- Devel - libvirt List Archives

[libvirt] [PATCH] Ignore additional fields in iscsiadm output
by Ján Tomko 21 Feb '14

21 Feb '14

There has been a new field introduced in iscsiadm --mode session output [1], but our regex only expects four fields. This breaks startup of iscsi pools: error: Failed to start pool iscsi error: internal error: cannot find session Fix this by ignoring anything after the fourth field. https://bugzilla.redhat.com/show_bug.cgi?id=1067173 [1] https://github.com/mikechristie/open-iscsi/commit/181af9a --- src/storage/storage_backend_iscsi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/storage/storage_backend_iscsi.c b/src/storage/storage_backend_iscsi.c index 556c2cc..1149e43 100644 --- a/src/storage/storage_backend_iscsi.c +++ b/src/storage/storage_backend_iscsi.c @@ -102,7 +102,7 @@ virStorageBackendISCSISession(virStoragePoolObjPtr pool, * Pull out 2nd and 4th fields */ const char *regexes[] = { - "^tcp:\\s+\\[(\\S+)\\]\\s+\\S+\\s+(\\S+)\\s*$" + "^tcp:\\s+\\[(\\S+)\\]\\s+\\S+\\s+(\\S+).*$" }; int vars[] = { 2, -- 1.8.3.2

2 2

[libvirt] [PATCH] Add a stub for virCgroupGetDomainTotalCpuStats
by Ján Tomko 21 Feb '14

21 Feb '14

Commit 6515889 broke the build on FreeBSD: In function `qemuDomainGetCPUStats': /../../src/qemu/qemu_driver.c:16102: undefined reference to `virCgroupGetDomainTotalCpuStats' --- Pushed as a build-breaker. src/util/vircgroup.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/src/util/vircgroup.c b/src/util/vircgroup.c index 268a4ae..2dba10c 100644 --- a/src/util/vircgroup.c +++ b/src/util/vircgroup.c @@ -4261,6 +4261,17 @@ virCgroupGetCpuacctStat(virCgroupPtr group ATTRIBUTE_UNUSED, int +virCgroupGetDomainTotalCpuStats(virCgroupPtr group ATTRIBUTE_UNUSED, + virTypedParameterPtr params ATTRIBUTE_UNUSED, + int nparams ATTRIBUTE_UNUSED) +{ + virReportSystemError(ENOSYS, "%s", + _("Control groups not supported on this platform")); + return -1; +} + + +int virCgroupSetFreezerState(virCgroupPtr group ATTRIBUTE_UNUSED, const char *state ATTRIBUTE_UNUSED) { -- 1.8.3.2

1 0

[libvirt] [PATCHv4 00/10] Add BlkIO and CPU/mem stat API implementations for lxc
by Thorsten Behrens 21 Feb '14

21 Feb '14

This patch set adds block io, memory and domain cpu statistics API slot implementations to the LXC driver, in order to get linux container monitoring and accounting a bit closer to qemu standards. The last patch is a tad quirky (happy to hear suggestions on alternative ways), in that it widens the permissible value set at the .domainBlockStats slot: for lxc guests, it is relatively likely to have zero disk devices, since host filesystems can be used via passthrough bind mounts. Therefore, passing the zero-length string as device path, is interpreted as 'return summary stats for the entire domains's block io'. v4 addresses the last remaining review comments. Thorsten Behrens (10): Add util virCgroupGetBlkioIo*Serviced methods. Implement domainMemoryStats API slot for LXC driver. Make qemuGetDomainTotalCPUStats a virCgroup function. Implement domainGetCPUStats for lxc driver. Implement lxcDomainBlockStats* for lxc driver Widening API change - accept empty path for virDomainBlockStats Add unit test for virCgroupGetBlkioIo*Serviced Add unit test for virCgroupGetMemoryUsage. Fix misspelled cpuacct.usage_percpu in cgroup mock. Add unit test for virCgroupGetPercpuStats. src/libvirt.c | 8 +- src/libvirt_private.syms | 4 + src/lxc/lxc_driver.c | 300 +++++++++++++++++++++++++++++++++ src/qemu/qemu_driver.c | 54 +----- src/util/vircgroup.c | 382 +++++++++++++++++++++++++++++++++++++++++++ src/util/vircgroup.h | 24 +++ tests/testutilslxc.h | 3 + tests/vircgroupmock.c | 100 ++++++++++- tests/vircgrouptest.c | 230 ++++++++++++++++++++++++++ tools/virsh-domain-monitor.c | 11 +- tools/virsh.pod | 5 +- 11 files changed, 1059 insertions(+), 62 deletions(-) -- 1.8.4.5

