- Devel - libvirt List Archives

[libvirt] Creating external snapshots with logical volumes.
by Dominique Ramaekers 29 Jun '16

29 Jun '16

Hi, I've already asked this question on the libvirt-users mailing list but got no answer. I think because the answer to my question is not obvious. I hope I get some help here... Currently, for making backups of my guests, I use external snapshots like this: A) virsh snapshot-create-as --domain PCVIRTdra --disk-only --atomic --quiesce --no-metadata --diskspec vda,file=/srv/poolVMS/PCVIRTdra-ONDERHOUD If I wanted to use a logical volume as destination behind the '--diskspec' option, can I do this like this? B) virsh snapshot-create-as --domain PCVIRTdra --disk-only --atomic --quiesce --no-metadata --diskspec vda,file=/dev/CmsrvVH3-vg/PCVIRTdra-ONDERHOUD I got the hit from libvirt itself. If I create a snapshot with command A and the diskspec file already exists, libvirt gives me an error like: " external snapshot file for disk vda already exists and is not a block device: /srv/poolVMS/PCVIRTdra-ONDERHOUD" So the 'and is not a block device' gave me the idea to use a logical volume instead of a file It seems to work, but if this is never intended to be used like this, I would rather not... I ask this question because it seems libvirt is actually recognizing the logical volume as a file if I use command B: root@CmsrvVH3:~# virsh domblklist PCVIRTdra --details Type Device Target Source ------------------------------------------------ file disk vda /dev/CmsrvVH3-vg/PCVIRTdra-ONDERHOUD file cdrom hdb /srv/poolVMS/Windows10_x64.iso file cdrom hdc /srv/poolVMS/virtiodrivers.iso Is this intentional? Is there a better way of doing this? Thanks in advance, Dominique.

1 0

[libvirt] [PATCH] virt-admin: Call virInitialize to fix startup crash
by Cole Robinson 29 Jun '16

29 Jun '16

Similar to what virsh and virt-login-shell do https://bugzilla.redhat.com/show_bug.cgi?id=1350315 --- I can't actually reproduce the bug, the backtrace is similar to 97973ebb7 which added the same fix for virt-login-shell, and that commit also mentions the randomness of reproducing... tools/virt-admin.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/tools/virt-admin.c b/tools/virt-admin.c index 4acac65..7bff5c3 100644 --- a/tools/virt-admin.c +++ b/tools/virt-admin.c @@ -1371,6 +1371,11 @@ main(int argc, char **argv) return EXIT_FAILURE; } + if (virInitialize() < 0) { + vshError(ctl, "%s", _("Failed to initialize libvirt")); + return EXIT_FAILURE; + } + virFileActivateDirOverride(argv[0]); if (!vshInit(ctl, cmdGroups, NULL)) -- 2.7.4

4 10

[libvirt] [PATCH] qemu: Let empty default VNC password work as documented
by Jiri Denemark 29 Jun '16

29 Jun '16

Setting an empty vnc_password in qemu.conf is documented as a way to disable VNC access, but QEMU does not seem to behave like that. Let's enforce the behavior by setting password expiration to "now". Note, this has no effect on setting an empty //graphics@passwd in domain XML. Users may use //graphics@passwdValidTo to enforce the same behavior. https://bugzilla.redhat.com/show_bug.cgi?id=1180092 Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_hotplug.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index e0b8230..91f48dc 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -3970,6 +3970,8 @@ qemuDomainChangeGraphicsPasswords(virQEMUDriverPtr driver, snprintf(expire_time, sizeof(expire_time), "now"); else snprintf(expire_time, sizeof(expire_time), "%lu", (long unsigned)auth->validTo); + } else if (!auth->passwd && defaultPasswd && defaultPasswd[0] == '\0') { + snprintf(expire_time, sizeof(expire_time), "now"); } else { snprintf(expire_time, sizeof(expire_time), "never"); } -- 2.9.0

3 6

[libvirt] [PATCH] qemu: Use proper async job to refresh virtio channels
by Jiri Denemark 29 Jun '16

