- Devel - libvirt List Archives

[libvirt] Question about migration max downtime in domXML
by Jim Fehlig 21 Jan '15

21 Jan '15

Hi All, I've had a few user requests to add migration max downtime to the domain XML. IIRC, this was discussed on the list quite some time ago, but I'm unable to find it in the archives. My recollection is that max downtime is considered a tunable vs config and as such not included in the domXML. But domXML does contain quite a few tunables, so perhaps my memory is incorrect. Does anyone recall the details? I'd be happy to work on this if folks find it acceptable. Regards, Jim

2 2

[libvirt] [PATCH] qemu: Propagate storage errors on migration
by Michal Privoznik 20 Jan '15

20 Jan '15

There's one bug that got more visible with my patches that automatically precreate storage on migration. The problem is, if there's not enough disk space on destination, we successfully finish the migration instead of failing. Firstly, we instruct qemu to copy the storage. And as it copies the data, one write() will return -ENOSPC, eventually, to which qemu reacts by emitting BLOCK_JOB_ERROR. The event is successfully ignored by us. Then, since the block job has finished, BLOCK_JOB_COMPLETED event is emitted too. Secondly, we are not checking for the block job completion as we should. Currently, we do the check by issuing 'query-block-jobs' and then looking in the command's output for the job we are interested in. If course, we don't fail if the job is not there, in which case the number of total bytes to be transferred and bytes already transferred, well they equal both to zero. This is actually the line causing the bug as it sees both numbers equal to each other and gives green light to the actual migration. The fix consist of two parts: In the first, code handling BLOCK_JOB_ERROR is added. It's a monitor event and we should have handler for that anyway. In the second part, the code doing drive mirroring is rewritten so it waits for the events instead of querying on the monitor repeatedly. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/conf/domain_conf.c | 3 ++- src/conf/domain_conf.h | 1 + src/qemu/qemu_migration.c | 25 +++++++------------------ src/qemu/qemu_monitor_json.c | 10 ++++++++++ src/qemu/qemu_process.c | 14 ++++++++++++++ 5 files changed, 34 insertions(+), 19 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index d1a483a..751a9b5 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -787,7 +787,8 @@ VIR_ENUM_IMPL(virDomainDiskMirrorState, VIR_DOMAIN_DISK_MIRROR_STATE_LAST, "none", "yes", "abort", - "pivot") + "pivot", + "error") VIR_ENUM_IMPL(virDomainLoader, VIR_DOMAIN_LOADER_TYPE_LAST, diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index ac1f4f8..f18fa80 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -654,6 +654,7 @@ typedef enum { VIR_DOMAIN_DISK_MIRROR_STATE_READY, /* Job in second phase */ VIR_DOMAIN_DISK_MIRROR_STATE_ABORT, /* Job aborted, waiting for event */ VIR_DOMAIN_DISK_MIRROR_STATE_PIVOT, /* Job pivoted, waiting for event */ + VIR_DOMAIN_DISK_MIRROR_STATE_ERROR, /* Job failed */ VIR_DOMAIN_DISK_MIRROR_STATE_LAST } virDomainDiskMirrorState; diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 77e0b35..e2309ea 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -1743,7 +1743,6 @@ qemuMigrationDriveMirror(virQEMUDriverPtr driver, for (i = 0; i < vm->def->ndisks; i++) { virDomainDiskDefPtr disk = vm->def->disks[i]; - virDomainBlockJobInfo info; /* skip shared, RO and source-less disks */ if (disk->src->shared || disk->src->readonly || @@ -1768,6 +1767,7 @@ qemuMigrationDriveMirror(virQEMUDriverPtr driver, if (mon_ret < 0) goto error; + disk->mirrorJob = VIR_DOMAIN_BLOCK_JOB_TYPE_COPY; lastGood = i; /* wait for completion */ @@ -1775,38 +1775,27 @@ qemuMigrationDriveMirror(virQEMUDriverPtr driver, /* Poll every 500ms for progress & to allow cancellation */ struct timespec ts = { .tv_sec = 0, .tv_nsec = 500 * 1000 * 1000ull }; - memset(&info, 0, sizeof(info)); - - if (qemuDomainObjEnterMonitorAsync(driver, vm, - QEMU_ASYNC_JOB_MIGRATION_OUT) < 0) - goto error; if (priv->job.asyncAbort) { /* explicitly do this *after* we entered the monitor, * as this is a critical section so we are guaranteed * priv->job.asyncAbort will not change */ - qemuDomainObjExitMonitor(driver, vm); priv->job.current->type = VIR_DOMAIN_JOB_CANCELLED; virReportError(VIR_ERR_OPERATION_ABORTED, _("%s: %s"), qemuDomainAsyncJobTypeToString(priv->job.asyncJob), _("canceled by client")); goto error; } - mon_ret = qemuMonitorBlockJobInfo(priv->mon, diskAlias, &info, - NULL); - qemuDomainObjExitMonitor(driver, vm); - if (mon_ret < 0) - goto error; - - if (info.cur == info.end) { + if (disk->mirrorState == VIR_DOMAIN_DISK_MIRROR_STATE_READY) { VIR_DEBUG("Drive mirroring of '%s' completed", diskAlias); break; + } else if (disk->mirrorState == VIR_DOMAIN_DISK_MIRROR_STATE_ERROR) { + virReportError(VIR_ERR_OPERATION_ABORTED, _("%s: %s"), + qemuDomainAsyncJobTypeToString(priv->job.asyncJob), + _("canceled by destination")); + goto error; } - /* XXX Frankly speaking, we should listen to the events, - * instead of doing this. But this works for now and we - * are doing something similar in migration itself anyway */ - virObjectUnlock(vm); nanosleep(&ts, NULL); diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index e567aa7..5d6bbca 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -76,6 +76,7 @@ static void qemuMonitorJSONHandlePMWakeup(qemuMonitorPtr mon, virJSONValuePtr da static void qemuMonitorJSONHandlePMSuspend(qemuMonitorPtr mon, virJSONValuePtr data); static void qemuMonitorJSONHandleBlockJobCompleted(qemuMonitorPtr mon, virJSONValuePtr data); static void qemuMonitorJSONHandleBlockJobCanceled(qemuMonitorPtr mon, virJSONValuePtr data); +static void qemuMonitorJSONHandleBlockJobError(qemuMonitorPtr mon, virJSONValuePtr data); static void qemuMonitorJSONHandleBlockJobReady(qemuMonitorPtr mon, virJSONValuePtr data); static void qemuMonitorJSONHandleBalloonChange(qemuMonitorPtr mon, virJSONValuePtr data); static void qemuMonitorJSONHandlePMSuspendDisk(qemuMonitorPtr mon, virJSONValuePtr data); @@ -94,6 +95,7 @@ static qemuEventHandler eventHandlers[] = { { "BLOCK_IO_ERROR", qemuMonitorJSONHandleIOError, }, { "BLOCK_JOB_CANCELLED", qemuMonitorJSONHandleBlockJobCanceled, }, { "BLOCK_JOB_COMPLETED", qemuMonitorJSONHandleBlockJobCompleted, }, + { "BLOCK_JOB_ERROR", qemuMonitorJSONHandleBlockJobError, }, { "BLOCK_JOB_READY", qemuMonitorJSONHandleBlockJobReady, }, { "DEVICE_DELETED", qemuMonitorJSONHandleDeviceDeleted, }, { "DEVICE_TRAY_MOVED", qemuMonitorJSONHandleTrayChange, }, @@ -842,6 +844,14 @@ qemuMonitorJSONHandleBlockJobCanceled(qemuMonitorPtr mon, } static void +qemuMonitorJSONHandleBlockJobError(qemuMonitorPtr mon, + virJSONValuePtr data) +{ + qemuMonitorJSONHandleBlockJobImpl(mon, data, + VIR_DOMAIN_BLOCK_JOB_FAILED); +} + +static void qemuMonitorJSONHandleBlockJobReady(qemuMonitorPtr mon, virJSONValuePtr data) { diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index c18204b..b29ecf1 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -1110,6 +1110,20 @@ qemuProcessHandleBlockJob(qemuMonitorPtr mon ATTRIBUTE_UNUSED, disk->mirrorJob = VIR_DOMAIN_BLOCK_JOB_TYPE_UNKNOWN; save = true; } + } else if (disk->mirrorJob == VIR_DOMAIN_BLOCK_JOB_TYPE_COPY) { + switch ((virConnectDomainEventBlockJobStatus) status) { + case VIR_DOMAIN_BLOCK_JOB_READY: + case VIR_DOMAIN_BLOCK_JOB_COMPLETED: + disk->mirrorState = VIR_DOMAIN_DISK_MIRROR_STATE_READY; + break; + case VIR_DOMAIN_BLOCK_JOB_CANCELED: + case VIR_DOMAIN_BLOCK_JOB_FAILED: + disk->mirrorState = VIR_DOMAIN_DISK_MIRROR_STATE_ERROR; + break; + case VIR_DOMAIN_BLOCK_JOB_LAST: + VIR_DEBUG("should not get here"); + break; + } } } -- 2.0.5

2 2

[libvirt] [PATCH] Fix virCgroupNewMachine prototype on non-Linux
by Ján Tomko 20 Jan '15

20 Jan '15

Commit 318df5a changed the prototype of virCgroupNewMachine without adjusting the stub function for platforms without cgroups. --- Pushed as a trivial build fix. src/util/vircgroup.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index f5f617e..fe34290 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -4073,6 +4073,8 @@ virCgroupNewMachine(const char *name ATTRIBUTE_UNUSED, const char *rootdir ATTRIBUTE_UNUSED, pid_t pidleader ATTRIBUTE_UNUSED, bool isContainer ATTRIBUTE_UNUSED, + size_t nnicindexes ATTRIBUTE_UNUSED, + int *nicindexes ATTRIBUTE_UNUSED, const char *partition ATTRIBUTE_UNUSED, int controllers ATTRIBUTE_UNUSED, virCgroupPtr *group ATTRIBUTE_UNUSED) -- 2.0.4

1 0

[libvirt] glibc detected invalid pointer
by shaowt 20 Jan '15

20 Jan '15

hi all: when i use vnc to connect VM, sometimes in libvirt logs, i found glibc detected invalid pointer error,and then this vm will be shutdown . “The information (including any attachments) contained in this e-mail is confidential and may be legally privileged . The message is intended solely for the addressee(s). If this e-mail is not originally intended for you, or received by you in error, please contact the sender by return e-mail and do not disclose its content to anyone and destroy all copies of the original message immediately. If you are not the intended recipient, you are hereby notified that any use, dissemination, or reproduction is strictly prohibited and may be unlawful. 本邮件（包括任何附件）包含机密信息，并受法律保护。此信息意在使收件人明确以下内容：如果您并非该收件人或错收此邮件，请联系发件人并退还此邮件，请不要向任何人泄露邮件内容，请立即删除所有邮件及其附件。严禁任何使用，传播或复制此邮件的行为，这些行为可能违反法律。“

2 3

[libvirt] [PATCH v2] network: Let domains be restricted to local DNS
by Josh Stone 20 Jan '15

20 Jan '15

This adds a new "localOnly" attribute on the domain element of the network xml. With this set to "yes", DNS requests under that domain will only be resolved by libvirt's dnsmasq, never forwarded upstream. This was how it worked before commit f69a6b987d616, and I found that functionality useful. For example, I have my host's NetworkManager dnsmasq configured to forward that domain to libvirt's dnsmasq, so I can easily resolve guest names from outside. But if libvirt's dnsmasq doesn't know a name and forwards it to the host, I'd get an endless forwarding loop. Now I can set localOnly="yes" to prevent the loop. Signed-off-by: Josh Stone <jistone(a)redhat.com> Cc: Laine Stump <laine(a)laine.org> --- docs/formatnetwork.html.in | 12 +++++++- docs/schemas/network.rng | 3 ++ src/conf/network_conf.c | 32 ++++++++++++++++++++-- src/conf/network_conf.h | 1 + src/network/bridge_driver.c | 5 ++++ .../nat-network-dns-local-domain.conf | 14 ++++++++++ .../nat-network-dns-local-domain.xml | 9 ++++++ tests/networkxml2conftest.c | 1 + 8 files changed, 74 insertions(+), 3 deletions(-) create mode 100644 tests/networkxml2confdata/nat-network-dns-local-domain.conf create mode 100644 tests/networkxml2confdata/nat-network-dns-local-domain.xml diff --git a/docs/formatnetwork.html.in b/docs/formatnetwork.html.in index dc438aee8622..defcdba00930 100644 --- a/docs/formatnetwork.html.in +++ b/docs/formatnetwork.html.in @@ -82,7 +82,7 @@ <pre> ... <bridge name="virbr0" stp="on" delay="5"/> - <domain name="example.com"/> + <domain name="example.com" localOnly="no"/> <forward mode="nat" dev="eth0"/> ...</pre> @@ -113,6 +113,16 @@ a <code><forward></code> mode of "nat" or "route" (or an isolated network with no <code><forward></code> element). Since 0.4.5 + + + If the optional <code>localOnly</code> attribute on the + <code>domain</code> element is "yes", then DNS requests under + this domain will only be resolved by the virtual network's own + DNS server - they will not be forwarded to the host's upstream + DNS server. If <code>localOnly</code> is "no", and by + default, unresolved requests will be forwarded. + Since 1.2.11 + </dd> <dt><code>forward</code></dt> <dd>Inclusion of the <code>forward</code> element indicates that diff --git a/docs/schemas/network.rng b/docs/schemas/network.rng index 4546f8037580..a1da28092375 100644 --- a/docs/schemas/network.rng +++ b/docs/schemas/network.rng @@ -225,6 +225,9 @@ <optional> <element name="domain"> <attribute name="name"><ref name="dnsName"/></attribute> + <optional> + <attribute name="localOnly"><ref name="virYesNo"/></attribute> + </optional> </element> </optional> diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index 97719ed536de..31b765ff2c51 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -2083,6 +2083,18 @@ virNetworkDefParseXML(xmlXPathContextPtr ctxt) /* Parse network domain information */ def->domain = virXPathString("string(./domain[1]/@name)", ctxt); + tmp = virXPathString("string(./domain[1]/@localOnly)", ctxt); + if (tmp) { + def->domain_local = virTristateBoolTypeFromString(tmp); + if (def->domain_local <= 0) { + virReportError(VIR_ERR_XML_ERROR, + _("Invalid domain localOnly setting '%s' " + "in network '%s'"), + tmp, def->name); + goto error; + } + VIR_FREE(tmp); + } if ((bandwidthNode = virXPathNode("./bandwidth", ctxt)) != NULL && (def->bandwidth = virNetDevBandwidthParse(bandwidthNode, -1)) == NULL) @@ -2805,8 +2817,24 @@ virNetworkDefFormatBuf(virBufferPtr buf, virBufferAsprintf(buf, "<mac address='%s'/>\n", macaddr); } - if (def->domain) - virBufferAsprintf(buf, "<domain name='%s'/>\n", def->domain); + if (def->domain) { + virBufferAsprintf(buf, "<domain name='%s'", def->domain); + + /* default to "no", but don't format it in the XML */ + if (def->domain_local) { + const char *local = virTristateBoolTypeToString(def->domain_local); + + if (!local) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unknown localOnly type %d in network"), + def->domain_local); + return -1; + } + virBufferAsprintf(buf, " localOnly='%s'", local); + } + + virBufferAddLit(buf, "/>\n"); + } if (virNetworkDNSDefFormat(buf, &def->dns) < 0) goto error; diff --git a/src/conf/network_conf.h b/src/conf/network_conf.h index 660cd2d10cd1..bb9724fddbb8 100644 --- a/src/conf/network_conf.h +++ b/src/conf/network_conf.h @@ -232,6 +232,7 @@ struct _virNetworkDef { char *bridge; /* Name of bridge device */ char *domain; + int domain_local; /* enum virTristateBool: yes disables dns forwarding */ unsigned long delay; /* Bridge forward delay (ms) */ bool stp; /* Spanning tree protocol */ virMacAddr mac; /* mac address of bridge device */ diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c index 9ccc9f8f8de6..f4158517c573 100644 --- a/src/network/bridge_driver.c +++ b/src/network/bridge_driver.c @@ -928,6 +928,11 @@ networkDnsmasqConfContents(virNetworkObjPtr network, } if (network->def->domain) { + if (network->def->domain_local == VIR_TRISTATE_BOOL_YES) { + virBufferAsprintf(&configbuf, + "local=/%s/\n", + network->def->domain); + } virBufferAsprintf(&configbuf, "domain=%s\n" "expand-hosts\n", diff --git a/tests/networkxml2confdata/nat-network-dns-local-domain.conf b/tests/networkxml2confdata/nat-network-dns-local-domain.conf new file mode 100644 index 000000000000..5f41b9186cbc --- /dev/null +++ b/tests/networkxml2confdata/nat-network-dns-local-domain.conf @@ -0,0 +1,14 @@ +##WARNING: THIS IS AN AUTO-GENERATED FILE. CHANGES TO IT ARE LIKELY TO BE +##OVERWRITTEN AND LOST. Changes to this configuration should be made using: +## virsh net-edit default +## or other application using the libvirt API. +## +## dnsmasq conf file created by libvirt +strict-order +local=/example.com/ +domain=example.com +expand-hosts +except-interface=lo +bind-dynamic +interface=virbr0 +addn-hosts=/var/lib/libvirt/dnsmasq/default.addnhosts diff --git a/tests/networkxml2confdata/nat-network-dns-local-domain.xml b/tests/networkxml2confdata/nat-network-dns-local-domain.xml new file mode 100644 index 000000000000..a92d71f1f2f6 --- /dev/null +++ b/tests/networkxml2confdata/nat-network-dns-local-domain.xml @@ -0,0 +1,9 @@ +<network> + <name>default</name> + <uuid>81ff0d90-c91e-6742-64da-4a736edb9a9c</uuid> + <forward dev='eth0' mode='nat'/> + <bridge name='virbr0' stp='on' delay='0' /> + <domain name='example.com' localOnly='yes'/> + <ip address='192.168.122.1' netmask='255.255.255.0'> + </ip> +</network> diff --git a/tests/networkxml2conftest.c b/tests/networkxml2conftest.c index 267513f6372c..280db306b6d6 100644 --- a/tests/networkxml2conftest.c +++ b/tests/networkxml2conftest.c @@ -134,6 +134,7 @@ mymain(void) DO_TEST("nat-network-dns-hosts", full); DO_TEST("nat-network-dns-forward-plain", full); DO_TEST("nat-network-dns-forwarders", full); + DO_TEST("nat-network-dns-local-domain", full); DO_TEST("dhcp6-network", dhcpv6); DO_TEST("dhcp6-nat-network", dhcpv6); DO_TEST("dhcp6host-routed-network", dhcpv6); -- 2.1.0

