- Devel - libvirt List Archives

[libvirt] [libvirt-guests] New feature ALWAYS_START
by Marek Lukács 23 Sep '15

23 Sep '15

Hi, It will be nice feature to have configuration option ALWAYS_START="$uri:$name $uri:name ..." in libvirt-guests configuration file. If ON_BOOT is "start" and if ALWAYS_START is not empty, it iterates over the ALWAYS_START and starts guests with same conditions (delays etc.) before it starts guests from LISTFILE. Benefits: - guests are started with delays - guests are started after host failure - guests are started in specific order (for example complex environment, when DB should be started before other guest, etc.) Regards, Marek Lukács

3 7

[libvirt] [PATCH v2] virsh: Fix job status indicator for 0 length block jobs
by Peter Krempa 23 Sep '15

23 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. Since the only place that actually uses the condition that suppresses the output is in migration, let's move the check there and thus add support for 0 of 0 equaling to 100%. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1196711 --- tools/virsh-domain.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 73c476d..fb138d5 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -1700,10 +1700,6 @@ virshPrintJobProgress(const char *label, unsigned long long remaining, { int progress; - if (total == 0) - /* migration has not been started */ - return; - if (remaining == 0) { /* migration has completed */ progress = 100; @@ -4401,7 +4397,7 @@ virshWatchJob(vshControl *ctl, ret = virDomainGetJobInfo(dom, &jobinfo); pthread_sigmask(SIG_SETMASK, &oldsigmask, NULL); if (ret == 0) { - if (verbose) + if (verbose && jobinfo.dataTotal > 0) virshPrintJobProgress(label, jobinfo.dataRemaining, jobinfo.dataTotal); -- 2.4.5

2 2

[libvirt] [PATCH v2] qemu: Refresh memory size only on fresh starts
by Peter Krempa 23 Sep '15

23 Sep '15

Qemu unfortunately doesn't update internal state right after migration and so the actual balloon size as returned by 'query-balloon' are invalid for a while after the CPUs are started after migration. If we'd refresh our internal state at this point we would report invalid current memory size until the next balloon event would arrive. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1242940 --- Version 2: - completely different from original posting due to unexpected qemu behavior src/qemu/qemu_process.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 7187dc1..b961f40 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -5023,7 +5023,8 @@ int qemuProcessStart(virConnectPtr conn, /* Since CPUs were not started yet, the balloon could not return the memory * to the host and thus cur_balloon needs to be updated so that GetXMLdesc * and friends return the correct size in case they can't grab the job */ - if (qemuProcessRefreshBalloonState(driver, vm, asyncJob) < 0) + if (!migrateFrom && !snapshot && + qemuProcessRefreshBalloonState(driver, vm, asyncJob) < 0) goto cleanup; VIR_DEBUG("Detecting actual memory size for video device"); -- 2.4.5

2 2

[libvirt] [PATCH] qemu: Align memory module sizes to 2MiB
by Peter Krempa 23 Sep '15

