxl_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index 5f69b49..e2797d5 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -992,6 +992,7 @@ libxlDomainCreateXML(virConnectPtr conn, const char *xml, if (!vm->persistent) { virDomainObjListRemove(driver->domains, vm); vm = NULL; + goto cleanup; } goto endjob; } -- 2.1.4

3 2

[libvirt] [PATCH] xen: fix race in refresh of config cache
by Daniel P. Berrange 11 Sep '15

11 Sep '15

The xenXMConfigCacheRefresh method scans /etc/xen and loads all config files it finds. It then scans its internal hash table and purges any (previously) loaded config files whose refresh timestamp does not match the timestamp recorded at the start of xenXMConfigCacheRefresh(). There is unfortunately a subtle flaw in this, because if loading the config files takes longer than 1 second, some of the config files will have a refresh timestamp that is 1 or more seconds different (newer) than is checked for. So we immediately purge a bunch of valid config files we just loaded. To avoid this flaw, we must pass the timestamp we record at the start of xenXMConfigCacheRefresh() into the xenXMConfigCacheAddFile() method, instead of letting the latter call time(NULL) again. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/xen/xen_inotify.c | 10 ++++++---- src/xen/xm_internal.c | 7 +++---- src/xen/xm_internal.h | 2 +- 3 files changed, 10 insertions(+), 9 deletions(-) diff --git a/src/xen/xen_inotify.c b/src/xen/xen_inotify.c index cd1e2df..d81a35d 100644 --- a/src/xen/xen_inotify.c +++ b/src/xen/xen_inotify.c @@ -217,11 +217,11 @@ xenInotifyRemoveDomainConfigInfo(virConnectPtr conn, const char *fname) } static int -xenInotifyAddDomainConfigInfo(virConnectPtr conn, const char *fname) +xenInotifyAddDomainConfigInfo(virConnectPtr conn, const char *fname, time_t now) { xenUnifiedPrivatePtr priv = conn->privateData; return priv->useXenConfigCache ? - xenXMConfigCacheAddFile(conn, fname) : + xenXMConfigCacheAddFile(conn, fname, now) : xenInotifyXendDomainsDirAddEntry(conn, fname); } @@ -238,6 +238,7 @@ xenInotifyEvent(int watch ATTRIBUTE_UNUSED, char *tmp, *name; virConnectPtr conn = data; xenUnifiedPrivatePtr priv = NULL; + time_t now = time(NULL); VIR_DEBUG("got inotify event"); @@ -300,7 +301,7 @@ xenInotifyEvent(int watch ATTRIBUTE_UNUSED, } } else if (e->mask & (IN_CREATE | IN_CLOSE_WRITE | IN_MOVED_TO)) { virObjectEventPtr event; - if (xenInotifyAddDomainConfigInfo(conn, fname) < 0) { + if (xenInotifyAddDomainConfigInfo(conn, fname, now) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Error adding file to config cache")); goto cleanup; @@ -344,6 +345,7 @@ xenInotifyOpen(virConnectPtr conn, char *path; xenUnifiedPrivatePtr priv = conn->privateData; int direrr; + time_t now = time(NULL); virCheckFlags(VIR_CONNECT_RO, -1); @@ -374,7 +376,7 @@ xenInotifyOpen(virConnectPtr conn, return -1; } - if (xenInotifyAddDomainConfigInfo(conn, path) < 0) { + if (xenInotifyAddDomainConfigInfo(conn, path, now) < 0) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Error adding file to config list")); closedir(dh); diff --git a/src/xen/xm_internal.c b/src/xen/xm_internal.c index 59b1cd4..07b9eb4 100644 --- a/src/xen/xm_internal.c +++ b/src/xen/xm_internal.c @@ -191,15 +191,14 @@ xenXMConfigCacheRemoveFile(virConnectPtr conn, const char *filename) * calling this function */ int -xenXMConfigCacheAddFile(virConnectPtr conn, const char *filename) +xenXMConfigCacheAddFile(virConnectPtr conn, const char *filename, time_t now) { xenUnifiedPrivatePtr priv = conn->privateData; xenXMConfCachePtr entry; struct stat st; int newborn = 0; - time_t now = time(NULL); - VIR_DEBUG("Adding file %s", filename); + VIR_DEBUG("Adding file %s %lld", filename, (long long)now); /* Get modified time */ if ((stat(filename, &st) < 0)) { @@ -373,7 +372,7 @@ xenXMConfigCacheRefresh(virConnectPtr conn) /* If we already have a matching entry and it is not modified, then carry on to next one*/ - if (xenXMConfigCacheAddFile(conn, path) < 0) { + if (xenXMConfigCacheAddFile(conn, path, now) < 0) { /* Ignoring errors, since a lot of stuff goes wrong in /etc/xen */ } diff --git a/src/xen/xm_internal.h b/src/xen/xm_internal.h index 25b4fd5..0246895 100644 --- a/src/xen/xm_internal.h +++ b/src/xen/xm_internal.h @@ -31,7 +31,7 @@ # include "domain_conf.h" int xenXMConfigCacheRefresh (virConnectPtr conn); -int xenXMConfigCacheAddFile(virConnectPtr conn, const char *filename); +int xenXMConfigCacheAddFile(virConnectPtr conn, const char *filename, time_t now); int xenXMConfigCacheRemoveFile(virConnectPtr conn, const char *filename); int xenXMOpen(virConnectPtr conn, virConnectAuthPtr auth, unsigned int flags); -- 2.4.3

