Cédric Bosdonnat [Tue, 25 Mar 2014 11:48:26 +0000 (12:48 +0100)]
Fix apparmor profile to make vfio pci passthrough work
See lp#1276719 for the bug description. As virt-aa-helper doesn't know
the VFIO groups to use for the guest, allow access to all
/dev/vfio/[0-9]* and /dev/vfio/vfio files if there is a potential need
for vfio
Fix Memory Leak in virQEMUCapsInitGuestFromBinary()
While running qemucaps2xmltest, it was found that valgrind pointed out
the following memory leaks:
==29896== 0 bytes in 1 blocks are definitely lost in loss record 1 of 65
==29896== at 0x4A0577B: calloc (vg_replace_malloc.c:593)
==29896== by 0x4C6B45E: virAllocN (viralloc.c:191)
==29896== by 0x4232A9: virQEMUCapsGetMachineTypesCaps (qemu_capabilities.c:1999)
==29896== by 0x4234E7: virQEMUCapsInitGuestFromBinary (qemu_capabilities.c:789)
==29896== by 0x41F10B: testQemuCapsXML (qemucaps2xmltest.c:118)
==29896== by 0x41FFD1: virtTestRun (testutils.c:201)
==29896== by 0x41EE7A: mymain (qemucaps2xmltest.c:203)
==29896== by 0x42074D: virtTestMain (testutils.c:789)
==29896== by 0x3E6CE1ED1C: (below main) (libc-start.c:226)
==29896==
==29896== 0 bytes in 1 blocks are definitely lost in loss record 2 of 65
==29896== at 0x4A0577B: calloc (vg_replace_malloc.c:593)
==29896== by 0x4C6B45E: virAllocN (viralloc.c:191)
==29896== by 0x4232A9: virQEMUCapsGetMachineTypesCaps (qemu_capabilities.c:1999)
==29896== by 0x4234E7: virQEMUCapsInitGuestFromBinary (qemu_capabilities.c:789)
==29896== by 0x41F10B: testQemuCapsXML (qemucaps2xmltest.c:118)
==29896== by 0x41FFD1: virtTestRun (testutils.c:201)
==29896== by 0x41EEA3: mymain (qemucaps2xmltest.c:204)
==29896== by 0x42074D: virtTestMain (testutils.c:789)
==29896== by 0x3E6CE1ED1C: (below main) (libc-start.c:226)
Li Yang [Thu, 27 Mar 2014 06:10:57 +0000 (14:10 +0800)]
Modify help information of virsh list command
Use 'virsh list domain --title' option can get domain's title,
not description, the original help information 'show short
domain description' will confuse users, so modify it to
'show domain title'
Signed-off-by: Li Yang <liyang.fnst@cn.fujitsu.com>
While running qemucaps2xmltest, it was found that valgrind pointed out
the following memory leaks:
==27045== 160 (112 direct, 48 indirect) bytes in 1 blocks are definitely lost in loss record 51 of 65
==27045== at 0x4A0577B: calloc (vg_replace_malloc.c:593)
==27045== by 0x4C6BACD: virAllocVar (viralloc.c:560)
==27045== by 0x4CAF095: virObjectNew (virobject.c:193)
==27045== by 0x421453: virQEMUCapsNew (qemu_capabilities.c:1805)
==27045== by 0x41F04F: testQemuCapsXML (qemucaps2xmltest.c:72)
==27045== by 0x41FFD1: virtTestRun (testutils.c:201)
==27045== by 0x41EE7A: mymain (qemucaps2xmltest.c:203)
==27045== by 0x42074D: virtTestMain (testutils.c:789)
==27045== by 0x3E6CE1ED1C: (below main) (libc-start.c:226)
==27045==
==27045== 160 (112 direct, 48 indirect) bytes in 1 blocks are definitely lost in loss record 52 of 65
==27045== at 0x4A0577B: calloc (vg_replace_malloc.c:593)
==27045== by 0x4C6BACD: virAllocVar (viralloc.c:560)
==27045== by 0x4CAF095: virObjectNew (virobject.c:193)
==27045== by 0x421453: virQEMUCapsNew (qemu_capabilities.c:1805)
==27045== by 0x41F04F: testQemuCapsXML (qemucaps2xmltest.c:72)
==27045== by 0x41FFD1: virtTestRun (testutils.c:201)
==27045== by 0x41EEA3: mymain (qemucaps2xmltest.c:204)
==27045== by 0x42074D: virtTestMain (testutils.c:789)
==27045== by 0x3E6CE1ED1C: (below main) (libc-start.c:226)
To ease mocking for bhyve unit tests move virBhyveTapGetRealDeviceName()
out of bhyve_command.c to virnetdevtap and rename it to
virNetDevTapGetRealDeviceName().