2 1

[libvirt] [java] [PATCH] JNA: add missing flags parameter to wrapped functions
by Claudio Bley 20 Jan '15

20 Jan '15

Some functions of the Libvirt interface were missing a required flags parameter as defined by the libvirt C API, namely: * virInterfaceCreate * virInterfaceDestroy * virNodeDeviceGetXMLDesc Calling these functions without the flags parameter would result in undefined behaviour, at best in an "unsupported flags" error. --- Reported by Matt Russell via personal email. Thanks, Matt! Pushing under the trivial rule. src/main/java/org/libvirt/Device.java | 2 +- src/main/java/org/libvirt/Interface.java | 4 ++-- src/main/java/org/libvirt/jna/Libvirt.java | 6 +++--- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/src/main/java/org/libvirt/Device.java b/src/main/java/org/libvirt/Device.java index a3b20b8..3c876c6 100644 --- a/src/main/java/org/libvirt/Device.java +++ b/src/main/java/org/libvirt/Device.java @@ -118,7 +118,7 @@ public class Device { * @throws LibvirtException */ public String getXMLDescription() throws LibvirtException { - return processError(libvirt.virNodeDeviceGetXMLDesc(VDP)); + return processError(libvirt.virNodeDeviceGetXMLDesc(VDP, 0)); } /** diff --git a/src/main/java/org/libvirt/Interface.java b/src/main/java/org/libvirt/Interface.java index e0ddc3f..4c9670c 100644 --- a/src/main/java/org/libvirt/Interface.java +++ b/src/main/java/org/libvirt/Interface.java @@ -52,7 +52,7 @@ public class Interface { * @throws LibvirtException */ public int create() throws LibvirtException { - return processError(libvirt.virInterfaceCreate(VIP)); + return processError(libvirt.virInterfaceCreate(VIP, 0)); } /** @@ -71,7 +71,7 @@ public class Interface { * @throws LibvirtException */ public int destroy() throws LibvirtException { - return processError(libvirt.virInterfaceDestroy(VIP)); + return processError(libvirt.virInterfaceDestroy(VIP, 0)); } @Override diff --git a/src/main/java/org/libvirt/jna/Libvirt.java b/src/main/java/org/libvirt/jna/Libvirt.java index 4d76071..a17bfc7 100644 --- a/src/main/java/org/libvirt/jna/Libvirt.java +++ b/src/main/java/org/libvirt/jna/Libvirt.java @@ -350,7 +350,7 @@ public interface Libvirt extends Library { String virNodeDeviceGetParent(DevicePointer virDevicePointer); int virNodeDeviceNumOfCaps(DevicePointer virDevicePointer); int virNodeDeviceListCaps(DevicePointer virDevicePointer, Pointer[] names, int maxNames); - String virNodeDeviceGetXMLDesc(DevicePointer virDevicePointer); + String virNodeDeviceGetXMLDesc(DevicePointer virDevicePointer, int flags); int virNodeDeviceFree(DevicePointer virDevicePointer); int virNodeDeviceDettach(DevicePointer virDevicePointer); int virNodeDeviceReAttach(DevicePointer virDevicePointer); @@ -402,9 +402,9 @@ public interface Libvirt extends Library { int virStorageVolResize(StorageVolPointer storageVolPtr, long capacity, int flags); // Interface Methods - int virInterfaceCreate(InterfacePointer virDevicePointer); + int virInterfaceCreate(InterfacePointer virDevicePointer, int flags); InterfacePointer virInterfaceDefineXML(ConnectionPointer virConnectPtr, String xml, int flags); - int virInterfaceDestroy(InterfacePointer virDevicePointer); + int virInterfaceDestroy(InterfacePointer virDevicePointer, int flags); int virInterfaceFree(InterfacePointer virDevicePointer); String virInterfaceGetName(InterfacePointer virInterfacePtr); String virInterfaceGetMACString(InterfacePointer virInterfacePtr); -- 2.2.2

1 0

[libvirt] libvirt 1.2.10 and latest EL6 qemu-kvm
by liedekef＠telenet.be 20 Jan '15

20 Jan '15

Hi all, I had libvirt 1.2.10 running without issues together with the qemu package qemu-kvm-0.12.1.2-2.415.el6_5.14.x86_64.rpm Now, after updating my server, qemu-kvm became version qemu-kvm-0.12.1.2-2.448.el6_6.x86_64 And as a result, when doing 'virsh start' (virsh define worked ok), the following error occured: 2015-01-19 15:05:26.539+0000: 4347: error : qemuMonitorJSONCheckError:379 : internal error: unable to execute QEMU command 'qom-list': The command qom-list has not been The only thing I could do (even downgrading libvirt to 1.2.9 didn't work), was to downgrade qemu-kvm to the older version again ... Any hints on this? Franky

2 1

[libvirt] [python PATCH] sanitytest: fix mapping test for virDomainDefineXMLFlags
by Pavel Hrdina 19 Jan '15

19 Jan '15

The libvirt API has in the name "virDomain" but it's correctly mapped into "virConnect" class. Create an exception in the sanity test. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- Pushed as build(test) breaker. sanitytest.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sanitytest.py b/sanitytest.py index 9907f3d..f021e5a 100644 --- a/sanitytest.py +++ b/sanitytest.py @@ -230,7 +230,7 @@ for name in sorted(basicklassmap): "LookupByID", "LookupByName", "LookupByKey", "LookupByPath", "LookupByMACString", "LookupByUsage", "LookupByVolume", "LookupSCSIHostByWWN", "Restore", "RestoreFlags", - "SaveImageDefineXML", "SaveImageGetXMLDesc"]: + "SaveImageDefineXML", "SaveImageGetXMLDesc", "DefineXMLFlags"]: if klass != "virDomain": func = klass[3:] + func -- 2.0.5

1 0

[libvirt] [PATCH v2] Make ZFS storage pool XML tests optional
by Gary R Hook 19 Jan '15

19 Jan '15

From: Gary R Hook <grhookatwork(a)gmail.com> Do not run ZFS tests when ZFS is unsupported in the environment. The recent patch b4af40226d09adeaf9e33a1d6594c4e8ce7f771d adds tests to storagepoolxml2xmltest for the optional ZFS feature. When ZFS is not included in the configuration these tests should not / cannot be run. Modify the test source file to check for use of the feature and compile in the tests accordingly. Signed-off-by: Gary R Hook <gary.hook(a)nimboxx.com> --- tests/storagepoolxml2xmltest.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/storagepoolxml2xmltest.c b/tests/storagepoolxml2xmltest.c index 52e2193..270f75d 100644 --- a/tests/storagepoolxml2xmltest.c +++ b/tests/storagepoolxml2xmltest.c @@ -106,8 +106,10 @@ mymain(void) DO_TEST("pool-gluster"); DO_TEST("pool-gluster-sub"); DO_TEST("pool-scsi-type-scsi-host-stable"); +#ifdef WITH_STORAGE_ZFS DO_TEST("pool-zfs"); DO_TEST("pool-zfs-sourcedev"); +#endif return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE; } -- 1.9.1

2 1

[libvirt] [PATCH] systemd: fix build without dbus
by Daniel P. Berrange 19 Jan '15