29 Jun '16

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/qemu/qemu_migration.c | 3 ++- src/qemu/qemu_process.c | 9 ++++++--- src/qemu/qemu_process.h | 3 ++- 3 files changed, 10 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index 83e1bfc..0b1770b 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -6225,7 +6225,8 @@ qemuMigrationFinish(virQEMUDriverPtr driver, if (qemuMigrationStopNBDServer(driver, vm, mig) < 0) goto endjob; - if (qemuRefreshVirtioChannelState(driver, vm) < 0) + if (qemuRefreshVirtioChannelState(driver, vm, + QEMU_ASYNC_JOB_MIGRATION_IN) < 0) goto endjob; if ((rc = qemuConnectAgent(driver, vm)) < 0) { diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 63da600..c8ff587 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -1974,13 +1974,16 @@ qemuProcessRefreshChannelVirtioState(virQEMUDriverPtr driver, int qemuRefreshVirtioChannelState(virQEMUDriverPtr driver, - virDomainObjPtr vm) + virDomainObjPtr vm, + qemuDomainAsyncJob asyncJob) { qemuDomainObjPrivatePtr priv = vm->privateData; virHashTablePtr info = NULL; int ret = -1; - qemuDomainObjEnterMonitor(driver, vm); + if (qemuDomainObjEnterMonitorAsync(driver, vm, asyncJob) < 0) + goto cleanup; + ret = qemuMonitorGetChardevInfo(priv->mon, &info); if (qemuDomainObjExitMonitor(driver, vm) < 0) ret = -1; @@ -3312,7 +3315,7 @@ qemuProcessReconnect(void *opaque) if (qemuProcessRefreshDisks(driver, obj, QEMU_ASYNC_JOB_NONE) < 0) goto error; - if (qemuRefreshVirtioChannelState(driver, obj) < 0) + if (qemuRefreshVirtioChannelState(driver, obj, QEMU_ASYNC_JOB_NONE) < 0) goto error; /* If querying of guest's RTC failed, report error, but do not kill the domain. */ diff --git a/src/qemu/qemu_process.h b/src/qemu/qemu_process.h index 37081ad..21f3b0c 100644 --- a/src/qemu/qemu_process.h +++ b/src/qemu/qemu_process.h @@ -180,7 +180,8 @@ int qemuProcessSetupIOThread(virDomainObjPtr vm, virDomainIOThreadIDDefPtr iothread); int qemuRefreshVirtioChannelState(virQEMUDriverPtr driver, - virDomainObjPtr vm); + virDomainObjPtr vm, + qemuDomainAsyncJob asyncJob); int qemuProcessRefreshBalloonState(virQEMUDriverPtr driver, virDomainObjPtr vm, -- 2.9.0

2 1

[libvirt] [PATCH rfc 0/2] devices: filesystem: type volume
by Olga Krishtal 29 Jun '16

29 Jun '16

In-Reply-To: Patches introduce the new type of filesystem device. The first patch adds the possibility to use storage pool volumes as a backend for filesystem, that is accesed from the guest. Such possibility allows to use volumes for container as a source of root directory. Usage: <devices> <filesystem type='volume'> <source pool='pool_name' volume='volume_name'/> ... <filesystem/> The second patch adds the support of filesystem type volume to virtuozzo containers. There is special type of volume for virtuozzo containers, that can be manipulated via storage api. Now, the volume of this type can be added to ct.xml description as filesystem source. However, there is a small issue with the pool/volume to path translation. Usuallyi, in storage pools, virStorageTranslateDiskSourcePool is used, but in the case of filesystem we do not have virDomainDiskDefPtr and and code deduplication in every driver,that uses this possibility, is unavoidable. eg. virStorageTranslatePoolLocal. Now I am looking for the way to avoid it somehow.