When searching for that message later, I found another series that had
been posted by Guannan Ren back in 2012 that somehow slipped between
the cracks:
That patch was very much out of date, but also pointed out some real
problems.
This patch fixes all the noted problems by refactoring
virNetworkDNSSrvDefParseXML() and networkDnsmasqConfContents(), then
verifies those fixes by added several new records to the test case.
Problems fixed:
* both service and protocol now have an underscore ("_") prepended on
the commandline, as required by RFC2782.
* when optional attributes aren't specified, the separating comma is
also now not placed on the dnsmasq commandline. If optional
attributes in the middle of the line are not specified, they are
replaced with a default value in the commandline (1 for port, 0 for
priority and weight).
(actually the would have generated an error, because "optional"
attributes weren't really optional.)
* The allowed characters for both service and protocol are now limited
to alphanumerics, plus a few special characters that are found in
existing names in /etc/services and /etc/protocols. (One exception
is that both of these files contain names with an embedded ".", but
"." can't be used in these fields of an SRV record because it is
used as a field separator and there is no method to escape a "."
into a field.) (Previously only the strings "tcp" and "udp" were
allowed for protocol, but this restriction has been removed, since
RFC2782 specifically says that it isn't limited to those, and that
anyway it is case insensitive.)
* the "domain" attribute is no longer required in order to recognize
the port, priority, and weight attributes during parsing. Only
"target" is required for this.
* if "target" isn't specified, port, priority, and weight are not
allowed (since they are meaningless - an empty target means "this
service is *not available* for this domain").
* port, priority, and weight are now truly optional, as the comments
originally suggested, but which was not actually true.
Michal Privoznik [Wed, 19 Mar 2014 14:46:07 +0000 (15:46 +0100)]
qemuDomainAttachDeviceFlags: Parse device xml as inactive
In all other drivers we are doing so. Moreover, we don't want to parse
runtime information in attach (even if the attach is meant as live)
because we are generating the runtime info ourselves. We can't trust
users they supply sane values anyway.
==1140== 9 bytes in 1 blocks are definitely lost in loss record 72 of 1,151
==1140== at 0x4A06C2B: malloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==1140== by 0x623C758: xmlStrndup (in /usr/lib64/libxml2.so.2.9.1)
==1140== by 0x50FD763: virXMLPropString (virxml.c:483)
==1140== by 0x510F8B7: virDomainDeviceInfoParseXML (domain_conf.c:3685)
==1140== by 0x511ACFD: virDomainChrDefParseXML (domain_conf.c:7535)
==1140== by 0x5121D13: virDomainDeviceDefParse (domain_conf.c:9918)
==1140== by 0x13AE6313: qemuDomainAttachDeviceFlags (qemu_driver.c:6926)
==1140== by 0x13AE65FA: qemuDomainAttachDevice (qemu_driver.c:7005)
==1140== by 0x51C77DA: virDomainAttachDevice (libvirt.c:10231)
==1140== by 0x127FDD: remoteDispatchDomainAttachDevice (remote_dispatch.h:2404)
==1140== by 0x127EC5: remoteDispatchDomainAttachDeviceHelper (remote_dispatch.h:2382)
==1140== by 0x5241F81: virNetServerProgramDispatchCall (virnetserverprogram.c:437)
When doing live attach, we are passing the inactive definition anyway
since we are passing the result of virDomainDeviceDefCopy() which does
inactive copy by default.
Moreover, we are doing the same mistake in qemuhotplugtest.
Just a side note - it makes perfect sense to parse the runtime info
like alias in qemuDomainDetachDevice and qemuDomainUpdateDeviceFlags()
as in some cases the only difference to distinguish two devices can be
just their alias.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Francesco Romani [Mon, 17 Mar 2014 15:19:44 +0000 (16:19 +0100)]
qemu: export disk snapshot support in capabilities
This patch adds an element to QEMU's capability XML, to
show if the underlying QEMU binary supports the live disk
snapshotting or not.
This allows any client to know ahead of time if the feature
is available.
Without this information available, the only way to check
for the snapshot support is to request one and check for
errors.
Signed-off-by: Francesco Romani <fromani@redhat.com>
John Ferlan [Tue, 25 Mar 2014 17:03:19 +0000 (13:03 -0400)]
Coverity: Resolve a FORWARD_NULL
Recent changes in the module seemed to have caused Coverity to reanalyze
certain parts of the code. Previously the code was modified via commit
id '11a11812' to resolve a different error (perhaps DEADCODE). Up through
commit id '7b3f1f8c' there were no issues.