19 Jan '15

The virDBusMethodCall method has a DBusError as one of its parameters. If the caller wants to pass a non-NULL value for this, it immediately makes the calling code require DBus at build time. This has led to breakage of non-DBus builds several times. It is desirable that only the virdbus.c file should need WITH_DBUS conditionals, so we must ideally remove the DBusError parameter from the method. We can't simply raise a libvirt error, since the whole point of this parameter is to give the callers a way to check if the error is one they want to ignore, without having the logs polluted with an error message. So, we add a virErrorPtr parameter which the caller can then either ignore or raise using virSetError. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/util/virdbus.c | 31 +++++++++++++++++++------------ src/util/virdbus.h | 4 ++-- src/util/virfirewall.c | 34 ++++++++++------------------------ src/util/virsystemd.c | 15 +++++++-------- 4 files changed, 38 insertions(+), 46 deletions(-) This is a build-breaker fix, but I'm not pushing since the fix is too complicated to go in without some review. diff --git a/src/util/virdbus.c b/src/util/virdbus.c index d9665c1..3522ae0 100644 --- a/src/util/virdbus.c +++ b/src/util/virdbus.c @@ -1522,7 +1522,7 @@ static int virDBusCall(DBusConnection *conn, DBusMessage *call, DBusMessage **replyout, - DBusError *error) + virErrorPtr error) { DBusMessage *reply = NULL; @@ -1530,8 +1530,9 @@ virDBusCall(DBusConnection *conn, int ret = -1; const char *iface, *member, *path, *dest; - if (!error) - dbus_error_init(&localerror); + dbus_error_init(&localerror); + if (error) + memset(error, 0, sizeof(*error)); iface = dbus_message_get_interface(call); member = dbus_message_get_member(call); @@ -1545,13 +1546,20 @@ virDBusCall(DBusConnection *conn, if (!(reply = dbus_connection_send_with_reply_and_block(conn, call, VIR_DBUS_METHOD_CALL_TIMEOUT_MILLIS, - error ? error : &localerror))) { + &localerror))) { PROBE(DBUS_METHOD_ERROR, "'%s.%s' on '%s' at '%s' error %s: %s", iface, member, path, dest, - error ? error->name : localerror.name, - error ? error->message : localerror.message); + localerror.name, + localerror.message); if (error) { + error->level = VIR_ERR_ERROR; + error->code = VIR_ERR_DBUS_SERVICE; + error->domain = VIR_FROM_DBUS; + if (VIR_STRDUP(error->message, localerror.message) < 0) + goto cleanup; + if (VIR_STRDUP(error->str1, localerror.name) < 0) + goto cleanup; ret = 0; } else { virReportError(VIR_ERR_DBUS_SERVICE, _("%s: %s"), member, @@ -1567,8 +1575,9 @@ virDBusCall(DBusConnection *conn, ret = 0; cleanup: - if (!error) - dbus_error_free(&localerror); + if (ret < 0 && error) + virResetError(error); + dbus_error_free(&localerror); if (reply) { if (ret == 0 && replyout) *replyout = reply; @@ -1616,7 +1625,7 @@ virDBusCall(DBusConnection *conn, */ int virDBusCallMethod(DBusConnection *conn, DBusMessage **replyout, - DBusError *error, + virErrorPtr error, const char *destination, const char *path, const char *iface, @@ -1634,8 +1643,6 @@ int virDBusCallMethod(DBusConnection *conn, if (ret < 0) goto cleanup; - ret = -1; - ret = virDBusCall(conn, call, replyout, error); cleanup: @@ -1832,7 +1839,7 @@ int virDBusCreateReply(DBusMessage **reply ATTRIBUTE_UNUSED, int virDBusCallMethod(DBusConnection *conn ATTRIBUTE_UNUSED, DBusMessage **reply ATTRIBUTE_UNUSED, - DBusError *error ATTRIBUTE_UNUSED, + virErrorPtr error ATTRIBUTE_UNUSED, const char *destination ATTRIBUTE_UNUSED, const char *path ATTRIBUTE_UNUSED, const char *iface ATTRIBUTE_UNUSED, diff --git a/src/util/virdbus.h b/src/util/virdbus.h index d0c7de2..e2b8d2b 100644 --- a/src/util/virdbus.h +++ b/src/util/virdbus.h @@ -28,7 +28,7 @@ # else # define DBusConnection void # define DBusMessage void -# define DBusError void +# define dbus_message_unref(m) do {} while (0) # endif # include "internal.h" @@ -62,7 +62,7 @@ int virDBusCreateReplyV(DBusMessage **reply, int virDBusCallMethod(DBusConnection *conn, DBusMessage **reply, - DBusError *error, + virErrorPtr error, const char *destination, const char *path, const char *iface, diff --git a/src/util/virfirewall.c b/src/util/virfirewall.c index b536912..a01b703 100644 --- a/src/util/virfirewall.c +++ b/src/util/virfirewall.c @@ -156,7 +156,6 @@ static int virFirewallValidateBackend(virFirewallBackend backend) { VIR_DEBUG("Validating backend %d", backend); -#if WITH_DBUS if (backend == VIR_FIREWALL_BACKEND_AUTOMATIC || backend == VIR_FIREWALL_BACKEND_FIREWALLD) { int rv = virDBusIsServiceRegistered(VIR_FIREWALL_FIREWALLD_SERVICE); @@ -180,16 +179,6 @@ virFirewallValidateBackend(virFirewallBackend backend) backend = VIR_FIREWALL_BACKEND_FIREWALLD; } } -#else - if (backend == VIR_FIREWALL_BACKEND_AUTOMATIC) { - VIR_DEBUG("DBus support disabled, trying direct backend"); - backend = VIR_FIREWALL_BACKEND_DIRECT; - } else if (backend == VIR_FIREWALL_BACKEND_FIREWALLD) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("firewalld firewall backend requested, but DBus support disabled")); - return -1; - } -#endif if (backend == VIR_FIREWALL_BACKEND_DIRECT) { const char *commands[] = { @@ -755,7 +744,6 @@ virFirewallApplyRuleDirect(virFirewallRulePtr rule, } -#ifdef WITH_DBUS static int virFirewallApplyRuleFirewallD(virFirewallRulePtr rule, bool ignoreErrors, @@ -764,13 +752,13 @@ virFirewallApplyRuleFirewallD(virFirewallRulePtr rule, const char *ipv = virFirewallLayerFirewallDTypeToString(rule->layer); DBusConnection *sysbus = virDBusGetSystemBus(); DBusMessage *reply = NULL; - DBusError error; + virError error; int ret = -1; if (!sysbus) return -1; - dbus_error_init(&error); + memset(&error, 0, sizeof(error)); if (!ipv) { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -789,10 +777,13 @@ virFirewallApplyRuleFirewallD(virFirewallRulePtr rule, "sa&s", ipv, (int)rule->argsLen, - rule->args) < 0) + rule->args) < 0) { + VIR_ERROR("Here fail"); goto cleanup; + } - if (dbus_error_is_set(&error)) { + VIR_ERROR("Error %d: %s\n", error.level, error.message); + if (error.level == VIR_ERR_ERROR) { /* * As of firewalld-0.3.9.3-1.fc20.noarch the name and * message fields in the error look like @@ -820,11 +811,9 @@ virFirewallApplyRuleFirewallD(virFirewallRulePtr rule, */ if (ignoreErrors) { VIR_DEBUG("Ignoring error '%s': '%s'", - error.name, error.message); + error.str1, error.message); } else { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("Unable to apply rule '%s'"), - error.message); + virSetError(&error); goto cleanup; } } else { @@ -835,12 +824,11 @@ virFirewallApplyRuleFirewallD(virFirewallRulePtr rule, ret = 0; cleanup: - dbus_error_free(&error); + virResetError(&error); if (reply) dbus_message_unref(reply); return ret; } -#endif static int virFirewallApplyRule(virFirewallPtr firewall, @@ -862,12 +850,10 @@ virFirewallApplyRule(virFirewallPtr firewall, if (virFirewallApplyRuleDirect(rule, ignoreErrors, &output) < 0) return -1; break; -#if WITH_DBUS case VIR_FIREWALL_BACKEND_FIREWALLD: if (virFirewallApplyRuleFirewallD(rule, ignoreErrors, &output) < 0) return -1; break; -#endif default: virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("Unexpected firewall engine backend")); diff --git a/src/util/virsystemd.c b/src/util/virsystemd.c index 3eea5c2..0b71b26 100644 --- a/src/util/virsystemd.c +++ b/src/util/virsystemd.c @@ -252,8 +252,8 @@ int virSystemdCreateMachine(const char *name, VIR_DEBUG("Attempting to create machine via systemd"); if (virAtomicIntGet(&hasCreateWithNetwork)) { - DBusError error; - dbus_error_init(&error); + virError error; + memset(&error, 0, sizeof(error)); if (virDBusCallMethod(conn, NULL, @@ -280,21 +280,20 @@ int virSystemdCreateMachine(const char *name, "Before", "as", 1, "libvirt-guests.service") < 0) goto cleanup; - if (dbus_error_is_set(&error)) { + if (error.level == VIR_ERR_ERROR) { if (STREQ_NULLABLE("org.freedesktop.DBus.Error.UnknownMethod", - error.name)) { + error.str1)) { VIR_INFO("CreateMachineWithNetwork isn't supported, switching " "to legacy CreateMachine method for systemd-machined"); - dbus_error_free(&error); + virResetError(&error); virAtomicIntSet(&hasCreateWithNetwork, 0); /* Could re-structure without Using goto, but this * avoids another atomic read which would trigger * another memory barrier */ goto fallback; } - virReportError(VIR_ERR_DBUS_SERVICE, - _("CreateMachineWithNetwork: %s"), - error.message ? error.message : _("unknown error")); + virSetError(&error); + virResetError(&error); goto cleanup; } } else { -- 2.1.0

2 6

[libvirt] [PATCH] docs: Fix docs about python bindings package
by Martin Kletzander 19 Jan '15

19 Jan '15

Since the day we removed python bindings from the core repository, the documentation was missing that information. Reported-by: Lingyu Zhu <lynuszhu(a)gmail.com> Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- docs/bindings.html.in | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/docs/bindings.html.in b/docs/bindings.html.in index 13b7d71..95cfe25 100644 --- a/docs/bindings.html.in +++ b/docs/bindings.html.in @@ -44,8 +44,10 @@ </li> <li> - Python: Libvirt comes with direct support for - the Python language. + Python: Libvirt's python bindings are split to a + separate <a href="http://libvirt.org/git/?p=libvirt-python.git">package</a> + since version 1.2.0, older versions came with direct support for the + Python language. If your libvirt is installed as packages, rather than compiled -- 2.2.1

2 1

[libvirt] [PATCH] qemu_conf: check for duplicate security drivers
by Ján Tomko 19 Jan '15

19 Jan '15

Using the same driver multiple times is pointless and it can result in confusing errors: $ virsh start test error: Failed to start domain test error: internal error: security label already defined for VM https://bugzilla.redhat.com/show_bug.cgi?id=1153891 --- src/qemu/qemu_conf.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c index 9539231..a24c5c5 100644 --- a/src/qemu/qemu_conf.c +++ b/src/qemu/qemu_conf.c @@ -460,7 +460,7 @@ int virQEMUDriverConfigLoadFile(virQEMUDriverConfigPtr cfg, p = virConfGetValue(conf, "security_driver"); if (p && p->type == VIR_CONF_LIST) { - size_t len; + size_t len, j; virConfValuePtr pp; /* Calc length and check items */ @@ -476,6 +476,13 @@ int virQEMUDriverConfigLoadFile(virQEMUDriverConfigPtr cfg, goto cleanup; for (i = 0, pp = p->list; pp; i++, pp = pp->next) { + for (j = 0; j < i; j++) { + if (STREQ(pp->str, cfg->securityDriverNames[j])) { + virReportError(VIR_ERR_CONF_SYNTAX, + _("Duplicate security driver %s"), pp->str); + goto cleanup; + } + } if (VIR_STRDUP(cfg->securityDriverNames[i], pp->str) < 0) goto cleanup; } -- 2.0.4

2 1

[libvirt] [PATCH 0/4] Misc work on systemd/lxc
by Daniel P. Berrange 19 Jan '15

19 Jan '15

This is just a few patches working towards an eventual goal of being ble to tell system the tap/veth devices associated with a qemu/lxc guest. Daniel P. Berrange (4): Add support for systemd-machined CreateMachineWithNetwork Add systemd/dtrace probes for DBus APIs Log dtrace/systemd probes at INFO level instead of DEBUG Change int to size_t in size var for tap/vhost FDs po/POTFILES.in | 1 + src/libvirt_probes.d | 5 ++ src/lxc/lxc_cgroup.c | 1 + src/qemu/qemu_cgroup.c | 1 + src/qemu/qemu_command.c | 18 +++---- src/qemu/qemu_command.h | 10 ++-- src/qemu/qemu_hotplug.c | 4 +- src/util/vircgroup.c | 8 ++++ src/util/vircgroup.h | 2 + src/util/virdbus.c | 26 +++++++++-- src/util/virnetdevtap.c | 8 ++-- src/util/virnetdevtap.h | 4 +- src/util/virprobe.h | 10 ++-- src/util/virsystemd.c | 122 +++++++++++++++++++++++++++++++++++++----------- src/util/virsystemd.h | 2 + tests/virsystemdtest.c | 36 ++++++++++++++ 16 files changed, 202 insertions(+), 56 deletions(-) -- 2.1.0

3 7

[libvirt] [PATCH] Conditionalize use of -Wno-suggest-attribute=format pragma
by Daniel P. Berrange 18 Jan '15

18 Jan '15

Many GCC versions don't understand -Wno-suggest-attribute=format so the pragma must only be used when supported --- m4/virt-compile-warnings.m4 | 6 ++++++ src/internal.h | 8 +++++++- 2 files changed, 13 insertions(+), 1 deletion(-) diff --git a/m4/virt-compile-warnings.m4 b/m4/virt-compile-warnings.m4 index 532a777..b412a42 100644 --- a/m4/virt-compile-warnings.m4 +++ b/m4/virt-compile-warnings.m4 @@ -215,6 +215,12 @@ AC_DEFUN([LIBVIRT_COMPILE_WARNINGS],[ ;; esac + case $WARN_CFLAGS in + *-Wsuggest-attribute=format*) + AC_DEFINE(HAVE_SUGGEST_ATTRIBUTE_FORMAT, [1], [Whether -Wsuggest-attribute=format works]) + ;; + esac + # Silence certain warnings in gnulib, and use improved glibc headers AC_DEFINE([lint], [1], [Define to 1 if the compiler is checking for lint.]) diff --git a/src/internal.h b/src/internal.h index 9855c49..4d473af 100644 --- a/src/internal.h +++ b/src/internal.h @@ -234,9 +234,15 @@ # define VIR_WARNINGS_NO_CAST_ALIGN \ _Pragma ("GCC diagnostic push") \ _Pragma ("GCC diagnostic ignored \"-Wcast-align\"") -# define VIR_WARNINGS_NO_PRINTF \ + +# if HAVE_SUGGEST_ATTRIBUTE_FORMAT +# define VIR_WARNINGS_NO_PRINTF \ _Pragma ("GCC diagnostic push") \ _Pragma ("GCC diagnostic ignored \"-Wsuggest-attribute=format\"") +# else +# define VIR_WARNINGS_NO_PRINTF \ + _Pragma ("GCC diagnostic push") +# endif # define VIR_WARNINGS_RESET \ _Pragma ("GCC diagnostic pop") -- 2.1.0

3 3

[libvirt] [PATCH V3] libxl: Set path to console on domain startup.
by Anthony PERARD 17 Jan '15

17 Jan '15

The path to the pty of a Xen PV console is set only in virDomainOpenConsole. But this is done too late. A call to virDomainGetXMLDesc done before OpenConsole will not have the path to the pty, but a call after OpenConsole will. e.g. of the current issue. Starting a domain with '<console type="pty"/>' Then: virDomainGetXMLDesc(): <devices> <console type='pty'> <target type='xen' port='0'/> </console> </devices> virDomainOpenConsole() virDomainGetXMLDesc(): <devices> <console type='pty' tty='/dev/pts/30'> <source path='/dev/pts/30'/> <target type='xen' port='0'/> </console> </devices> The patch intend to have the TTY path on the first call of GetXMLDesc. This is done by setting up the path at domain start up instead of in OpenConsole. https://bugzilla.redhat.com/show_bug.cgi?id=1170743 Signed-off-by: Anthony PERARD <anthony.perard(a)citrix.com> --- Change in V3: Using aop_console_how from libxl_domain_create_new() Ignore empty string that can return libxl_console_get_tty() Change in V2: Adding bug report link. Reword the last part of the patch description. Cleanup the code. Use VIR_FREE before VIR_STRDUP. Remove the code from OpenConsole as it is now a duplicate. CC: Jim Fehlig <jfehlig(a)suse.com> CC: Ian Campbell <Ian.Campbell(a)citrix.com> CC: Ian Jackson <Ian.Jackson(a)eu.citrix.com> --- src/libxl/libxl_domain.c | 47 ++++++++++++++++++++++++++++++++++++++++++++--- src/libxl/libxl_driver.c | 15 --------------- 2 files changed, 44 insertions(+), 18 deletions(-) diff --git a/src/libxl/libxl_domain.c b/src/libxl/libxl_domain.c index 9185117..804f9b9 100644 --- a/src/libxl/libxl_domain.c +++ b/src/libxl/libxl_domain.c @@ -1149,6 +1149,42 @@ libxlDomainFreeMem(libxlDomainObjPrivatePtr priv, libxl_domain_config *d_config) return ret; } +static void +libxlConsoleCallback(libxl_ctx *ctx, libxl_event* ev, void *for_callback) +{ + virDomainObjPtr vm = for_callback; + libxlDomainObjPrivatePtr priv = vm->privateData; + int i; + + virObjectLock(vm); + for (i = 0; i < vm->def->nconsoles; i++) { + virDomainChrDefPtr chr = vm->def->consoles[i]; + if (chr && chr->source.type == VIR_DOMAIN_CHR_TYPE_PTY) { + libxl_console_type console_type; + char *console = NULL; + int ret; + + console_type = + (chr->targetType == VIR_DOMAIN_CHR_CONSOLE_TARGET_TYPE_SERIAL ? + LIBXL_CONSOLE_TYPE_SERIAL : LIBXL_CONSOLE_TYPE_PV); + ret = libxl_console_get_tty(priv->ctx, ev->domid, + chr->target.port, console_type, + &console); + if (!ret) { + VIR_FREE(chr->source.data.file.path); + if (console && console[0] != '\0') { + ignore_value(VIR_STRDUP(chr->source.data.file.path, + console)); + } + } + VIR_FREE(console); + } + } + virObjectUnlock(vm); + libxl_event_free(ctx, ev); +} + + /* * Start a domain through libxenlight. * @@ -1173,6 +1209,7 @@ libxlDomainStart(libxlDriverPrivatePtr driver, virDomainObjPtr vm, libxl_domain_restore_params params; #endif virHostdevManagerPtr hostdev_mgr = driver->hostdevMgr; + libxl_asyncprogress_how aop_console_how; libxl_domain_config_init(&d_config); @@ -1242,17 +1279,21 @@ libxlDomainStart(libxlDriverPrivatePtr driver, virDomainObjPtr vm, /* Unlock virDomainObj while creating the domain */ virObjectUnlock(vm); + + aop_console_how.for_callback = vm; + aop_console_how.callback = libxlConsoleCallback; if (restore_fd < 0) { ret = libxl_domain_create_new(priv->ctx, &d_config, - &domid, NULL, NULL); + &domid, NULL, &aop_console_how); } else { #ifdef LIBXL_HAVE_DOMAIN_CREATE_RESTORE_PARAMS params.checkpointed_stream = 0; ret = libxl_domain_create_restore(priv->ctx, &d_config, &domid, - restore_fd, &params, NULL, NULL); + restore_fd, &params, NULL, + &aop_console_how); #else ret = libxl_domain_create_restore(priv->ctx, &d_config, &domid, - restore_fd, NULL, NULL); + restore_fd, NULL, &aop_console_how); #endif } virObjectLock(vm); diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index cad5101..fc0949d 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -3982,10 +3982,8 @@ libxlDomainOpenConsole(virDomainPtr dom, { virDomainObjPtr vm = NULL; int ret = -1; - libxl_console_type console_type; virDomainChrDefPtr chr = NULL; libxlDomainObjPrivatePtr priv; - char *console = NULL; virCheckFlags(VIR_DOMAIN_CONSOLE_FORCE, -1); @@ -4027,18 +4025,6 @@ libxlDomainOpenConsole(virDomainPtr dom, goto cleanup; } - console_type = - (chr->targetType == VIR_DOMAIN_CHR_CONSOLE_TARGET_TYPE_SERIAL ? - LIBXL_CONSOLE_TYPE_SERIAL : LIBXL_CONSOLE_TYPE_PV); - - ret = libxl_console_get_tty(priv->ctx, vm->def->id, chr->target.port, - console_type, &console); - if (ret) - goto cleanup; - - if (VIR_STRDUP(chr->source.data.file.path, console) < 0) - goto cleanup; - /* handle mutually exclusive access to console devices */ ret = virChrdevOpen(priv->devs, &chr->source, @@ -4052,7 +4038,6 @@ libxlDomainOpenConsole(virDomainPtr dom, } cleanup: - VIR_FREE(console); if (vm) virObjectUnlock(vm); return ret; -- Anthony PERARD