2 1

[libvirt] [PATCH] tests: Don't use testutils in mock libraries
by Martin Kletzander 11 Sep '15

11 Sep '15

Mock libraries are not built with testutils.c, but there's one which uses VIR_TEST_DEBUG. But because that debug should be an error, if we change it, then it will not only be more semantically correct, but mingw compiler will be happier as well. It also follows suit with all other mock libraries. For few other things, used in this file, need libvirt.la to be added into LIBADD for mingw as well. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- tests/Makefile.am | 1 + tests/virportallocatortest.c | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/tests/Makefile.am b/tests/Makefile.am index bde7f5baf1fe..4af38fe9c14b 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -1020,6 +1020,7 @@ libvirportallocatormock_la_SOURCES = \ virportallocatortest.c libvirportallocatormock_la_CFLAGS = $(AM_CFLAGS) -DMOCK_HELPER=1 libvirportallocatormock_la_LDFLAGS = $(MOCKLIBS_LDFLAGS) +libvirportallocatormock_la_LIBADD = ../src/libvirt.la vircgrouptest_SOURCES = \ vircgrouptest.c testutils.h testutils.c diff --git a/tests/virportallocatortest.c b/tests/virportallocatortest.c index 1125b0aeedda..58b2155994f9 100644 --- a/tests/virportallocatortest.c +++ b/tests/virportallocatortest.c @@ -49,7 +49,7 @@ static void init_syms(void) realsocket = dlsym(RTLD_NEXT, "socket"); if (!realsocket) { - VIR_TEST_DEBUG("Unable to find 'socket' symbol\n"); + fprintf(stderr, "Unable to find 'socket' symbol\n"); abort(); } -- 2.5.1

2 1

[libvirt] (no subject)
by Wido den Hollander 10 Sep '15

10 Sep '15

The current code of libvirt-java does not build with Java 7. This patch updates the build.properties and INSTALL that we require Java 8 to build libvirt-java.

3 8

[libvirt] [java] [PATCH 0/2] Fix a few compilation problems
by Claudio Bley 10 Sep '15

10 Sep '15

Initially reported by Wido den Hollander here[1]. [1]: http://www.redhat.com/archives/libvir-list/2015-September/msg00052.html Claudio Bley (2): Fix compilation error for unreported exceptions Fix compilation error when using -source 1.x src/main/java/org/libvirt/Library.java | 2 +- src/main/java/org/libvirt/event/DomainEvent.java | 4 +++- src/main/java/org/libvirt/event/DomainEventType.java | 7 ------- 3 files changed, 4 insertions(+), 9 deletions(-) -- 2.5.1