4 21

[libvirt] [libvirt-python PATCH] generator: Skip exporting only sentinels
by Martin Kletzander 21 Feb '14

21 Feb '14

When enum type has '_LAST' in its name, but is not the last type in that enum, it's skipped even though it shouldn't be. Currently, this is the case for only VIR_NETWORK_UPDATE_COMMAND_ADD_LAST inside an enum virNetworkUpdateCommand. Also, since _LAST types can have other enums instead of values, that needs to be filtered out using a try-except when converting the value. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- generator.py | 19 ++++++++++++++----- 1 file changed, 14 insertions(+), 5 deletions(-) diff --git a/generator.py b/generator.py index 42f3913..0e9600f 100755 --- a/generator.py +++ b/generator.py @@ -227,10 +227,9 @@ def enum(type, name, value): value = 1 elif value == 'VIR_DOMAIN_AFFECT_CONFIG': value = 2 - if name[-5:] != '_LAST': - if onlyOverrides and name not in enums[type]: - return - enums[type][name] = value + if onlyOverrides and name not in enums[type]: + return + enums[type][name] = value def lxc_enum(type, name, value): if type not in lxc_enums: @@ -1765,13 +1764,23 @@ def buildWrappers(module): # # Generate enum constants # + def enumsSortKey(data): + value = data[1] + try: + value = int(value) + except ValueError: + value = float('inf') + return value + enumvals = list(enums.items()) if enumvals is not None: enumvals.sort(key=lambda x: x[0]) for type,enum in enumvals: classes.write("# %s\n" % type) items = list(enum.items()) - items.sort(key=lambda i: int(i[1])) + items.sort(key=enumsSortKey) + if items[-1][0].endswith('_LAST'): + del items[-1] for name,value in items: classes.write("%s = %s\n" % (name,value)) classes.write("\n") -- 1.9.0

2 2

[libvirt] [PATCH] virsh: fix memory leak in virsh when starting a guest with invalid fd
by Jincheng Miao 21 Feb '14

21 Feb '14

When start a guest with --pass-fd, if the argument of --pass-fd is invalid, virsh will exit, but doesn't free the variable 'dom'. The valgrind said: ... ==24569== 63 (56 direct, 7 indirect) bytes in 1 blocks are definitely lost in loss record 130 of 234 ==24569== at 0x4C2A1D4: calloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so) ==24569== by 0x4E879A4: virAllocVar (viralloc.c:544) ==24569== by 0x4EBD625: virObjectNew (virobject.c:190) ==24569== by 0x4F3A18A: virGetDomain (datatypes.c:226) ==24569== by 0x4F9311F: remoteDomainLookupByName (remote_driver.c:6636) ==24569== by 0x4F44F20: virDomainLookupByName (libvirt.c:2277) ==24569== by 0x12F616: vshCommandOptDomainBy (virsh-domain.c:105) ==24569== by 0x131C79: cmdStart (virsh-domain.c:3330) ==24569== by 0x12C4AB: vshCommandRun (virsh.c:1752) ==24569== by 0x127001: main (virsh.c:3218) https://bugzilla.redhat.com/show_bug.cgi?id=1067338 Signed-off-by: Jincheng Miao <jmiao(a)redhat.com> --- tools/virsh-domain.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index c3db94c..57653a2 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -3428,11 +3428,11 @@ cmdStart(vshControl *ctl, const vshCmd *cmd) if (virDomainGetID(dom) != (unsigned int)-1) { vshError(ctl, "%s", _("Domain is already active")); virDomainFree(dom); - return false; + goto cleanup; } if (cmdStartGetFDs(ctl, cmd, &nfds, &fds) < 0) - return false; + goto cleanup; if (vshCommandOptBool(cmd, "paused")) flags |= VIR_DOMAIN_START_PAUSED; -- 1.8.5.3