2 1

[libvirt] Git repository browsing online?
by Gary R Hook 17 Jan '15

17 Jan '15

Warning: possibly rhetorical or already answered question ahead. Why is the git repository that is located here: http://libvirt.org/git/?p=libvirt-appdev-guide.git;a=summary which is pointed to by this page here: http://www.libvirt.org/downloads.html Only valid up to 13 Aug 2012? If I pull a copy down from the link on the downloads page: git clone git://libvirt.org/libvirt-appdev-guide.git I get what appears to be a current clone. I'm sure I have missed something somewhere, but shouldn't/couldn't the downloads page be up-to-date? Even if only to say, "no online browsing" ? -- Gary R Hook Senior Kernel Engineer NIMBOXX, Inc

3 4

[libvirt] [PATCH] Make ZFS storage pool XML tests optional
by Gary R Hook 17 Jan '15

17 Jan '15

From: Gary R Hook <grhookatwork(a)gmail.com> Do not run ZFS tests when ZFS is unsupported in the environment. The recent patch b4af40226d09adeaf9e33a1d6594c4e8ce7f771d adds tests to storagepoolxml2xmltest for the optional ZFS feature. When ZFS is not included in the configuration these tests should not / cannot be run. Modify Makefile.am and the test source file to check for use of the feature and compile in the tests accordingly. Signed-off-by: Gary R Hook <gary.hook(a)nimboxx.com> --- tests/Makefile.am | 4 ++++ tests/storagepoolxml2xmltest.c | 2 ++ 2 files changed, 6 insertions(+) diff --git a/tests/Makefile.am b/tests/Makefile.am index e9418ea..8f8df27 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -780,6 +780,10 @@ storagepoolxml2xmltest_SOURCES = \ storagepoolxml2xmltest.c \ testutils.c testutils.h storagepoolxml2xmltest_LDADD = $(LDADDS) +storagepoolxml2xmltest_CFLAGS = $(AM_CFLAGS) +if WITH_STORAGE_ZFS +storagepoolxml2xmltest_CFLAGS += -DWITH_STORAGE_ZFS +endif nodedevxml2xmltest_SOURCES = \ nodedevxml2xmltest.c \ diff --git a/tests/storagepoolxml2xmltest.c b/tests/storagepoolxml2xmltest.c index 52e2193..270f75d 100644 --- a/tests/storagepoolxml2xmltest.c +++ b/tests/storagepoolxml2xmltest.c @@ -106,8 +106,10 @@ mymain(void) DO_TEST("pool-gluster"); DO_TEST("pool-gluster-sub"); DO_TEST("pool-scsi-type-scsi-host-stable"); +#ifdef WITH_STORAGE_ZFS DO_TEST("pool-zfs"); DO_TEST("pool-zfs-sourcedev"); +#endif return ret == 0 ? EXIT_SUCCESS : EXIT_FAILURE; } -- 1.9.1

2 4

[libvirt] [PATCHv2] Coverity fix: properly check for 0 ipv6 address.
by Cédric Bosdonnat 16 Jan '15

16 Jan '15

--- src/util/virsocketaddr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/util/virsocketaddr.c b/src/util/virsocketaddr.c index 91bcadf..67ed330 100644 --- a/src/util/virsocketaddr.c +++ b/src/util/virsocketaddr.c @@ -850,7 +850,7 @@ virSocketAddrGetIpPrefix(const virSocketAddr *address, } return -1; } else if (VIR_SOCKET_ADDR_IS_FAMILY(address, AF_INET6)) { - if (address->data.inet6.sin6_addr.s6_addr == 0) + if (virSocketAddrIsWildcard(address)) return 0; return 64; } -- 2.1.2

3 2

[libvirt] [PATCH v3 0/3] Add schema validation for domain XML
by Daniel P. Berrange 16 Jan '15

16 Jan '15

A followup to https://www.redhat.com/archives/libvir-list/2015-January/msg00203.html In this posting, the 3 remaining patches only are posted: - Catch errors parsing RNG schema - Use virFileFindResource for findning RNG - Don't enable validation by default in virsh define/create - Other misc typos Daniel P. Berrange (3): Add virXMLValidateAgainstSchema helper method Add support for schema validation when passing in XML virsh: add support for domain XML validation include/libvirt/libvirt-domain.h | 5 +++ include/libvirt/virterror.h | 1 + src/bhyve/bhyve_driver.c | 16 ++++++-- src/conf/domain_conf.c | 11 +++++ src/conf/domain_conf.h | 1 + src/esx/esx_driver.c | 8 +++- src/internal.h | 4 ++ src/libvirt-domain.c | 2 +- src/libvirt_private.syms | 1 + src/libxl/libxl_driver.c | 17 ++++++-- src/lxc/lxc_driver.c | 18 +++++++-- src/openvz/openvz_driver.c | 16 ++++++-- src/parallels/parallels_driver.c | 8 +++- src/phyp/phyp_driver.c | 8 +++- src/qemu/qemu_driver.c | 16 ++++++-- src/test/test_driver.c | 16 ++++++-- src/uml/uml_driver.c | 17 ++++++-- src/util/virerror.c | 6 +++ src/util/virxml.c | 79 ++++++++++++++++++++++++++++++++++++ src/util/virxml.h | 5 +++ src/vbox/vbox_common.c | 8 +++- src/vmware/vmware_driver.c | 16 ++++++-- src/xen/xen_driver.c | 16 ++++++-- src/xenapi/xenapi_driver.c | 17 ++++++-- tools/virsh-domain.c | 87 +++++++++++++++++++++++++++++++++++----- 25 files changed, 341 insertions(+), 58 deletions(-) -- 2.1.0

3 11

[libvirt] [PATCHv2 0/7] Sharing code for domain and network routes
by Cédric Bosdonnat 16 Jan '15

16 Jan '15

Hi guys, Here are a few patches to have common route definitions for domains and networks. What has changed since v1: * Split into several patches for backportability as suggested by Laine * Moved the virNetworkRouteDef struct definition in the c file to make sure all routes are created with the virNetworkRouteDefCreate (or Parse) function. This way we can get rid of the redundant checks in the format function. * Moved the prefix computing code into a new virNetworkRouteDefGetPrefix function. * Applied and tweaked John's patches. * Fix the virSocketAddrGetIpPrefix 0.0.0.0 special case as suggested by Laine. Cédric Bosdonnat (5): Fix ipv6 regex in RNG schemas to match '::' Move network route definition to networkcommon.rng Move code related to network routes to networkcommon_conf.[ch] Use the network route definitions for domains virSocketAddrGetIpPrefix 0.0.0.0 special case John Ferlan (2): domain_conf: Resolve Coverity RESOURCE_LEAK domain_conf: Check errors from virSocketAddrFormat docs/formatdomain.html.in | 9 +- docs/schemas/basictypes.rng | 2 +- docs/schemas/domaincommon.rng | 29 +- docs/schemas/network.rng | 20 +- docs/schemas/networkcommon.rng | 22 ++ po/POTFILES.in | 1 + src/Makefile.am | 3 +- src/conf/domain_conf.c | 136 ++----- src/conf/domain_conf.h | 14 +- src/conf/network_conf.c | 297 +--------------- src/conf/network_conf.h | 22 +- src/conf/networkcommon_conf.c | 414 ++++++++++++++++++++++ src/conf/networkcommon_conf.h | 72 ++++ src/libvirt_private.syms | 11 + src/lxc/lxc_container.c | 22 +- src/lxc/lxc_native.c | 20 +- src/network/bridge_driver.c | 44 +-- src/util/virsocketaddr.c | 6 + tests/lxcconf2xmldata/lxcconf2xml-physnetwork.xml | 4 +- tests/lxcconf2xmldata/lxcconf2xml-simple.xml | 4 +- tests/lxcxml2xmldata/lxc-hostdev.xml | 4 +- tests/lxcxml2xmldata/lxc-idmap.xml | 4 +- 22 files changed, 633 insertions(+), 527 deletions(-) create mode 100644 src/conf/networkcommon_conf.c create mode 100644 src/conf/networkcommon_conf.h -- 2.1.2

5 25

[libvirt] [PATCHv3 0/3] Fix vmdef usage after ExitMonitor
by Ján Tomko 16 Jan '15

16 Jan '15

The patches 1-2, 6-8 from v2 have been pushed already Device removal and device detach are now squashed together. Both attach and detach skip audit when the domain crashed. (Left to be cleaned up by a separate series) Patches checking the return value in all other non-fatal cases were squashed together with the patch adding ATTRIBUTE_RETURN_CHECK. Rebased to master, added qemuProcessUpdateVideoRamSize, and fixed numerous wrong return values and a few missing EndJobs. Ján Tomko (3): Fix vmdef usage after domain crash in monitor on device detach Fix vmdef usage after domain crash in monitor on device attach Always check return value of qemuDomainObjExitMonitor src/qemu/qemu_domain.c | 11 +-- src/qemu/qemu_domain.h | 3 +- src/qemu/qemu_driver.c | 166 ++++++++++++++++++++++--------------- src/qemu/qemu_hotplug.c | 202 ++++++++++++++++++++++++++++++---------------- src/qemu/qemu_hotplug.h | 6 +- src/qemu/qemu_migration.c | 139 ++++++++++++++++--------------- src/qemu/qemu_process.c | 115 +++++++++++++------------- 7 files changed, 373 insertions(+), 269 deletions(-) -- 2.0.4

2 5

[libvirt] [PATCH] ipv6 check for :: address
by Cédric Bosdonnat 16 Jan '15

16 Jan '15

Hi John, Here is a possible fix for the coverity issue you mentioned. Since ipv6 addresses are too big to be converted to a single integer, I just looped over the array... not sure that is the most elegant way to do it. Cédric Bosdonnat (1): Coverity fix: properly check for 0 ipv6 address. src/util/virsocketaddr.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) -- 2.1.2

2 2

[libvirt] [PATCH v2 0/5] parallels: manage container's filesystems
by Dmitry Guryanov 16 Jan '15

16 Jan '15

This patch series adds ability to manage container's filesystems: you can add or remove fs from existing container or create new one, based on existing disk image. Dmitry Guryanov (5): add ploop fs driver type parallels: dump info about container filesystems parallels: commit with PVCF_DETACH_HDD_BUNDLE flag parallels: allow to add filesystems to container parallels: create container from existing image docs/formatdomain.html.in | 3 +- docs/schemas/domaincommon.rng | 1 + docs/schemas/storagecommon.rng | 1 + src/conf/domain_conf.c | 3 +- src/conf/domain_conf.h | 1 + src/parallels/parallels_sdk.c | 209 +++++++++++++++++++-- src/qemu/qemu_command.c | 1 + .../domain-parallels-ct-simple.xml | 5 + 8 files changed, 206 insertions(+), 18 deletions(-) -- 2.1.0

2 9

[libvirt] [PATCH 0/3] Fix adding PCI bridge controller to extend default PCI bus
by Erik Skultety 16 Jan '15

16 Jan '15

Erik Skultety (3): qemu: Tweak auto adding PCI bridge controller when extending default PCI bus qemu: qemuDomainAssignPCIAddresses fix dead code qemu: Add XML test for too many PCI devices on default PCI bus src/qemu/qemu_command.c | 26 ++- tests/qemuxml2argvdata/qemuxml2argv-pci-many.args | 63 +++++++ tests/qemuxml2argvdata/qemuxml2argv-pci-many.xml | 214 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 2 + 4 files changed, 301 insertions(+), 4 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pci-many.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pci-many.xml -- 1.9.3

3 7

[libvirt] [PATCH] network: Resolve Coverity FORWARD_NULL
by John Ferlan 16 Jan '15

16 Jan '15

Commit id 'ca481a6f' added virNetworkRouteDefFree which may be called in an error path from lxcAddNetworkRouteDefinition with 'route = NULL'. So just add the (!def) at the top to resolve. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/conf/networkcommon_conf.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/conf/networkcommon_conf.c b/src/conf/networkcommon_conf.c index 3f0896d..7b7a851 100644 --- a/src/conf/networkcommon_conf.c +++ b/src/conf/networkcommon_conf.c @@ -52,6 +52,8 @@ struct _virNetworkRouteDef { void virNetworkRouteDefFree(virNetworkRouteDefPtr def) { + if (!def) + return; VIR_FREE(def->family); VIR_FREE(def); } -- 2.1.0

2 1

[libvirt] [PATCH 0/2] qemu: Add support for PMU feature
by Martin Kletzander 16 Jan '15