4 8

[libvirt] [PATCH] util: swap After/Before clauses in virSystemdCreateMachine
by Quentin Machu 29 Jun '16

29 Jun '16

An unexpected behavior in systemd has been introduced between the versions 225 and 229: when non-existing units are specified in After/Before clauses, systemd fails if the After clause is listed first. What is expected is that systemd simply ignores these clauses. Because the After clause is specified before the Before clause in virSystemdCreateMachine and because libvirt does not tolerate missing the systemd-machined registration, it leads to fatal boot failures, whereas instances could start properly without these units - only the shutdown workflow could be affected. This fixes the following error in some particular cases: `libvirtError: Cannot set property Before, or unknown property.`. If accepted, this bug fix should be backported. https://bugzilla.redhat.com/show_bug.cgi?id=1350909 --- src/util/virsystemd.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/util/virsystemd.c b/src/util/virsystemd.c index 871db7e..3a146f8 100644 --- a/src/util/virsystemd.c +++ b/src/util/virsystemd.c @@ -365,8 +365,8 @@ int virSystemdCreateMachine(const char *name, nnicindexes, nicindexes, 3, "Slice", "s", slicename, - "After", "as", 1, "libvirtd.service", - "Before", "as", 1, "libvirt-guests.service") < 0) + "Before", "as", 1, "libvirt-guests.service", + "After", "as", 1, "libvirtd.service") < 0) goto cleanup; if (error.level == VIR_ERR_ERROR) { @@ -406,8 +406,8 @@ int virSystemdCreateMachine(const char *name, rootdir ? rootdir : "", 3, "Slice", "s", slicename, - "After", "as", 1, "libvirtd.service", - "Before", "as", 1, "libvirt-guests.service") < 0) + "Before", "as", 1, "libvirt-guests.service", + "After", "as", 1, "libvirtd.service") < 0) goto cleanup; } -- 1.8.3.1

2 1

[libvirt] [PATCH] Fix possible invalid read in adminClientGetInfo
by Ján Tomko 29 Jun '16

29 Jun '16

virNetServerClientGetInfo returns the client's remote address as a string, which is a part of the client object. Use VIR_STRDUP to make a copy which can be freely accessed even after the virNetServerClient object is unlocked. --- daemon/admin_server.c | 3 ++- src/rpc/virnetserverclient.c | 8 ++++++-- src/rpc/virnetserverclient.h | 2 +- 3 files changed, 9 insertions(+), 4 deletions(-) diff --git a/daemon/admin_server.c b/daemon/admin_server.c index cb9079c..9f24f68 100644 --- a/daemon/admin_server.c +++ b/daemon/admin_server.c @@ -221,7 +221,7 @@ adminClientGetInfo(virNetServerClientPtr client, int ret = -1; int maxparams = 0; bool readonly; - const char *sock_addr = NULL; + char *sock_addr = NULL; const char *attr = NULL; virTypedParameterPtr tmpparams = NULL; virIdentityPtr identity = NULL; @@ -300,6 +300,7 @@ adminClientGetInfo(virNetServerClientPtr client, cleanup: virObjectUnref(identity); + VIR_FREE(sock_addr); return ret; } diff --git a/src/rpc/virnetserverclient.c b/src/rpc/virnetserverclient.c index 39fe860..81da82c 100644 --- a/src/rpc/virnetserverclient.c +++ b/src/rpc/virnetserverclient.c @@ -1606,20 +1606,24 @@ virNetServerClientGetTransport(virNetServerClientPtr client) int virNetServerClientGetInfo(virNetServerClientPtr client, - bool *readonly, const char **sock_addr, + bool *readonly, char **sock_addr, virIdentityPtr *identity) { int ret = -1; + const char *addr; virObjectLock(client); *readonly = client->readonly; - if (!(*sock_addr = virNetServerClientRemoteAddrStringURI(client))) { + if (!(addr = virNetServerClientRemoteAddrStringURI(client))) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("No network socket associated with client")); goto cleanup; } + if (VIR_STRDUP(*sock_addr, addr) < 0) + goto cleanup; + if (!client->identity) { virReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("No identity information available for client")); diff --git a/src/rpc/virnetserverclient.h b/src/rpc/virnetserverclient.h index c243a68..a53cc00 100644 --- a/src/rpc/virnetserverclient.h +++ b/src/rpc/virnetserverclient.h @@ -149,7 +149,7 @@ bool virNetServerClientNeedAuth(virNetServerClientPtr client); int virNetServerClientGetTransport(virNetServerClientPtr client); int virNetServerClientGetInfo(virNetServerClientPtr client, - bool *readonly, const char **sock_addr, + bool *readonly, char **sock_addr, virIdentityPtr *identity); #endif /* __VIR_NET_SERVER_CLIENT_H__ */ -- 2.7.3