1 2

[libvirt] [PATCH] docs: Update devguide.html to point to the new Python dev guide
by Daniel P. Berrange 10 Sep '15

10 Sep '15

We have a new libvirt-appdev-guide-python which we need to promote to users. Rewrite the existing page to mention it too. Also use the new URL location which is automatically refreshed once a day. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- docs/devguide.html.in | 56 +++++++++++++++++++-------------------------------- 1 file changed, 21 insertions(+), 35 deletions(-) diff --git a/docs/devguide.html.in b/docs/devguide.html.in index 122fe24..639e495 100644 --- a/docs/devguide.html.in +++ b/docs/devguide.html.in @@ -2,55 +2,41 @@ <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"> <html xmlns="http://www.w3.org/1999/xhtml"> <body> - <h1>libvirt Application Development Guide</h1> - - <ul id="toc"></ul> - - <p> - The guide is both a learning tool for developing with libvirt and an - API reference document. It is a work in progress, composed by a - professional author from contributions written by members of the - libvirt team. - </p> + <h1>libvirt Application Development Guides</h1> <p> - Contributions to the guide are <b>VERY</b> welcome. If you'd like to get - your name on this and demonstrate your virtualisation prowess, a solid - contribution to the content here will do it. :) + The libvirt API is accessible from a number of programming languages. + At this time, there are application development guides available + which cover the C API and the Python API. Of the two, the Python guide + is currently the more comprehensive document. </p> - <h2><a name="online">Browsable online</a></h2> - <ul> - <li><a href="http://libvirt.org/guide/html/"> - HTML format using multiple pages</a></li> - <li><a href="http://libvirt.org/guide/html-single/"> - HTML format using one big page</a></li> - <li><a href="http://libvirt.org/guide/pdf/Application_Development_Guide.pdf"> - PDF format</a></li> - <li><a href="http://libvirt.org/guide/libvirt-0.7.5-Application_Development_Guide-en-US.…"> - ePub format</a></li> - <li><a href="http://libvirt.org/guide/txt/Application_Development_Guide.txt"> - Plain text format</a></li> - <li><a href="http://libvirt.org/guide/libvirt-Application_Development_Guide-0.7.5-web-en…"> - Source RPM format</a></li> + <li><a href="http://libvirt.org/docs/libvirt-appdev-guide/en-US/html/">Application Development Guide (C language) HTML</a></li> + <li><a href="http://libvirt.org/docs/libvirt-appdev-guide/en-US/pdf/">Application Development Guide (C language) PDF</a></li> + <li><a href="http://libvirt.org/docs/libvirt-appdev-guide-python/en-US/html/">Application Development Guide (Python language) HTML</a></li> + <li><a href="http://libvirt.org/docs/libvirt-appdev-guide-python/en-US/pdf/">Application Development Guide (Python language) PDF</a></li> </ul> - <h2><a name="git">GIT source repository</a></h2> + <h2>Contributing content</h2> <p> - The source is in a git repository: + These guides are written in DocBook and published with the + publican tool, which is also used for Fedora and Red Hat + documentation. The original content is provided in GIT and + any contributions to the guide are welcome. </p> <pre> - git clone git://libvirt.org/libvirt-appdev-guide.git</pre> + # C language + $ git clone <a href="http://libvirt.org/git/?p=libvirt-appdev-guide.git">git://libvirt.org/libvirt-appdev-guide.git</a> - <p> - Browsable here: - </p> + # Python language + $ git clone <a href="http://libvirt.org/git/?p=libvirt-appdev-guide-python.git">git://libvirt.org/libvirt-appdev-guide-python.git</a> - <pre> - <a href="http://libvirt.org/git/?p=libvirt-appdev-guide.git;a=summary">http://libvirt.org/git/?p=libvirt-appdev-guide.git;a=summary</a></pre> + # Publican Style/Theme + $ git clone <a href="http://libvirt.org/git/?p=libvirt-publican.git">git://libvirt.org/libvirt-publican.git</a> + </pre> </body> </html> -- 2.4.3

2 1

[libvirt] [PATCH v3 0/4] Try several possible default network devices
by Andrea Bolognani 10 Sep '15