16 Jan '15

This is based on top of: https://www.redhat.com/archives/libvir-list/2015-January/msg00307.html Martin Kletzander (2): docs, schema, conf: Add support for PMU feature qemu: Add support for enabling/disabling PMU docs/formatdomain.html.in | 6 ++++++ docs/schemas/domaincommon.rng | 11 ++++++++++ src/conf/domain_conf.c | 8 +++++-- src/conf/domain_conf.h | 3 ++- src/qemu/qemu_command.c | 12 ++++++++++- .../qemuxml2argv-pmu-feature-off.args | 5 +++++ .../qemuxml2argv-pmu-feature-off.xml | 25 ++++++++++++++++++++++ .../qemuxml2argvdata/qemuxml2argv-pmu-feature.args | 5 +++++ .../qemuxml2argvdata/qemuxml2argv-pmu-feature.xml | 25 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 3 +++ .../qemuxml2xmlout-pmu-feature.xml | 25 ++++++++++++++++++++++ tests/qemuxml2xmltest.c | 3 +++ 12 files changed, 127 insertions(+), 4 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pmu-feature-off.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pmu-feature-off.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pmu-feature.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pmu-feature.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-pmu-feature.xml -- 2.2.1

2 3

[libvirt] [PATCH v3] virsh: add support for domain XML validation
by Daniel P. Berrange 16 Jan '15

16 Jan '15

The 'virsh edit' command gets XML validation enabled by default, with a --skip-validate option to disable it. The 'virsh define' and 'virsh create' commands get a --validate option to enable it, to avoid regressions for existing scripts. The quality of error reporting from libxml2 varies depending on the type of XML error made. Sometimes it is quite clear and useful, other times it is obscure & inaccurate. At least the user will see an error now, rather than having their XML modification silently disappear. --- tools/virsh-domain.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 52 insertions(+), 5 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 6733cfa..bab44fe 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -133,6 +133,28 @@ vshCommandOptDomainBy(vshControl *ctl, const vshCmd *cmd, return vshLookupDomainInternal(ctl, cmd->def->name, n, flags); } +static virDomainPtr +vshDomainDefine(virConnectPtr conn, const char *xml, unsigned int flags) +{ + virDomainPtr dom; + if (flags) { + dom = virDomainDefineXMLFlags(conn, xml, flags); + /* If validate is the only flag, just drop it and + * try again. + */ + if (!dom) { + virErrorPtr err = virGetLastError(); + if (err && + (err->code == VIR_ERR_NO_SUPPORT) && + (flags == VIR_DOMAIN_DEFINE_VALIDATE)) + dom = virDomainDefineXML(conn, xml); + } + } else { + dom = virDomainDefineXML(conn, xml); + } + return dom; +} + VIR_ENUM_DECL(vshDomainVcpuState) VIR_ENUM_IMPL(vshDomainVcpuState, VIR_VCPU_LAST, @@ -7154,6 +7176,10 @@ static const vshCmdOptDef opts_create[] = { .type = VSH_OT_STRING, .help = N_("pass file descriptors N,M,... to the guest") }, + {.name = "validate", + .type = VSH_OT_BOOL, + .help = N_("validate the XML against the schema") + }, {.name = NULL} }; @@ -7167,7 +7193,7 @@ cmdCreate(vshControl *ctl, const vshCmd *cmd) #ifndef WIN32 bool console = vshCommandOptBool(cmd, "console"); #endif - unsigned int flags = VIR_DOMAIN_NONE; + unsigned int flags = 0; size_t nfds = 0; int *fds = NULL; @@ -7184,6 +7210,8 @@ cmdCreate(vshControl *ctl, const vshCmd *cmd) flags |= VIR_DOMAIN_START_PAUSED; if (vshCommandOptBool(cmd, "autodestroy")) flags |= VIR_DOMAIN_START_AUTODESTROY; + if (vshCommandOptBool(cmd, "validate")) + flags |= VIR_DOMAIN_START_VALIDATE; if (nfds) dom = virDomainCreateXMLWithFiles(ctl->conn, buffer, nfds, fds, flags); @@ -7229,6 +7257,10 @@ static const vshCmdOptDef opts_define[] = { .flags = VSH_OFLAG_REQ, .help = N_("file containing an XML domain description") }, + {.name = "validate", + .type = VSH_OT_BOOL, + .help = N_("validate the XML against the schema") + }, {.name = NULL} }; @@ -7239,14 +7271,21 @@ cmdDefine(vshControl *ctl, const vshCmd *cmd) const char *from = NULL; bool ret = true; char *buffer; + unsigned int flags = 0; if (vshCommandOptStringReq(ctl, cmd, "file", &from) < 0) return false; + if (vshCommandOptBool(cmd, "validate")) + flags |= VIR_DOMAIN_DEFINE_VALIDATE; + if (virFileReadAll(from, VSH_MAX_XML_FILE, &buffer) < 0) return false; - dom = virDomainDefineXML(ctl->conn, buffer); + if (flags) + dom = virDomainDefineXMLFlags(ctl->conn, buffer, flags); + else + dom = virDomainDefineXML(ctl->conn, buffer); VIR_FREE(buffer); if (dom != NULL) { @@ -11105,6 +11144,10 @@ static const vshCmdOptDef opts_edit[] = { .flags = VSH_OFLAG_REQ, .help = N_("domain name, id or uuid") }, + {.name = "skip-validate", + .type = VSH_OT_BOOL, + .help = N_("skip validation of the XML against the schema") + }, {.name = NULL} }; @@ -11114,13 +11157,17 @@ cmdEdit(vshControl *ctl, const vshCmd *cmd) bool ret = false; virDomainPtr dom = NULL; virDomainPtr dom_edited = NULL; - unsigned int flags = VIR_DOMAIN_XML_SECURE | VIR_DOMAIN_XML_INACTIVE; + unsigned int query_flags = VIR_DOMAIN_XML_SECURE | VIR_DOMAIN_XML_INACTIVE; + unsigned int define_flags = VIR_DOMAIN_DEFINE_VALIDATE; dom = vshCommandOptDomain(ctl, cmd, NULL); if (dom == NULL) goto cleanup; -#define EDIT_GET_XML virDomainGetXMLDesc(dom, flags) + if (vshCommandOptBool(cmd, "skip-validate")) + define_flags &= ~VIR_DOMAIN_DEFINE_VALIDATE; + +#define EDIT_GET_XML virDomainGetXMLDesc(dom, query_flags) #define EDIT_NOT_CHANGED \ do { \ vshPrint(ctl, _("Domain %s XML configuration not changed.\n"), \ @@ -11129,7 +11176,7 @@ cmdEdit(vshControl *ctl, const vshCmd *cmd) goto edit_cleanup; \ } while (0) #define EDIT_DEFINE \ - (dom_edited = virDomainDefineXML(ctl->conn, doc_edited)) + (dom_edited = vshDomainDefine(ctl->conn, doc_edited, define_flags)) #include "virsh-edit.c" vshPrint(ctl, _("Domain %s XML configuration edited.\n"), -- 2.1.0

2 1

[libvirt] [PATCH] qemu: fix fail to start vm with one cell use memdev and another not
by Luyao Huang 16 Jan '15

16 Jan '15

https://bugzilla.redhat.com/show_bug.cgi?id=1182467 We changed our build numa qemu cmd line way in f309db, But qemu cannot accept one cell use memdev and another not, so this cause a issue when we start a vm with one cell have hugpage or nodemask settings and others not. After this patch we will use memdev for all cells if there is at least one cell need this. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/qemu/qemu_command.c | 88 +++++++++++++--------- .../qemuxml2argv-numatune-memnode-no-memory.args | 3 +- 2 files changed, 55 insertions(+), 36 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 3cafbfb..11ca355 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -6613,14 +6613,18 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, virBitmapPtr nodeset) { size_t i, j; - virBuffer buf = VIR_BUFFER_INITIALIZER; + virBuffer buf[def->cpu->ncells]; virDomainHugePagePtr master_hugepage = NULL; char *cpumask = NULL, *tmpmask = NULL, *next = NULL; char *nodemask = NULL; char *mem_path = NULL; int ret = -1; + bool need_memdev = false; const long system_page_size = sysconf(_SC_PAGESIZE) / 1024; + for (i = 0; i < def->cpu->ncells; i++) + memset(&buf[i], 0, sizeof(buf[i])); + if (virDomainNumatuneHasPerNodeBinding(def->numatune) && !(virQEMUCapsGet(qemuCaps, QEMU_CAPS_OBJECT_MEMORY_RAM) || virQEMUCapsGet(qemuCaps, QEMU_CAPS_OBJECT_MEMORY_FILE))) { @@ -6672,20 +6676,8 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, def->cpu->cells[i].mem = cellmem * 1024; virMemAccess memAccess = def->cpu->cells[i].memAccess; - VIR_FREE(cpumask); VIR_FREE(nodemask); - if (!(cpumask = virBitmapFormat(def->cpu->cells[i].cpumask))) - goto cleanup; - - if (strchr(cpumask, ',') && - !virQEMUCapsGet(qemuCaps, QEMU_CAPS_NUMA)) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", - _("disjoint NUMA cpu ranges are not supported " - "with this QEMU")); - goto cleanup; - } - if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_OBJECT_MEMORY_RAM) || virQEMUCapsGet(qemuCaps, QEMU_CAPS_OBJECT_MEMORY_FILE)) { virDomainNumatuneMemMode mode; @@ -6752,17 +6744,17 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, if (!(mem_path = qemuGetHugepagePath(&cfg->hugetlbfs[j]))) goto cleanup; - virBufferAsprintf(&buf, + virBufferAsprintf(&buf[i], "memory-backend-file,prealloc=yes,mem-path=%s", mem_path); switch (memAccess) { case VIR_MEM_ACCESS_SHARED: - virBufferAddLit(&buf, ",share=on"); + virBufferAddLit(&buf[i], ",share=on"); break; case VIR_MEM_ACCESS_PRIVATE: - virBufferAddLit(&buf, ",share=off"); + virBufferAddLit(&buf[i], ",share=off"); break; case VIR_MEM_ACCESS_DEFAULT: @@ -6777,10 +6769,10 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, "only with hugepages")); goto cleanup; } - virBufferAddLit(&buf, "memory-backend-ram"); + virBufferAddLit(&buf[i], "memory-backend-ram"); } - virBufferAsprintf(&buf, ",size=%lluM,id=ram-node%zu", cellmem, i); + virBufferAsprintf(&buf[i], ",size=%lluM,id=ram-node%zu", cellmem, i); if (virDomainNumatuneMaybeFormatNodeset(def->numatune, nodeset, &nodemask, i) < 0) @@ -6798,19 +6790,15 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, for (tmpmask = nodemask; tmpmask; tmpmask = next) { if ((next = strchr(tmpmask, ','))) *(next++) = '\0'; - virBufferAddLit(&buf, ",host-nodes="); - virBufferAdd(&buf, tmpmask, -1); + virBufferAddLit(&buf[i], ",host-nodes="); + virBufferAdd(&buf[i], tmpmask, -1); } - virBufferAsprintf(&buf, ",policy=%s", policy); + virBufferAsprintf(&buf[i], ",policy=%s", policy); } - if (hugepage || nodemask) { - virCommandAddArg(cmd, "-object"); - virCommandAddArgBuffer(cmd, &buf); - } else { - virBufferFreeAndReset(&buf); - } + if (hugepage || nodemask) + need_memdev = true; } else { if (memAccess) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", @@ -6819,23 +6807,52 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, goto cleanup; } } + } + + /*We have finnished build args for all cells, and know + *if we can use memdev or no need use it. If there is + *one cells have hugepage or nodemask settings, we need + *build all cells with -object memory-backend-ram, if not + *just use -numa.*/ + for (i = 0; i < def->cpu->ncells; i++) { + unsigned long long cellmem = VIR_DIV_UP(def->cpu->cells[i].mem, 1024); + def->cpu->cells[i].mem = cellmem * 1024; + VIR_FREE(cpumask); + + if (need_memdev) { + virCommandAddArg(cmd, "-object"); + virCommandAddArgBuffer(cmd, &buf[i]); + } else { + virBufferFreeAndReset(&buf[i]); + } virCommandAddArg(cmd, "-numa"); - virBufferAsprintf(&buf, "node,nodeid=%zu", i); + virBufferAsprintf(&buf[i], "node,nodeid=%zu", i); + + if (!(cpumask = virBitmapFormat(def->cpu->cells[i].cpumask))) + goto cleanup; + + if (strchr(cpumask, ',') && + !virQEMUCapsGet(qemuCaps, QEMU_CAPS_NUMA)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("disjoint NUMA cpu ranges are not supported " + "with this QEMU")); + goto cleanup; + } for (tmpmask = cpumask; tmpmask; tmpmask = next) { if ((next = strchr(tmpmask, ','))) *(next++) = '\0'; - virBufferAddLit(&buf, ",cpus="); - virBufferAdd(&buf, tmpmask, -1); + virBufferAddLit(&buf[i], ",cpus="); + virBufferAdd(&buf[i], tmpmask, -1); } - if (hugepage || nodemask) - virBufferAsprintf(&buf, ",memdev=ram-node%zu", i); + if (need_memdev) + virBufferAsprintf(&buf[i], ",memdev=ram-node%zu", i); else - virBufferAsprintf(&buf, ",mem=%llu", cellmem); + virBufferAsprintf(&buf[i], ",mem=%llu", cellmem); - virCommandAddArgBuffer(cmd, &buf); + virCommandAddArgBuffer(cmd, &buf[i]); } ret = 0; @@ -6843,7 +6860,8 @@ qemuBuildNumaArgStr(virQEMUDriverConfigPtr cfg, VIR_FREE(cpumask); VIR_FREE(nodemask); VIR_FREE(mem_path); - virBufferFreeAndReset(&buf); + for (i = 0; i < def->cpu->ncells; i++) + virBufferFreeAndReset(&buf[i]); return ret; } diff --git a/tests/qemuxml2argvdata/qemuxml2argv-numatune-memnode-no-memory.args b/tests/qemuxml2argvdata/qemuxml2argv-numatune-memnode-no-memory.args index 2addf97..b0e274c 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-numatune-memnode-no-memory.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-numatune-memnode-no-memory.args @@ -2,6 +2,7 @@ LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \ /usr/bin/kvm -S -M pc -m 64 -smp 2 \ -object memory-backend-ram,size=32M,id=ram-node0,host-nodes=3,policy=preferred \ -numa node,nodeid=0,cpus=0,memdev=ram-node0 \ --numa node,nodeid=1,cpus=1,mem=32 \ +-object memory-backend-ram,size=32M,id=ram-node1 \ +-numa node,nodeid=1,cpus=1,memdev=ram-node1 \ -nographic -monitor unix:/tmp/test-monitor,server,nowait \ -no-acpi -boot c -usb -net none -serial none -parallel none -- 1.8.3.1

4 5

[libvirt] [PATCH] Allow shrinking of file based volumes
by Daniel P. Berrange 16 Jan '15