23 Sep '15

My original implementation was based on a qemu version that still did not have all the checks in place. Using sizes that would align to odd megabyte increments will produce the following error: qemu-kvm: -device pc-dimm,node=0,memdev=memdimm0,id=dimm0: backend memory size must be multiple of 0x200000 qemu-kvm: -device pc-dimm,node=0,memdev=memdimm0,id=dimm0: Device 'pc-dimm' could not be initialized Introduce an alignment retrieval function for memory devices and use it to align the devices separately and modify a test case to verify it. --- src/qemu/qemu_domain.c | 19 ++++++++++++++++++- .../qemuxml2argv-memory-hotplug-dimm.xml | 2 +- 2 files changed, 19 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 7d92f3a..7680c87 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -3376,6 +3376,21 @@ qemuDomainGetMemorySizeAlignment(virDomainDefPtr def) } +static unsigned long long +qemuDomainGetMemoryModuleSizeAlignment(const virDomainDef *def, + const virDomainMemoryDef *mem ATTRIBUTE_UNUSED) +{ + /* PPC requires the memory sizes to be rounded to 256MiB increments, so + * round them to the size always. */ + if (ARCH_IS_PPC64(def->os.arch)) + return 256 * 1024; + + /* dimm memory modules require 2MiB alignment rather than the 1MiB we are + * using elsewhere. */ + return 2048; +} + + int qemuDomainAlignMemorySizes(virDomainDefPtr def) { @@ -3402,8 +3417,10 @@ qemuDomainAlignMemorySizes(virDomainDefPtr def) def->mem.max_memory = VIR_ROUND_UP(def->mem.max_memory, align); /* Align memory module sizes */ - for (i = 0; i < def->nmems; i++) + for (i = 0; i < def->nmems; i++) { + align = qemuDomainGetMemoryModuleSizeAlignment(def, def->mems[i]); def->mems[i]->size = VIR_ROUND_UP(def->mems[i]->size, align); + } return 0; } diff --git a/tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-dimm.xml b/tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-dimm.xml index 3f468ec..fbcac84 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-dimm.xml +++ b/tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-dimm.xml @@ -36,7 +36,7 @@ <memballoon model='virtio'/> <memory model='dimm'> <target> - <size unit='KiB'>524287</size> + <size unit='KiB'>523264</size> <node>0</node> </target> </memory> -- 2.4.5

2 1

[libvirt] [PATCH 0/3] Improve keepalive timeout handling in clients
by Jiri Denemark 23 Sep '15

23 Sep '15

Jiri Denemark (3): client rpc: Report proper error for keepalive disconnections client rpc: Process pending data on error virsh: Notify users about disconnects src/rpc/virkeepalive.c | 10 +++++----- src/rpc/virnetclient.c | 30 +++++++++++++++++++++++++----- tools/virsh.c | 35 ++++++++++++++++++++++++++++++++--- 3 files changed, 62 insertions(+), 13 deletions(-) -- 2.5.2

2 6

[libvirt] [sandbox 00/11] Adding a virt-builder source
by Cédric Bosdonnat 23 Sep '15

23 Sep '15

Hi all, Here are a few patches to add a virt-builder source to virt-sandbox-image. There are also patches fixing a few issues here and there. Cédric Bosdonnat (11): Fix memory leak virt-sandbox-image: fix error string formatting. Make virt-sandbox-image executable virt-sandbox-image: remove undefined default_disk_format virt-sandbox-image: add a source post_run hook virt-sandbox-image: move DockerSource _format_disk to Source virt-sandbox-image: smarter source name computing virt-sandbox-image: add a virt-builder source virt-sandbox-image: automatically call download and create if needed virt-sandbox-image: fix exception catching virt-sandbox-image: add error handling for sources import bin/virt-sandbox-image | 0 libvirt-sandbox.spec.in | 1 + libvirt-sandbox/image/cli.py | 154 +++++++++++---------- libvirt-sandbox/image/sources/DockerSource.py | 32 +++-- libvirt-sandbox/image/sources/Makefile.am | 1 + libvirt-sandbox/image/sources/Source.py | 50 +++++++ libvirt-sandbox/image/sources/VirtBuilderSource.py | 143 +++++++++++++++++++ libvirt-sandbox/image/template.py | 24 +++- libvirt-sandbox/libvirt-sandbox-builder.c | 1 + 9 files changed, 311 insertions(+), 95 deletions(-) mode change 100644 => 100755 bin/virt-sandbox-image create mode 100644 libvirt-sandbox/image/sources/VirtBuilderSource.py -- 2.1.4

2 22

[libvirt] [PATCH] conf: escape string for disk driver name attribute
by Luyao Huang 23 Sep '15