10 Sep '15

Changes in v3: * detect rtl8139 and e1000 availability as well * don't try to use virtio-net based on availability alone: try rtl8139, e1000 and virtio-net in this order instead * patch 4/4 contains an ugly bit we'll be able to get rid of once capabilities can be properly mocked in the test suite Changes in v2[1]: * detect virtio-net availability * try to user it if available Cheers. [1] https://www.redhat.com/archives/libvir-list/2015-September/msg00058.html Andrea Bolognani (4): qemu: Introduce QEMU_CAPS_DEVICE_RTL8139 qemu: Introduce QEMU_CAPS_DEVICE_E1000 qemu: Introduce QEMU_CAPS_DEVICE_VIRTIO_NET qemu: Try several network devices when looking for a default src/qemu/qemu_capabilities.c | 12 +++++++++- src/qemu/qemu_capabilities.h | 3 +++ src/qemu/qemu_domain.c | 32 +++++++++++++++++++++++---- tests/qemucapabilitiesdata/caps_1.2.2-1.caps | 3 +++ tests/qemucapabilitiesdata/caps_1.3.1-1.caps | 3 +++ tests/qemucapabilitiesdata/caps_1.4.2-1.caps | 3 +++ tests/qemucapabilitiesdata/caps_1.5.3-1.caps | 3 +++ tests/qemucapabilitiesdata/caps_1.6.0-1.caps | 3 +++ tests/qemucapabilitiesdata/caps_1.6.50-1.caps | 3 +++ tests/qemucapabilitiesdata/caps_2.1.1-1.caps | 3 +++ tests/qemuhelptest.c | 30 ++++++++++++++++++++----- 11 files changed, 87 insertions(+), 11 deletions(-) -- 2.4.3

2 8

[libvirt] [PATCH sandbox v5 00/20] Docker image support
by Daniel P. Berrange 10 Sep '15

10 Sep '15

This is an update of Eren's v4 patch series to provide Docker image support. In this v5 I have incorporated fixes for all of the feedback I gave Eren against the v4, so I think this series is ready for merging now. One thing I still want to look at separately is how applicable this design is to other container formats, in particular the 'appc' specification https://github.com/appc/spec/blob/master/spec/discovery.md Before we make a release containing the docker impl, I want to be confident we've not done anything silly which will cause us compat problems if we want to extend to cover appc later. Daniel P Berrange (1): Add virt-sandbox-image Daniel P. Berrange (1): Rename 'name' to 'template' to disambiguate Eren Yagdiran (18): Fix docker authentication handling Image: Add Hooking Mechanism Image: virt-sandbox-image default dir constants Image: Add download function Image: Refactor create function Image: Add delete function Image: Add get_command function to Source Image: Add run args Image: Add check_connect function Image: Add get_disk function to Source Image: Add run function Image: Add network support Image: man file for virt-sandbox-image Add config for environment variables Add environment parameter to virt-sandbox init-common: Exporting custom environment variables Image: Add custom environment support Image: Add Volume Support .gitignore | 1 + bin/Makefile.am | 8 +- bin/virt-sandbox-image | 8 + bin/virt-sandbox-image.pod | 169 +++++++++++ bin/virt-sandbox.c | 14 + configure.ac | 3 + libvirt-sandbox.spec.in | 3 + libvirt-sandbox/Makefile.am | 2 +- libvirt-sandbox/image/Makefile.am | 10 + libvirt-sandbox/image/__init__.py | 0 libvirt-sandbox/image/cli.py | 293 ++++++++++++++++++ libvirt-sandbox/image/sources/DockerSource.py | 416 ++++++++++++++++++++++++++ libvirt-sandbox/image/sources/Makefile.am | 9 + libvirt-sandbox/image/sources/Source.py | 117 ++++++++ libvirt-sandbox/image/sources/__init__.py | 0 libvirt-sandbox/libvirt-sandbox-config.c | 171 ++++++++++- libvirt-sandbox/libvirt-sandbox-config.h | 13 + libvirt-sandbox/libvirt-sandbox-init-common.c | 18 ++ libvirt-sandbox/libvirt-sandbox.sym | 9 + libvirt-sandbox/tests/test-config.c | 10 + po/POTFILES.in | 1 + 21 files changed, 1272 insertions(+), 3 deletions(-) create mode 100644 bin/virt-sandbox-image create mode 100644 bin/virt-sandbox-image.pod create mode 100644 libvirt-sandbox/image/Makefile.am create mode 100644 libvirt-sandbox/image/__init__.py create mode 100755 libvirt-sandbox/image/cli.py create mode 100644 libvirt-sandbox/image/sources/DockerSource.py create mode 100644 libvirt-sandbox/image/sources/Makefile.am create mode 100644 libvirt-sandbox/image/sources/Source.py create mode 100644 libvirt-sandbox/image/sources/__init__.py -- 2.4.3