16 Jan '15

While the main storage driver code allows the flag VIR_STORAGE_VOL_RESIZE_SHRINK to be set, none of the backend drivers are supporting it. At the very least this can work for plain file based volumes since we just ftruncate() them to the new size. It does not work with qcow2 volumes, but we can arguably delegate to qemu-img for error reporting for that instead of second guessing this for ourselves: $ virsh vol-resize --shrink /home/berrange/VirtualMachines/demo.qcow2 2G error: Failed to change size of volume 'demo.qcow2' to 2G error: internal error: Child process (/usr/bin/qemu-img resize /home/berrange/VirtualMachines/demo.qcow2 2147483648) unexpected exit status 1: qemu-img: qcow2 doesn't support shrinking images yet qemu-img: This image does not support resize See also https://bugzilla.redhat.com/show_bug.cgi?id=1021802 --- src/storage/storage_backend_fs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/storage/storage_backend_fs.c b/src/storage/storage_backend_fs.c index 34f2153..79543a6 100644 --- a/src/storage/storage_backend_fs.c +++ b/src/storage/storage_backend_fs.c @@ -1262,7 +1262,8 @@ virStorageBackendFileSystemVolResize(virConnectPtr conn ATTRIBUTE_UNUSED, unsigned long long capacity, unsigned int flags) { - virCheckFlags(VIR_STORAGE_VOL_RESIZE_ALLOCATE, -1); + virCheckFlags(VIR_STORAGE_VOL_RESIZE_ALLOCATE | + VIR_STORAGE_VOL_RESIZE_SHRINK, -1); bool pre_allocate = flags & VIR_STORAGE_VOL_RESIZE_ALLOCATE; -- 2.1.0

2 1

[libvirt] [PATCH 0/2] Simplify and test CPU feature formatting code
by Ján Tomko 16 Jan '15

16 Jan '15

Ján Tomko (2): Add an XML test for host-model CPU with features Simplify condition for formatting CPU features src/conf/cpu_conf.c | 42 +++++++++++----------- .../qemuxml2argv-cpu-host-model-features.xml | 33 +++++++++++++++++ tests/qemuxml2xmltest.c | 1 + 3 files changed, 54 insertions(+), 22 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-cpu-host-model-features.xml -- 2.0.4

2 4

[libvirt] [PATCH v2 0/2] check IOMMU group devices usage when preparing device for vfio passthrough
by Shivaprasad G Bhat 16 Jan '15

16 Jan '15

==================================================================== If a device in the same iommu group is in use by a different vm, the guest boot fails with the below error during vfio passthrough. bash-4.2$ virsh start demo error: Failed to start domain demo error: internal error: process exited while connecting to monitor: 2014-12-02T13:43:52.020136Z qemu-system-x86_64: -device vfio-pci,host=00:1c.3,id=hostdev0,bus=pci.0,addr=0x5: vfio: error opening /dev/vfio/7: Device or resource busy Solution: ===================================================================== The patch iterates through the iommu group devices and errors out cleanly mentioning the device and the guest which is using the device. With Patch bash-4.2$ virsh start demo error: Failed to start domain demo error: Requested operation is not valid: PCI device 0000:0d:00.0 is in use by driver QEMU, domain vm10 --- Shivaprasad G Bhat (2): Implement virPCIDeviceGetAddress function check IOMMU group devices usage when preparing device for vfio passthrough src/util/virhostdev.c | 89 ++++++++++++++++++++++++++++++------------------- src/util/virpci.c | 16 +++++++++ src/util/virpci.h | 1 + 3 files changed, 72 insertions(+), 34 deletions(-) -- Signature

3 6

[libvirt] [PATCH 0/3] Add stateDir for storage driver and use for disk backend
by John Ferlan 16 Jan '15

16 Jan '15

https://bugzilla.redhat.com/show_bug.cgi?id=1138516 Using the network (and domain) for reference, add a 'stateDir' for the Storage Driver in order to save off pool and/or volume specific information into a live running area. This is useful for the disk backend where volume creation can define a specific name and target format type to use; however, that information cannot be encoded into the 'parted' data used to create the partition(s) on the disk. Thus the data is lost for refresh and restart/reload. The following patches will add the stateDir, then the infrastructure for storage volume xml to be saved off, and finally saving/using that data for the refresh, restart/reload path for the disk backend driver. John Ferlan (3): storage: Create a stateDir storage: Introduce virStorageVol{Save|Delete}Status storage: Adjustments to disk backend to use stateDir src/conf/storage_conf.c | 112 +++++++++++++++++++++++++++++++++---- src/conf/storage_conf.h | 13 ++++- src/libvirt_private.syms | 2 + src/parallels/parallels_storage.c | 5 +- src/storage/storage_backend_disk.c | 90 ++++++++++++++++++++++++++++- src/storage/storage_driver.c | 32 +++++++++-- 6 files changed, 232 insertions(+), 22 deletions(-) -- 2.1.0

2 7

[libvirt] [PATCH] virNetworkDefUpdateIPDHCPHost: Don't crash when updating network
by Michal Privoznik 16 Jan '15

16 Jan '15

https://bugzilla.redhat.com/show_bug.cgi?id=1182486 When updating a network and adding new ip-dhcp-host entry, the deamon may crash. The problem is, we iterate over existing <host/> entries trying to compare MAC addresses to see if there's already an existing rule. However, not all entries are required to have MAC address. For instance, the following is perfectly valid entry: <host id='00:04:58:fd:e4:15:1b:09:4c:0e:09:af:e4:d3:8c:b8:ca:1e' name='redhatipv6.redhat.com' ip='2001:db8:ca2:2::119'/> When the checking loop iterates over this, the entry's MAC address is accessed directly. Well, the fix is obvious - check if the address is defined before trying to compare it. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/conf/network_conf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/conf/network_conf.c b/src/conf/network_conf.c index 26fe18d..3b719de 100644 --- a/src/conf/network_conf.c +++ b/src/conf/network_conf.c @@ -3601,7 +3601,7 @@ virNetworkDefUpdateIPDHCPHost(virNetworkDefPtr def, /* log error if an entry with same name/address/ip already exists */ for (i = 0; i < ipdef->nhosts; i++) { - if ((host.mac && + if ((host.mac && ipdef->hosts[i].mac && !virMacAddrCompare(host.mac, ipdef->hosts[i].mac)) || (host.name && STREQ_NULLABLE(host.name, ipdef->hosts[i].name)) || -- 2.0.5

2 1

[libvirt] [python PATCH] build: make it easier to backport event ids
by Eric Blake 16 Jan '15

16 Jan '15

In some cases, it is very easy for downstream distros to backport enum values without requiring a .so bump. Keying the conditional code off of the upstream version where the enum value was added is not ideal, because downstream then has to patch that the feature is available in their build that still reports an earlier version number. For example, if RHEL 7 backports events from 1.2.11 into a build based on 1.2.8, building the python bindings would warn: libvirt-override.c: In function ‘libvirt_virConnectDomainEventRegisterAny’: libvirt-override.c:6653:5: warning: enumeration value ‘VIR_DOMAIN_EVENT_ID_TUNABLE’ not handled in switch [-Wswitch] switch ((virDomainEventID) eventID) { ^ libvirt-override.c:6653:5: warning: enumeration value ‘VIR_DOMAIN_EVENT_ID_AGENT_LIFECYCLE’ not handled in switch [-Wswitch] The solution is simple - use feature-based probes instead of version probes. Since we already scrape the XML API document of whatever libvirt build we are binding, and that XML already documents any downstream enum additions, we can use those as the features for gating conditional compilation. * generator.py (enum): Track event id names. (buildStubs): Output define wrappers for events. * libvirt-override.c (libvirt_virConnectDomainEventBalloonChangeCallback) (libvirt_virConnectDomainEventPMSuspendDiskCallback) (libvirt_virConnectDomainEventDeviceRemovedCallback) (libvirt_virConnectDomainEventTunableCallback) (libvirt_virConnectDomainEventAgentLifecycleCallback) (libvirt_virConnectDomainEventRegisterAny): Use them. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- I mentioned this idea on the list a while back, and finally got time to code it up in a much nicer way than my initial attempt that would have polluted libvirt.h proper. https://www.redhat.com/archives/libvir-list/2014-October/msg00186.html I'm relatively weak at python, and welcome to suggestions on how to avoid the long line in generator.py (maybe a regex match on VIR_.*_EVENT_ID_, but what's the canonical way to write that in python?) generator.py | 11 +++++++++-- libvirt-override.c | 46 +++++++++++++++++++++++----------------------- 2 files changed, 32 insertions(+), 25 deletions(-) diff --git a/generator.py b/generator.py index cf044c9..4fd6d55 100755 --- a/generator.py +++ b/generator.py @@ -9,6 +9,7 @@ qemu_functions = {} enums = {} # { enumType: { enumConstant: enumValue } } lxc_enums = {} # { enumType: { enumConstant: enumValue } } qemu_enums = {} # { enumType: { enumConstant: enumValue } } +event_ids = [] import os import sys @@ -219,6 +220,8 @@ def lxc_function(name, desc, ret, args, file, module, cond): def enum(type, name, value): if type not in enums: enums[type] = {} + if name.startswith('VIR_DOMAIN_EVENT_ID_') or name.startswith('VIR_NETWORK_EVENT_ID_'): + event_ids.append(name) if value == 'VIR_TYPED_PARAM_INT': value = 1 elif value == 'VIR_TYPED_PARAM_UINT': @@ -910,10 +913,10 @@ def buildStubs(module, api_xml): wrapper_file = "build/%s.c" % module include = open(header_file, "w") - include.write("/* Generated */\n\n") + include.write("/* Generated by generator.py */\n\n") export = open(export_file, "w") - export.write("/* Generated */\n\n") + export.write("/* Generated by generator.py */\n\n") wrapper = open(wrapper_file, "w") wrapper.write("/* Generated by generator.py */\n\n") @@ -943,6 +946,10 @@ def buildStubs(module, api_xml): # Write C pointer conversion functions. for classname in primary_classes: print_c_pointer(classname, wrapper, export, include) + # Write define wrappers around event id enums + include.write("\n") + for event_id in event_ids: + include.write("#define %s %s\n" % (event_id, event_id)) include.close() export.close() diff --git a/libvirt-override.c b/libvirt-override.c index e51c44d..88cb527 100644 --- a/libvirt-override.c +++ b/libvirt-override.c @@ -4,7 +4,7 @@ * entry points where an automatically generated stub is * unpractical * - * Copyright (C) 2005, 2007-2014 Red Hat, Inc. + * Copyright (C) 2005, 2007-2015 Red Hat, Inc. * * Daniel Veillard <veillard(a)redhat.com> */ @@ -6346,7 +6346,7 @@ libvirt_virConnectDomainEventPMSuspendCallback(virConnectPtr conn ATTRIBUTE_UNUS } -#if LIBVIR_CHECK_VERSION(0, 10, 0) +#ifdef VIR_DOMAIN_EVENT_ID_BALLOON_CHANGE static int libvirt_virConnectDomainEventBalloonChangeCallback(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, @@ -6398,9 +6398,9 @@ libvirt_virConnectDomainEventBalloonChangeCallback(virConnectPtr conn ATTRIBUTE_ LIBVIRT_RELEASE_THREAD_STATE; return ret; } -#endif /* LIBVIR_CHECK_VERSION(0, 10, 0) */ +#endif /* VIR_DOMAIN_EVENT_ID_BALLOON_CHANGE */ -#if LIBVIR_CHECK_VERSION(1, 0, 0) +#ifdef VIR_DOMAIN_EVENT_ID_PMSUSPEND_DISK static int libvirt_virConnectDomainEventPMSuspendDiskCallback(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, @@ -6452,9 +6452,9 @@ libvirt_virConnectDomainEventPMSuspendDiskCallback(virConnectPtr conn ATTRIBUTE_ LIBVIRT_RELEASE_THREAD_STATE; return ret; } -#endif /* LIBVIR_CHECK_VERSION(1, 0, 0) */ +#endif /* VIR_DOMAIN_EVENT_ID_PMSUSPEND_DISK */ -#if LIBVIR_CHECK_VERSION(1, 1, 1) +#ifdef VIR_DOMAIN_EVENT_ID_DEVICE_REMOVED static int libvirt_virConnectDomainEventDeviceRemovedCallback(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, @@ -6504,9 +6504,9 @@ libvirt_virConnectDomainEventDeviceRemovedCallback(virConnectPtr conn ATTRIBUTE_ LIBVIRT_RELEASE_THREAD_STATE; return ret; } -#endif /* LIBVIR_CHECK_VERSION(1, 1, 1) */ +#endif /* VIR_DOMAIN_EVENT_ID_DEVICE_REMOVED */ -#if LIBVIR_CHECK_VERSION(1, 2, 9) +#ifdef VIR_DOMAIN_EVENT_ID_TUNABLE static int libvirt_virConnectDomainEventTunableCallback(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, @@ -6564,9 +6564,9 @@ libvirt_virConnectDomainEventTunableCallback(virConnectPtr conn ATTRIBUTE_UNUSED return ret; } -#endif /* LIBVIR_CHECK_VERSION(1, 2, 9) */ +#endif /* VIR_DOMAIN_EVENT_ID_TUNABLE */ -#if LIBVIR_CHECK_VERSION(1, 2, 11) +#ifdef VIR_DOMAIN_EVENT_ID_AGENT_LIFECYCLE static int libvirt_virConnectDomainEventAgentLifecycleCallback(virConnectPtr conn ATTRIBUTE_UNUSED, virDomainPtr dom, @@ -6618,7 +6618,7 @@ libvirt_virConnectDomainEventAgentLifecycleCallback(virConnectPtr conn ATTRIBUTE return ret; } -#endif /* LIBVIR_CHECK_VERSION(1, 2, 11) */ +#endif /* VIR_DOMAIN_EVENT_ID_AGENT_LIFECYCLE */ static PyObject * @@ -6676,9 +6676,9 @@ libvirt_virConnectDomainEventRegisterAny(ATTRIBUTE_UNUSED PyObject *self, cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventGenericCallback); break; case VIR_DOMAIN_EVENT_ID_BLOCK_JOB: -#if LIBVIR_CHECK_VERSION(1, 2, 6) +#ifdef VIR_DOMAIN_EVENT_ID_BLOCK_JOB_2 case VIR_DOMAIN_EVENT_ID_BLOCK_JOB_2: -#endif /* LIBVIR_CHECK_VERSION(1, 2, 6) */ +#endif /* VIR_DOMAIN_EVENT_ID_BLOCK_JOB_2 */ cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventBlockJobCallback); break; case VIR_DOMAIN_EVENT_ID_DISK_CHANGE: @@ -6693,31 +6693,31 @@ libvirt_virConnectDomainEventRegisterAny(ATTRIBUTE_UNUSED PyObject *self, case VIR_DOMAIN_EVENT_ID_PMSUSPEND: cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventPMSuspendCallback); break; -#if LIBVIR_CHECK_VERSION(0, 10, 0) +#ifdef VIR_DOMAIN_EVENT_ID_BALLOON_CHANGE case VIR_DOMAIN_EVENT_ID_BALLOON_CHANGE: cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventBalloonChangeCallback); break; -#endif /* LIBVIR_CHECK_VERSION(0, 10, 0) */ -#if LIBVIR_CHECK_VERSION(1, 0, 0) +#endif /* VIR_DOMAIN_EVENT_ID_BALLOON_CHANGE */ +#ifdef VIR_DOMAIN_EVENT_ID_PMSUSPEND_DISK case VIR_DOMAIN_EVENT_ID_PMSUSPEND_DISK: cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventPMSuspendDiskCallback); break; -#endif /* LIBVIR_CHECK_VERSION(1, 0, 0) */ -#if LIBVIR_CHECK_VERSION(1, 1, 1) +#endif /* VIR_DOMAIN_EVENT_ID_PMSUSPEND_DISK */ +#ifdef VIR_DOMAIN_EVENT_ID_DEVICE_REMOVED case VIR_DOMAIN_EVENT_ID_DEVICE_REMOVED: cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventDeviceRemovedCallback); break; -#endif /* LIBVIR_CHECK_VERSION(1, 1, 1) */ -#if LIBVIR_CHECK_VERSION(1, 2, 9) +#endif /* VIR_DOMAIN_EVENT_ID_DEVICE_REMOVED */ +#ifdef VIR_DOMAIN_EVENT_ID_TUNABLE case VIR_DOMAIN_EVENT_ID_TUNABLE: cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventTunableCallback); break; -#endif /* LIBVIR_CHECK_VERSION(1, 2, 9) */ -#if LIBVIR_CHECK_VERSION(1, 2, 11) +#endif /* VIR_DOMAIN_EVENT_ID_TUNABLE */ +#ifdef VIR_DOMAIN_EVENT_ID_AGENT_LIFECYCLE case VIR_DOMAIN_EVENT_ID_AGENT_LIFECYCLE: cb = VIR_DOMAIN_EVENT_CALLBACK(libvirt_virConnectDomainEventAgentLifecycleCallback); break; -#endif /* LIBVIR_CHECK_VERSION(1, 2, 11) */ +#endif /* VIR_DOMAIN_EVENT_ID_AGENT_LIFECYCLE */ case VIR_DOMAIN_EVENT_ID_LAST: break; } -- 2.1.0