2 2

[libvirt] [PATCH] maint: update to latest gnulib
by Eric Blake 20 Feb '14

20 Feb '14

Among other things, gnulib now allows overriding of sed during 'make syntax-check'. * .gnulib: Update to latest, for maint.mk improvements. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- Pushing under the maintainer rule, before we freeze. I still need to revive Roman's work on using $(SED) and/or portable sed constructs in our cfg.mk, but that can occur after freeze if necessary, while the hard part of the gnulib submodule should be done before freeze. * .gnulib d5fec6c...3f51bf4 (9): > maintainer-makefiles: use $(SED) for syntax check > update from texinfo > savedir: add sorting arg to savedir, streamsavedir; remove fdsavedir > autoupdate > update from texinfo > update from texinfo > file-type: add support for doors and other less-common file types > update from texinfo > autoupdate .gnulib | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/.gnulib b/.gnulib index d5fec6c..3f51bf4 160000 --- a/.gnulib +++ b/.gnulib @@ -1 +1 @@ -Subproject commit d5fec6c22f03c6a73d62260c9ce091c10c0a9cbd +Subproject commit 3f51bf41c8be8b310f57caff371377414701d5cc -- 1.8.5.3

1 0

Re: [libvirt] [libvirt-users] event-test.py cannot detects domain shutdown
by Cole Robinson 20 Feb '14

20 Feb '14

On 02/18/2014 05:12 AM, Kim Larry wrote: > 2014년 2월 11일 오후 10:00에 Eric Blake <eblake(a)redhat.com> 작성: > >> On 02/11/2014 12:20 AM, Kim Larry wrote: >> >>>>> libvirtd version is 1.1.4 and using Xen for hypervisor. >>>> I'm not as familiar with the xen hypervisor as with qemu; it may just be >>>> the case that no one has coded the xen driver to send a shutdown event. >>>> At any rate, I can confirm that with the qemu driver, shutdown events >>>> are delivered. >> >>> It is a bit weird that all except shutdown event are coded... >>> If it is true, do I have to just wait? >> >> https://rwmj.wordpress.com/2010/11/08/want-help-dont-email-me-directly/ >> >> Or you could look into the source code and provide a patch; but >> discussing this on the list is a better way to get a reaction from >> someone familiar with the xen code. >> >> -- >> Eric Blake eblake redhat com +1-919-301-3266 >> Libvirt virtualization library http://libvirt.org > > > Ah, I forgot to add libvirt-users in Cc... sorry. > > The thing I found today is that if libvirt uses xend driver, shutdown events > are delivered, but if libvirt uses libxl drvier, doesn't show up anything. It > seems there are bugs on shutdown event, so I did dig into the libvirt source > briefly, but I couldn't find where libxl driver calls event callback. > > Any help will be greatly appreciated. Especially someone who is familiar with > xen code. > Jim, does this sound familiar? - Cole

2 4

[libvirt] [PATCH] libxl: always use libxlVmCleanupJob in shutdown thread
by Jim Fehlig 20 Feb '14