3 48

[libvirt] [PATCH] target-i386: Enable "check" mode by default
by Eduardo Habkost 09 Sep '15

09 Sep '15

Current default behavior of QEMU is to silently disable features that are not supported by the host when a CPU model is requested in the command-line. This means that in addition to risking breaking guest ABI by default, we are silent about it. I would like to enable "enforce" by default, but this can easily break existing production systems because of the way libvirt makes assumptions about CPU models today (this will change in the future, once QEMU provide a proper interface for checking if a CPU model is runnable). But there's no reason we should be silent about it. So, change target-i386 to enable "check" mode by default so at least we have some warning printed to stderr (and hopefully logged somewhere) when QEMU disables a feature that is not supported by the host system. Signed-off-by: Eduardo Habkost <ehabkost(a)redhat.com> --- target-i386/cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/target-i386/cpu.c b/target-i386/cpu.c index cfb8aa7..32fee00 100644 --- a/target-i386/cpu.c +++ b/target-i386/cpu.c @@ -3121,7 +3121,7 @@ static Property x86_cpu_properties[] = { DEFINE_PROP_BOOL("hv-relaxed", X86CPU, hyperv_relaxed_timing, false), DEFINE_PROP_BOOL("hv-vapic", X86CPU, hyperv_vapic, false), DEFINE_PROP_BOOL("hv-time", X86CPU, hyperv_time, false), - DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, false), + DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true), DEFINE_PROP_BOOL("enforce", X86CPU, enforce_cpuid, false), DEFINE_PROP_BOOL("kvm", X86CPU, expose_kvm, true), DEFINE_PROP_UINT32("level", X86CPU, env.cpuid_level, 0), -- 2.1.0

3 4

[libvirt] [PATCH RFC 0/2] libxl: migration v3 and p2p support
by Joao Martins 09 Sep '15

09 Sep '15

Hey Jim, These series introduce support for peer2peer and plain v3 migration without extensible parameters. This patch series is organized as follows: The first patch introduces the P2P where the source host must take care of connection handling with destination libvirtd and is responsible for performing the migration. I introduced an additional method for the P2P Perform phase since the default Perform of libxl is then reused. The sequence is the same as v3 so it is possible to migrate to non-P2P host. libxlDoMigrateP2P actually does the migration using a virConnectPtr to the destination and thus The second patch is mostly mechanical, and functional wise is very much the same as the Params variant. The added value of introducing plain v3 is having access to older APIs such virDomainMigrateToURI2 and virDomainMigrate (in P2P mode) which are currently in use by Openstack/virt-manager. My testing was performed in xen stable-4.5 and 4.4.3 (with the added patches in the Xen CI loop) plus Openstack Kilo. In addition I backported the patches and tested also on libvirt 1.2.14 (whic is also ran by the Xen CI Openstack loop). Any comments or suggestions are welcome, Thanks! Joao Joao Martins (2): libxl: add p2p migration libxl: add v3 migration w/o params src/libxl/libxl_driver.c | 226 +++++++++++++++++++++++++++++++++++++++++++- src/libxl/libxl_migration.c | 220 ++++++++++++++++++++++++++++++++++++++++++ src/libxl/libxl_migration.h | 11 +++ 3 files changed, 454 insertions(+), 3 deletions(-) -- 2.1.4

2 6