3 5

[libvirt] [PATCH] lxc: trivially support DomainHasManagedSaveImage
by Ján Tomko 15 Jan '15

15 Jan '15

Return 0 instead of ERR_NO_SUPPORT to avoid spamming daemon logs when 'virsh dominfo' is run. https://bugzilla.redhat.com/show_bug.cgi?id=1095637 --- src/lxc/lxc_driver.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 07ddce3..b5aa1cc 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -5717,6 +5717,29 @@ lxcNodeAllocPages(virConnectPtr conn, } +static int +lxcDomainHasManagedSaveImage(virDomainPtr dom, unsigned int flags) +{ + virDomainObjPtr vm = NULL; + int ret = -1; + + virCheckFlags(0, -1); + + if (!(vm = lxcDomObjFromDomain(dom))) + return ret; + + if (virDomainHasManagedSaveImageEnsureACL(dom->conn, vm->def) < 0) + goto cleanup; + + ret = 0; + + cleanup: + if (vm) + virObjectUnlock(vm); + return ret; +} + + /* Function Tables */ static virHypervisorDriver lxcDriver = { .no = VIR_DRV_LXC, @@ -5810,6 +5833,7 @@ static virHypervisorDriver lxcDriver = { .domainLxcOpenNamespace = lxcDomainLxcOpenNamespace, /* 1.0.2 */ .nodeGetFreePages = lxcNodeGetFreePages, /* 1.2.6 */ .nodeAllocPages = lxcNodeAllocPages, /* 1.2.9 */ + .domainHasManagedSaveImage = lxcDomainHasManagedSaveImage, /* 1.2.12 */ }; static virStateDriver lxcStateDriver = { -- 2.0.4

2 2

[libvirt] [PATCH] qemu: forget free priv->origname when get fail in qemuMigrationPrepareAny
by Luyao Huang 15 Jan '15

15 Jan '15

https://bugzilla.redhat.com/show_bug.cgi?id=1181182 When we meet error in qemuMigrationPrepareAny and goto cleanup with rc < 0, we forget clear the priv->origname and this will make this vm migrate fail next time because leave a wrong origname in priv, and will Generate a wrong cookie when do migrate next time. This patch will make priv->origname is NULL when migrate fail in target host. Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/qemu/qemu_migration.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 77e0b35..a282c43 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -3104,6 +3104,7 @@ qemuMigrationPrepareAny(virQEMUDriverPtr driver, if (ret < 0 && priv) { /* priv is set right after vm is added to the list of domains * and there is no 'goto cleanup;' in the middle of those */ + VIR_FREE(priv->origname); virPortAllocatorRelease(driver->migrationPorts, priv->nbdPort); priv->nbdPort = 0; qemuDomainRemoveInactive(driver, vm); -- 1.8.3.1

2 1

[libvirt] [PATCH] qemu: snapshot: inactive external snapshot can't work after libvirtd restart
by Shanzhi Yu 15 Jan '15

15 Jan '15

When create inactive external snapshot, after update disk definitions, virDomainSaveConfig is needed, if not after restart libvirtd the new snapshot file definitions in xml will be lost. Reproduce steps: 1. prepare a shut off guest $ virsh domstate rhel7 && virsh domblklist rhel7 shut off Target Source ------------------------------------------------ vda /var/lib/libvirt/images/rhel7.img 2. create external disk snapshot $ virsh snapshot-create rhel7 --disk-only && virsh domblklist rhel7 Domain snapshot 1417882967 created Target Source ------------------------------------------------ vda /var/lib/libvirt/images/rhel7.1417882967 3. restart libvirtd then check guest source file $ service libvirtd restart && virsh domblklist rhel7 Redirecting to /bin/systemctl restart libvirtd.service Target Source ------------------------------------------------ vda /var/lib/libvirt/images/rhel7.img This was first reported by Eric Blake http://www.redhat.com/archives/libvir-list/2014-December/msg00369.html Signed-off-by: Shanzhi Yu <shyu(a)redhat.com> --- src/qemu/qemu_driver.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 9152cf5..9f8ea0a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -12847,6 +12847,9 @@ qemuDomainSnapshotCreateInactiveExternal(virQEMUDriverPtr driver, goto cleanup; } defdisk->src->format = snapdisk->src->format; + + if (virDomainSaveConfig(cfg->configDir, vm->def) < 0) + goto cleanup; } } -- 2.1.0

2 2

[libvirt] Plans for next release
by Daniel Veillard 15 Jan '15

15 Jan '15

It's start to think pushing a new release. Last one was mid-december so we ought to push one this month, I will be travelling the last week and since February is short I suppose pushing the release a bit early is not a problem. We have around 180 commits already, and I would like to push for 1.2.12 around Monday 26, so possibly entering freeze Tuesday or Wednesday next week. If this plan is a problem, please raise your concerns, thanks, Daniel P.S.: I will be at FOSDEM at the end of the month -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH V5 0/3] Parser for xen-xl config format
by Jim Fehlig 15 Jan '15

15 Jan '15

V5 of xen-xl parser. V4 was close https://www.redhat.com/archives/libvir-list/2015-January/msg00429.html but the tests did not exersice the spice parsing/formatting code. It is now fixed in V5, along with addressing a few V4 comments from eblake. Jim Fehlig (1): Introduce support for parsing/formatting Xen xl config format Kiarie Kahurani (2): tests: Tests for the xen-xl parser libxl: Add support for parsing/formating Xen XL config configure.ac | 3 + po/POTFILES.in | 1 + src/Makefile.am | 11 + src/libvirt_xenxlconfig.syms | 12 + src/libxl/libxl_driver.c | 32 ++- src/xenconfig/xen_common.c | 3 +- src/xenconfig/xen_xl.c | 515 +++++++++++++++++++++++++++++++++++ src/xenconfig/xen_xl.h | 35 +++ tests/Makefile.am | 11 + tests/testutilsxen.c | 50 ++++ tests/testutilsxen.h | 9 +- tests/xlconfigdata/test-new-disk.cfg | 26 ++ tests/xlconfigdata/test-new-disk.xml | 51 ++++ tests/xlconfigdata/test-spice.cfg | 31 +++ tests/xlconfigdata/test-spice.xml | 46 ++++ tests/xlconfigtest.c | 225 +++++++++++++++ tools/virsh.pod | 8 +- 17 files changed, 1055 insertions(+), 14 deletions(-) create mode 100644 src/libvirt_xenxlconfig.syms create mode 100644 src/xenconfig/xen_xl.c create mode 100644 src/xenconfig/xen_xl.h create mode 100644 tests/xlconfigdata/test-new-disk.cfg create mode 100644 tests/xlconfigdata/test-new-disk.xml create mode 100644 tests/xlconfigdata/test-spice.cfg create mode 100644 tests/xlconfigdata/test-spice.xml create mode 100644 tests/xlconfigtest.c -- 1.8.4.5

2 5

[libvirt] Attempting to pivot disk with incomplete mirror results in hang
by Scott Sullivan 15 Jan '15

15 Jan '15

I have what appears to be a bug when pivoting a disk during a block copy that is not yet 100% finished, resulting in the pivot command hanging. I have verified this problem on libvirt 1.2.10. Here's what Im seeing (transient guest): 1.) Start the block copy: [root@host ~]# virsh blockcopy f20-SPICE vda /dev/sdc --format=raw --blockdev Block Copy started [root@host ~]# 2.) Query status to see it works/is started [root@host ~]# virsh blockjob --info f20-SPICE vda Block Copy: [ 1 %] [root@host ~]# 3.) Attempt the pivot before mirror reaches 100% (this makes cmd hang) [root@test-parent-kvm ~]# virsh blockjob f20-SPICE vda --pivot ^^^ cmd is now hanging ------------------------------------------------------------- When its hanging, I see this in /var/log/libvirt/libvirtd.log 19:29:33.376+0000: 1845: error : qemuDomainBlockPivot:15367 : block copy still active: disk 'vda' not ready for pivot yet 19:30:31.000+0000: 1842: warning : qemuDomainObjBeginJobInternal:1376 : Cannot start job (query, none) for domain f20-SPICE; current job is (modify, none) owned by (1845, 0) 19:30:31.000+0000: 1842: error : qemuDomainObjBeginJobInternal:1381 : Timed out during operation: cannot acquire state change lock This then makes other commands querying this domain hang as well, such as: virsh blockjob --info f20-SPICE vda With this being put into log: 19:30:31.000+0000: 1842: error : qemuDomainObjBeginJobInternal:1381 : Timed out during operation: cannot acquire state change lock 19:34:45.568+0000: 1841: error : virNetSocketReadWire:1571 : End of file while reading data: Input/output error Is this expected behavior? It would be nice if on step #3 instead of hanging the command returned in an error state saying cannot pivot because the mirror is not yet 100% or similar. I am happy to test later versions or patches as needed.

2 5

[libvirt] [PATCH V4 0/3] Parser for xen-xl config format
by Jim Fehlig 14 Jan '15

14 Jan '15

It's been a long, twisting road to V4 of the Xen xl parser. V3 [1] was based on a flex-based parser that was copied from the Xen project and proved to be a bit challenging to integrate properly with autotools. But as it turns out, Xen provides an interface to the parser via libxlutil. I hadn't realized this interface was available for external consumption since the corresponding header file was never installed. Patch sent to xen-devel [2] to install the header, but in the meantime need to declare gthe imported libxlutil functions as extern (see patch 1). V4 uses libxlutil, which has simplified the series quite a bit and eliminates the potential of the copied flex parser diverging from the canonical source in xen.git. [1] https://www.redhat.com/archives/libvir-list/2014-December/msg00765.html [2] http://lists.xen.org/archives/html/xen-devel/2015-01/msg00690.html Jim Fehlig (1): Introduce support for parsing/formatting Xen xl config format Kiarie Kahurani (2): tests: Tests for the xen-xl parser libxl: Add support for parsing/formating Xen XL config configure.ac | 3 + po/POTFILES.in | 1 + src/Makefile.am | 11 + src/libvirt_xenxlconfig.syms | 12 + src/libxl/libxl_driver.c | 32 ++- src/xenconfig/xen_common.c | 3 +- src/xenconfig/xen_xl.c | 515 +++++++++++++++++++++++++++++++++++ src/xenconfig/xen_xl.h | 35 +++ tests/Makefile.am | 11 + tests/testutilsxen.c | 50 ++++ tests/testutilsxen.h | 9 +- tests/xlconfigdata/test-new-disk.cfg | 26 ++ tests/xlconfigdata/test-new-disk.xml | 51 ++++ tests/xlconfigdata/test-spice.cfg | 32 +++ tests/xlconfigdata/test-spice.xml | 45 +++ tests/xlconfigtest.c | 225 +++++++++++++++ tools/virsh.pod | 8 +- 17 files changed, 1055 insertions(+), 14 deletions(-) create mode 100644 src/libvirt_xenxlconfig.syms create mode 100644 src/xenconfig/xen_xl.c create mode 100644 src/xenconfig/xen_xl.h create mode 100644 tests/xlconfigdata/test-new-disk.cfg create mode 100644 tests/xlconfigdata/test-new-disk.xml create mode 100644 tests/xlconfigdata/test-spice.cfg create mode 100644 tests/xlconfigdata/test-spice.xml create mode 100644 tests/xlconfigtest.c -- 1.8.4.5

3 12

[libvirt] question about the flow of live migration with non-shared storage in libvirt
by Zhang Haoyu 14 Jan '15

14 Jan '15

Hi, Someone can help detail the the flow of live migration with non-shared storage in libvirt? Especially, the flow of migration of non-shared storage part? Thanks, Zhang Haoyu

2 1

[libvirt] [PATCHv2 00/14] Fix vmdef usage after ExitMonitor
by Ján Tomko 14 Jan '15

14 Jan '15

Patches 3-8 fix possible crash/invalid memory access if QEMU crashes while we're in the monitor. Patches 9-13 change all other callers of qemuDomainObjExitMonitor to exit early in that case, but should not fix any real issues. They are necessary to turn on ATTRIBUTE_RETURN_CHECK for the ExitMonitor call. https://bugzilla.redhat.com/show_bug.cgi?id=1161024 Ján Tomko (14): Check for domain liveness in qemuDomainObjExitMonitor Mark the domain as active in qemuhotplugtest Fix vmdef usage after domain crash in monitor on device removal Fix vmdef usage after domain crash in monitor on device detach Fix vmdef usage after domain crash in monitor on device attach Fix vmdef usage while in monitor in qemuDomainHotplugVcpus Fix vmdef usage while in monitor in BlockStat* APIs Fix vmdef usage while in monitor in qemu process Exit early after domain crash in monitor on device hotplug Exit early after domain crash in monitor on migration Exit early after domain crash in monitor in qemu_process Exit early after domain crash in monitor in qemu_driver Exit early after domain crash in monitor on snapshots Add ATTRIBUTE_RETURN_CHECK to qemuDomainObjExitMonitor src/qemu/THREADS.txt | 5 ++ src/qemu/qemu_domain.c | 27 +++++-- src/qemu/qemu_domain.h | 7 +- src/qemu/qemu_driver.c | 196 +++++++++++++++++++++++++++++----------------- src/qemu/qemu_hotplug.c | 183 ++++++++++++++++++++++++++----------------- src/qemu/qemu_hotplug.h | 6 +- src/qemu/qemu_migration.c | 137 ++++++++++++++++---------------- src/qemu/qemu_process.c | 128 +++++++++++++++++------------- tests/qemuhotplugtest.c | 6 ++ 9 files changed, 410 insertions(+), 285 deletions(-) -- 2.0.4