2 1

[libvirt] [PATCH 0/2] qemu: Use bootindex whenever possible
by Jiri Denemark 29 Jun '16

29 Jun '16

Jiri Denemark (2): qemu: Use bootindex whenever possible qemu: Drop emitBootindex parameter src/qemu/qemu_command.c | 89 ++++++++++------------ .../qemuxml2argv-boot-menu-enable-bootindex.args | 23 ++++++ .../qemuxml2argv-boot-menu-enable-bootindex.xml | 28 +++++++ tests/qemuxml2argvtest.c | 2 +- 4 files changed, 91 insertions(+), 51 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-boot-menu-enable-bootindex.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-boot-menu-enable-bootindex.xml -- 2.9.0

2 6

[libvirt] [PATCH] conf: Allocate structure for 'perf' events in virDomainDefNew
by Peter Krempa 29 Jun '16

29 Jun '16

Some code paths already assume that it is allocated since it was always allocated by virDomainPerfDefParseXML. Move allocation to virDomainDefNew to handle cases that don't go through the XML parser. This fixes crash when attempting to connect to an existing process via virDomainQemuAttach. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1350688 --- src/conf/domain_conf.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index ef266af..fb622cd 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -2746,6 +2746,9 @@ virDomainDefNew(void) if (!(ret->numa = virDomainNumaNew())) goto error; + if (VIR_ALLOC(ret->perf) < 0) + goto error; + ret->mem.hard_limit = VIR_DOMAIN_MEMORY_PARAM_UNLIMITED; ret->mem.soft_limit = VIR_DOMAIN_MEMORY_PARAM_UNLIMITED; ret->mem.swap_hard_limit = VIR_DOMAIN_MEMORY_PARAM_UNLIMITED; @@ -13136,9 +13139,6 @@ virDomainPerfDefParseXML(virDomainDefPtr def, if ((n = virXPathNodeSet("./perf/event", ctxt, &nodes)) < 0) return n; - if (VIR_ALLOC(def->perf) < 0) - goto cleanup; - for (i = 0; i < n; i++) { if (virDomainPerfEventDefParseXML(def->perf, nodes[i]) < 0) goto cleanup; @@ -13147,8 +13147,6 @@ virDomainPerfDefParseXML(virDomainDefPtr def, ret = 0; cleanup: - if (ret < 0) - VIR_FREE(def->perf); VIR_FREE(nodes); return ret; } @@ -23388,8 +23386,7 @@ virDomainDefFormatInternal(virDomainDefPtr def, virBufferAddLit(buf, "</pm>\n"); } - if (def->perf) - virDomainPerfDefFormat(buf, def->perf); + virDomainPerfDefFormat(buf, def->perf); virBufferAddLit(buf, "<devices>\n"); virBufferAdjustIndent(buf, 2); -- 2.9.0

2 2

[libvirt] [PATCH 0/6] events: don't mandate a uuid
by Cole Robinson 28 Jun '16

28 Jun '16

The generic event infrastructure has the notion of lookup-by-uuid baked into the API, but not all objects have a uuid (nodedev, interfaces). This series enables callers to specify a string key for match purposes. Existing users of the binary uuid value now pass in a converted uuid string. Cole Robinson (6): events: Privatize virObjectEventCallback events: Add virObjectEventCallbackFree events: Cleanup callback variable name events: Pass in UUID as a string events: Add explicit lookup 'key' value events: Rename argument uuid->key src/conf/domain_event.c | 28 +++++-- src/conf/network_event.c | 16 +++- src/conf/object_event.c | 170 +++++++++++++++++++++------------------- src/conf/object_event.h | 3 - src/conf/object_event_private.h | 8 +- src/conf/storage_event.c | 16 +++- 6 files changed, 145 insertions(+), 96 deletions(-) -- 2.7.4

2 9