20 Feb '14

Commit e4a0e900 missed calling libxlVmCleanupJob in the shutdown handler when processing a reboot event. Signed-off-by: Jim Fehlig <jfehlig(a)suse.com> --- src/libxl/libxl_driver.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index f7ca91c..9f87e71 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -393,7 +393,7 @@ libxlDomainShutdownThread(void *opaque) break; case LIBXL_SHUTDOWN_REASON_REBOOT: libxl_domain_destroy(ctx, vm->def->id, NULL); - libxlVmCleanup(driver, vm, VIR_DOMAIN_SHUTOFF_SHUTDOWN); + libxlVmCleanupJob(driver, vm, VIR_DOMAIN_SHUTOFF_SHUTDOWN); libxlVmStart(driver, vm, 0, -1); break; default: -- 1.8.1.4

2 2

[libvirt] [PATCH] qemu: adjust maxmem/maxvcpu computation
by Eric Blake 20 Feb '14

20 Feb '14

https://bugzilla.redhat.com/show_bug.cgi?id=1038363 If a domain has a different maximum for persistent and live maxmem or max vcpus, then it is possible to hit cases where libvirt refuses to adjust the current values or gets halfway through the adjustment before failing. Better is to determine up front if the change is possible for all requested flags. * src/qemu/qemu_driver.c (qemuDomainSetMemoryFlags): Compute correct maximum if both live and config are being set. (qemuDomainSetVcpusFlags): Likewise. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- src/qemu/qemu_driver.c | 21 ++++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index a0b3ce3..288b98b 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2242,8 +2242,16 @@ static int qemuDomainSetMemoryFlags(virDomainPtr dom, unsigned long newmem, } else { /* resize the current memory */ + unsigned long oldmax = 0; - if (newmem > vm->def->mem.max_balloon) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) + oldmax = vm->def->mem.max_balloon; + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { + if (!oldmax || oldmax > persistentDef->mem.max_balloon) + oldmax = persistentDef->mem.max_balloon; + } + + if (newmem > oldmax) { virReportError(VIR_ERR_INVALID_ARG, "%s", _("cannot set memory higher than max memory")); goto endjob; @@ -4121,6 +4129,7 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, virDomainDefPtr persistentDef; int ret = -1; bool maximum; + unsigned int maxvcpus = 0; virQEMUDriverConfigPtr cfg = NULL; virCapsPtr caps = NULL; qemuAgentCPUInfoPtr cpuinfo = NULL; @@ -4168,11 +4177,17 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, goto endjob; } - if (!maximum && nvcpus > vm->def->maxvcpus) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) + maxvcpus = vm->def->maxvcpus; + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { + if (!maxvcpus || maxvcpus > persistentDef->maxvcpus) + maxvcpus = persistentDef->maxvcpus; + } + if (!maximum && nvcpus > maxvcpus) { virReportError(VIR_ERR_INVALID_ARG, _("requested vcpus is greater than max allowable" " vcpus for the domain: %d > %d"), - nvcpus, vm->def->maxvcpus); + nvcpus, maxvcpus); goto endjob; } -- 1.8.5.3

2 3

[libvirt] [PATCH 0/4] add 'virsh event' and friends
by Eric Blake 20 Feb '14

20 Feb '14

Inspired by my work on qemu monitor events; Dan correctly argued that if we're going to expose those through virsh, we also need to expose regular events. Patch 5/4 still coming, which adds handlers for the remaining 15 domain event types. Eric Blake (4): virsh: common code for parsing --seconds virsh: common code for waiting for an event virsh: add event command, for lifecycle events virsh: add net-event command tools/virsh-domain.c | 410 +++++++++++++++++++++++++++++++++++++++++++------- tools/virsh-network.c | 171 ++++++++++++++++++++- tools/virsh.c | 186 ++++++++++++++++++++++- tools/virsh.h | 17 ++- tools/virsh.pod | 30 ++++ 5 files changed, 752 insertions(+), 62 deletions(-) -- 1.8.5.3

2 12