2 35

[libvirt] [PATCH 0/2] check IOMMU group devices usage during vfio device passthrough
by Shivaprasad G Bhat 14 Jan '15

14 Jan '15

Problem: ==================================================================== If a device in the same iommu group is in use by a different vm, the guest boot fails with the below error during vfio passthrough. bash-4.2$ virsh start demo error: Failed to start domain demo error: internal error: process exited while connecting to monitor: 2014-12-02T13:43:52.020136Z qemu-system-x86_64: -device vfio-pci,host=00:1c.3,id=hostdev0,bus=pci.0,addr=0x5: vfio: error opening /dev/vfio/7: Device or resource busy Solution: ===================================================================== The patch iterates through the iommu group devices and errors out cleanly mentioning the device and the guest which is using the device. With Patch bash-4.2$ virsh start demo error: Failed to start domain demo error: Requested operation is not valid: PCI device 0000:0d:00.0 is in use by driver QEMU, domain vm10 --- Shivaprasad G Bhat (2): Move struct _virPCIDevice definition from virpci.c to virpci.h check IOMMU group devices usage when preparing device for vfio passthrough src/util/virhostdev.c | 87 +++++++++++++++++++++++++++++-------------------- src/util/virpci.c | 29 ---------------- src/util/virpci.h | 30 +++++++++++++++++ 3 files changed, 82 insertions(+), 64 deletions(-) -- Signature

3 6

[libvirt] [PATCH v3 0/2] Fix migration to older libvirt
by Pavel Hrdina 14 Jan '15

14 Jan '15

Recently I've implemented new feature that we can set "vgamem_mb" for QXL device and also I've fixed libvirt to honor the 'vram' attribute and pass that value to QEMU process. But with this change the migration to older libvirt stopped working because QEMU silently updates the video memory size if the value is too low or there are some dependencies (for example QXL device has to have 'ram' size twice as 'vgamem'). To fix the migration we need to load the updated values from QEMU and store them into the status XML. v2: - don't fail if JSON monitor isn't available v1: - removed unnecessary movement of qemuMonitorJSONObjectProperty - completely rewritten the second patch Pavel Hrdina (2): qemu_monitor: introduce new function to get QOM path qemu_process: detect updated video ram size values from QEMU src/qemu/qemu_monitor.c | 206 ++++++++++++++++++++++++++++--------------- src/qemu/qemu_monitor.h | 4 + src/qemu/qemu_monitor_json.c | 69 +++++++++++++++ src/qemu/qemu_monitor_json.h | 3 + src/qemu/qemu_process.c | 67 ++++++++++++++ 5 files changed, 279 insertions(+), 70 deletions(-) -- 2.0.5

2 7

[libvirt] [PATCH] qemu_driver: fix setting vcpus for offline domain
by Pavel Hrdina 14 Jan '15

14 Jan '15

Commit e3435caf fixed hot-plugging of vcpus with strict memory pinning on NUMA hosts, but unfortunately it also broke updating number of vcpus for offline guests using our API. The issue is that we try to create a cpu cgroup for non-running guest which fails as there are no cgroups for that domain. We should create cgroups and update cpuset.mems only if we are hot-plugging. Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com> --- src/qemu/qemu_driver.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index cdf4173..3a343d0 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4576,21 +4576,23 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, priv = vm->privateData; - if (virCgroupNewEmulator(priv->cgroup, false, &cgroup_temp) < 0) - goto cleanup; - - if (!(all_nodes = virNumaGetHostNodeset())) - goto cleanup; - - if (!(all_nodes_str = virBitmapFormat(all_nodes))) - goto cleanup; - if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; - if (virCgroupGetCpusetMems(cgroup_temp, &mem_mask) < 0 || - virCgroupSetCpusetMems(cgroup_temp, all_nodes_str) < 0) - goto endjob; + if (flags & VIR_DOMAIN_AFFECT_LIVE) { + if (virCgroupNewEmulator(priv->cgroup, false, &cgroup_temp) < 0) + goto cleanup; + + if (!(all_nodes = virNumaGetHostNodeset())) + goto cleanup; + + if (!(all_nodes_str = virBitmapFormat(all_nodes))) + goto cleanup; + + if (virCgroupGetCpusetMems(cgroup_temp, &mem_mask) < 0 || + virCgroupSetCpusetMems(cgroup_temp, all_nodes_str) < 0) + goto endjob; + } maximum = (flags & VIR_DOMAIN_VCPU_MAXIMUM) != 0; flags &= ~VIR_DOMAIN_VCPU_MAXIMUM; -- 2.0.5

2 2

[libvirt] [PATCH v2 0/4] UEFI and NVRAM store file format
by Michal Privoznik 14 Jan '15

14 Jan '15

diff to v1: -Laszlo's review worked in (notably, even UEFI can have a different format than raw) Michal Privoznik (4): conf: Make virDomainLoaderDefParseXML parse <nvram/> too conf: Allow UEFI and NVRAM format selection qemu: Adapt to UEFI and NVRAM store file format change domaincaps: Expose UEFI and NVRAM store file format capabilities docs/formatdomain.html.in | 18 +++++-- docs/formatdomaincaps.html.in | 26 ++++++++++ docs/schemas/domaincaps.rng | 13 +++++ docs/schemas/domaincommon.rng | 16 ++++++ src/conf/domain_capabilities.c | 18 +++++++ src/conf/domain_capabilities.h | 2 + src/conf/domain_conf.c | 57 +++++++++++++++++----- src/conf/domain_conf.h | 11 +++++ src/libvirt_private.syms | 2 + src/qemu/qemu_capabilities.c | 10 +++- src/qemu/qemu_command.c | 12 +++-- tests/domaincapsschemadata/domaincaps-full.xml | 10 ++++ .../domaincaps-qemu_1.6.50-1.xml | 10 ++++ tests/domaincapstest.c | 2 + .../qemuxml2argv-bios-nvram-qcow2.args | 10 ++++ .../qemuxml2argv-bios-nvram-qcow2.xml | 40 +++++++++++++++ tests/qemuxml2argvtest.c | 2 + .../qemuxml2xmlout-bios-nvram.xml | 40 +++++++++++++++ tests/qemuxml2xmltest.c | 2 +- 19 files changed, 279 insertions(+), 22 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-bios-nvram-qcow2.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-bios-nvram-qcow2.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-bios-nvram.xml -- 2.0.5

2 9

[libvirt] [PATCH v2] qemu, lxc: Warn if setting QoS on unsupported vNIC types
by Michal Privoznik 14 Jan '15

14 Jan '15

https://bugzilla.redhat.com/show_bug.cgi?id=1165993 So, there are still plenty of vNIC types that we don't know how to set bandwidth on. Let's warn explicitly in case user has requested it instead of pretending everything was set. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Diff to v1: -switch from virReportError(); goto error; to VIR_WARN(). src/lxc/lxc_driver.c | 18 +++++++++++++----- src/lxc/lxc_process.c | 18 +++++++++++++----- src/qemu/qemu_command.c | 19 +++++++++++++------ src/qemu/qemu_hotplug.c | 18 +++++++++++++----- 4 files changed, 52 insertions(+), 21 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index f503fcf..07ddce3 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -4148,6 +4148,7 @@ lxcDomainAttachDeviceNetLive(virConnectPtr conn, virLXCDomainObjPrivatePtr priv = vm->privateData; int ret = -1; int actualType; + virNetDevBandwidthPtr actualBandwidth; char *veth = NULL; if (!priv->initpid) { @@ -4195,11 +4196,18 @@ lxcDomainAttachDeviceNetLive(virConnectPtr conn, _("Network device type is not supported")); goto cleanup; } - /* set network bandwidth */ - if (virNetDevSupportBandwidth(actualType) && - virNetDevBandwidthSet(net->ifname, - virDomainNetGetActualBandwidth(net), false) < 0) - goto cleanup; + /* Set bandwidth or warn if requested and not supported. */ + actualBandwidth = virDomainNetGetActualBandwidth(net); + if (actualBandwidth) { + if (virNetDevSupportBandwidth(actualType)) { + if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false) < 0) + goto cleanup; + } else { + VIR_WARN("setting bandwidth on interfaces of " + "type '%s' is not implemented yet", + virDomainNetTypeToString(actualType)); + } + } if (virNetDevSetNamespace(veth, priv->initpid) < 0) { virDomainAuditNet(vm, NULL, net, "attach", false); diff --git a/src/lxc/lxc_process.c b/src/lxc/lxc_process.c index af08de9..89e9249 100644 --- a/src/lxc/lxc_process.c +++ b/src/lxc/lxc_process.c @@ -369,6 +369,7 @@ static int virLXCProcessSetupInterfaces(virConnectPtr conn, for (i = 0; i < def->nnets; i++) { char *veth = NULL; + virNetDevBandwidthPtr actualBandwidth; /* If appropriate, grab a physical device from the configured * network's pool of devices, or resolve bridge device name * to the one defined in the network definition. @@ -422,11 +423,18 @@ static int virLXCProcessSetupInterfaces(virConnectPtr conn, } - /* set network bandwidth */ - if (virNetDevSupportBandwidth(type) && - virNetDevBandwidthSet(net->ifname, - virDomainNetGetActualBandwidth(net), false) < 0) - goto cleanup; + /* Set bandwidth or warn if requested and not supported. */ + actualBandwidth = virDomainNetGetActualBandwidth(net); + if (actualBandwidth) { + if (virNetDevSupportBandwidth(type)) { + if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false) < 0) + goto cleanup; + } else { + VIR_WARN("setting bandwidth on interfaces of " + "type '%s' is not implemented yet", + virDomainNetTypeToString(type)); + } + } (*veths)[(*nveths)-1] = veth; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 3346e95..3cafbfb 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -7365,6 +7365,7 @@ qemuBuildInterfaceCommandLine(virCommandPtr cmd, char **tapfdName = NULL; char **vhostfdName = NULL; int actualType = virDomainNetGetActualType(net); + virNetDevBandwidthPtr actualBandwidth; size_t i; if (actualType == VIR_DOMAIN_NET_TYPE_VHOSTUSER) @@ -7415,12 +7416,18 @@ qemuBuildInterfaceCommandLine(virCommandPtr cmd, goto cleanup; } - /* Set Bandwidth */ - if (virNetDevSupportBandwidth(actualType) && - virNetDevBandwidthSet(net->ifname, - virDomainNetGetActualBandwidth(net), - false) < 0) - goto cleanup; + /* Set bandwidth or warn if requested and not supported. */ + actualBandwidth = virDomainNetGetActualBandwidth(net); + if (actualBandwidth) { + if (virNetDevSupportBandwidth(actualType)) { + if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false) < 0) + goto cleanup; + } else { + VIR_WARN("setting bandwidth on interfaces of " + "type '%s' is not implemented yet", + virDomainNetTypeToString(actualType)); + } + } if ((actualType == VIR_DOMAIN_NET_TYPE_NETWORK || actualType == VIR_DOMAIN_NET_TYPE_BRIDGE || diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 1714341..bf1f69a 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -838,6 +838,7 @@ int qemuDomainAttachNetDevice(virConnectPtr conn, bool releaseaddr = false; bool iface_connected = false; int actualType; + virNetDevBandwidthPtr actualBandwidth; virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); size_t i; @@ -926,11 +927,18 @@ int qemuDomainAttachNetDevice(virConnectPtr conn, if (qemuInterfaceStartDevice(net) < 0) goto cleanup; - /* Set Bandwidth */ - if (virNetDevSupportBandwidth(actualType) && - virNetDevBandwidthSet(net->ifname, - virDomainNetGetActualBandwidth(net), false) < 0) - goto cleanup; + /* Set bandwidth or warn if requested and not supported. */ + actualBandwidth = virDomainNetGetActualBandwidth(net); + if (actualBandwidth) { + if (virNetDevSupportBandwidth(actualType)) { + if (virNetDevBandwidthSet(net->ifname, actualBandwidth, false) < 0) + goto cleanup; + } else { + VIR_WARN("setting bandwidth on interfaces of " + "type '%s' is not implemented yet", + virDomainNetTypeToString(actualType)); + } + } for (i = 0; i < tapfdSize; i++) { if (virSecurityManagerSetTapFDLabel(driver->securityManager, -- 2.0.5

2 1

[libvirt] [PATCH 0/2] Forbid starting VMs with bandwidth on unsupported vNIC type
by Michal Privoznik 14 Jan '15

14 Jan '15

The patch 1/2 is trivial. The patch 2/2 is controversial and I expect heavy discussion on it :-) Thing is, there might be somebody misusing our XML to store the QoS setting, and do the actual setting themselves in startup hook. Now with this patch, we will forbid such usage scenario. Michal Privoznik (2): conf: Increase virNetDevBandwidthParse intelligence qemu, lxc: Fail explicitly if setting QoS on unsupported vNIC types src/conf/domain_conf.c | 10 ++++++---- src/conf/netdev_bandwidth_conf.c | 38 +++++++++++++++++++++++--------------- src/conf/netdev_bandwidth_conf.h | 7 ++++--- src/conf/network_conf.c | 7 +++---- src/lxc/lxc_driver.c | 20 +++++++++++++++----- src/lxc/lxc_process.c | 20 +++++++++++++++----- src/qemu/qemu_command.c | 21 +++++++++++++++------ src/qemu/qemu_hotplug.c | 20 +++++++++++++++----- tests/virnetdevbandwidthtest.c | 12 ++++++------ 9 files changed, 102 insertions(+), 53 deletions(-) -- 2.0.5

4 7

[libvirt] It does not support hot-plugging VHOST-USER networkcard in libvirt?
by zhang bo 14 Jan '15

14 Jan '15

It works well if we use qmp command to directly interact with qemu 2.1 to hot-add vhost-user network card. However, libvirt seems not support hotplugging well. Steps to use qmp commands to directly(without libvirtd) interact with qemu 2.1 : 1 (qemu) chardev-add backend=socket,id=charnet1,path=/var/run/vhost-user/tap1 2 (qemu) netdev_add vhost-user,id=hostnet1,chardev=charnet1 3 (qemu) device_add virtio-net-pci,netdev=hostnet1,id=net1 In libvirt, we found that, Hot-plug is finished with a success result, but the network card doesn't work afterwards, it could not send/receive IOs. As we further digged into the problem, the 1st step of qmp commands is missing if we use libvirt to do the hot-plug job. 1 (qemu) chardev-add backend=socket,id=charnet1,path=/var/run/vhost-user/tap1 After looking into the function qemuDomainAttachNetDevice(), we found that it does have 1) qemuMonitorAddNetdev() 2) qemuMonitorAddDevice() except for *ChardevAdd*() The question is : why does qemuDomainAttachNetDevice() not do the *ChardevAdd*() work? It seems necessary to add chardev to add a vhost-user network card.

2 1