23 Sep '15

Just like e92e5ba1, this attribute was missed. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/conf/domain_conf.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index a3b3ccb..85c2d0d 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -18867,8 +18867,7 @@ virDomainDiskDefFormat(virBufferPtr buf, def->ioeventfd || def->event_idx || def->copy_on_read || def->discard || def->iothread) { virBufferAddLit(buf, "<driver"); - if (def->src->driverName) - virBufferAsprintf(buf, " name='%s'", def->src->driverName); + virBufferEscapeString(buf, " name='%s'", def->src->driverName); if (def->src->format > 0) virBufferAsprintf(buf, " type='%s'", virStorageFileFormatTypeToString(def->src->format)); -- 1.8.3.1

3 2

[libvirt] [PATCH] Use VIR_DIV_UP macro where possible
by Martin Kletzander 23 Sep '15

23 Sep '15

Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- Will push as trivial in a second. src/rpc/virnetclientstream.c | 2 +- src/util/virbitmap.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/rpc/virnetclientstream.c b/src/rpc/virnetclientstream.c index 1cc9002b0ad7..64e9cd221be2 100644 --- a/src/rpc/virnetclientstream.c +++ b/src/rpc/virnetclientstream.c @@ -279,7 +279,7 @@ int virNetClientStreamQueuePacket(virNetClientStreamPtr st, goto end; } - pieces = (length + size - 1) / size; + pieces = VIR_DIV_UP(length, size); for (piece = 0; piece < pieces; piece++) { if (size > length - offset) size = length - offset; diff --git a/src/util/virbitmap.c b/src/util/virbitmap.c index 4d270a910c19..57135a09f71e 100644 --- a/src/util/virbitmap.c +++ b/src/util/virbitmap.c @@ -71,7 +71,7 @@ virBitmapNewQuiet(size_t size) if (SIZE_MAX - VIR_BITMAP_BITS_PER_UNIT < size || size == 0) return NULL; - sz = (size + VIR_BITMAP_BITS_PER_UNIT - 1) / VIR_BITMAP_BITS_PER_UNIT; + sz = VIR_DIV_UP(size, VIR_BITMAP_BITS_PER_UNIT); if (VIR_ALLOC_QUIET(bitmap) < 0) return NULL; -- 2.5.3

1 0

[libvirt] [PATCH 0/5] Couple of 'virsh create' fixes
by Michal Privoznik 23 Sep '15

23 Sep '15

*** BLURB HERE *** Michal Privoznik (5): qemuDomainCreateXML: Don't remove persistent domains on error qemuDomainCreateXML: Make domain definition transient qemu: Move vm->persistent check into qemuDomainRemoveInactive lxcDomainCreateXMLWithFiles: Don't remove persistent domains on error lxcDomainCreateXMLWithFiles: Make domain definition transient src/lxc/lxc_driver.c | 7 +++++-- src/qemu/qemu_domain.c | 9 ++++++++- src/qemu/qemu_driver.c | 37 +++++++++++++++---------------------- src/qemu/qemu_migration.c | 14 +++++--------- src/qemu/qemu_process.c | 12 ++++-------- 5 files changed, 37 insertions(+), 42 deletions(-) -- 2.4.9

2 7

[libvirt] [PATCH] qemuDomainDeviceDefPostParse: Drop useless check
by Michal Privoznik 23 Sep '15

23 Sep '15

Now that virQEMUDriverCreateXMLConf is never called with NULL (after 086f37e97aab) we can safely drop useless check in qemuDomainDeviceDefPostParse as we are guaranteed to be always called with the driver initialized. Therefore checking if driver is NULL makes no sense. Moreover, if we mix it with direct driver dereference. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/qemu/qemu_domain.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 7d92f3a..8e3a7d3 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -1233,12 +1233,9 @@ qemuDomainDeviceDefPostParse(virDomainDeviceDefPtr dev, { virQEMUDriverPtr driver = opaque; virQEMUCapsPtr qemuCaps = NULL; - virQEMUDriverConfigPtr cfg = NULL; + virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); int ret = -1; - if (driver) - cfg = virQEMUDriverGetConfig(driver); - qemuCaps = virQEMUCapsCacheLookup(driver->qemuCapsCache, def->emulator); if (dev->type == VIR_DOMAIN_DEVICE_NET && -- 2.4.9

2 1