The new error indicats the 'outbuf' was checked for NULL and then complains
because of the dereference. Adding checks for non-NULL prior to the deref
resulted in a DEADCODE message.
So, resolve using an sa_assert() to keep Coverity quiet especially since
it doesn't understand that outbuf will change as a result of a successful
virCommandRun() call.
John Ferlan [Tue, 25 Mar 2014 16:57:58 +0000 (12:57 -0400)]
Coverity: Resolve a CHECKED_RETURN message
Recent changes to the module seemed to have caused Coverity to find a new
issue regarding the failure to check the return from a sendmsg. The code
doesn't seem to care about the return status, so just added an ignore_value
to keep Coverity quiet.
Stefan Bader [Tue, 25 Mar 2014 15:28:36 +0000 (16:28 +0100)]
libxl: Create log directory earlier
Commit d9f19c30d054c86b15a304f4118baa4fa75af9d2 moved a lot of the
configuration setup into libxlDriverConfigNew().
However that tries to create the libxl/libxl-driver.log before the
libxl directory gets created in libxlStateInitialize().
This causes the daemon to fail on systems that have not had the directory
created before.
Move the code to create the libxl directory into libxlDriverConfigNew().
Signed-off-by: Stefan Bader <stefan.bader@canonical.com>
When lock failure is detected by sanlock, our sanlock_helper kill script
will try to restart (shutdown followed by start) the affected domain
when RESTART action is configured for it. While shutting down kills QEMU
and removes all its leases (which is what sanlock wants to happen),
trying to start it again just hangs because libvirt tries reacquire the
locks in the failed lock space. Hence, this action cannot be supported
by sanlock driver.
Sanlock expects that the configured kill script either kills the PID on
lock failure or removes all locks the PID owns. If none of the two
options happen, sanlock will reboot the host. Although IGNORE action is
supposed to ignore the request to kill the PID or remove all leases,
it's certainly not designed to cause the host to be rebooted. That said,
IGNORE action is incompatible with sanlock and should be forbidden by
libvirt.
Peter Krempa [Tue, 25 Mar 2014 07:17:27 +0000 (08:17 +0100)]
util: Sanitize ATTRIBUTE_NONNULL use in viriscsi.h
Some of the function attributes marked as nonnull actually explicitly
handle the arguments for NULL. All changed functions handle missing
"initiatoriqn" argument well and virISCSIScanTargets also handles well
if the return pointers are missing. Remove some of the liberaly used
ATTRIBUTE_NONNULLs as coverity and possibly other compilers that honor
the attribute fail to compile the code.
Qiao Nuohan [Sun, 23 Mar 2014 03:51:12 +0000 (11:51 +0800)]
add new virDomainCoreDumpWithFormat API
--memory-only option is introduced without compression supported. Now qemu
has support for dumping domain's memory in kdump-compressed format. This
patch adds a new virDomainCoreDumpWithFormat API, so that the format in
which qemu dumps domain's memory can be specified.
Signed-off-by: Qiao Nuohan <qiaonuohan@cn.fujitsu.com> Signed-off-by: Eric Blake <eblake@redhat.com>
Eric Blake [Wed, 19 Mar 2014 17:11:16 +0000 (11:11 -0600)]
conf: prepare to track multiple host source files per <disk>
It's finally time to start tracking disk backing chains in
<domain> XML. The first step is to start refactoring code
so that we have an object more convenient for representing
each host source resource in the context of a single guest
<disk>. Ultimately, I plan to move the new type into src/util
where it can be reused by virStorageFile, but to make the
transition easier to review, this patch just creates the
new type then fixes everything until it compiles again.
Eric Blake [Mon, 17 Mar 2014 18:53:15 +0000 (12:53 -0600)]
conf: use disk source accessors in conf/
Part of a series of cleanups to use new accessor methods.
Several places in domain_conf.c still open-code raw field access,
but that code will be touched later with the diskDef struct split
so I'm avoiding churn here.
Eric Blake [Mon, 17 Mar 2014 17:39:57 +0000 (11:39 -0600)]
conf: accessors for common source information
A future patch will split virDomainDiskDef, in order to track
multiple host resources per guest <disk>. To reduce the size
of that patch, I've factored out the four most common accesses
into functions, so that I can incrementally upgrade the code
base to use the accessors, and so that code that doesn't care
about the distinction of per-file details won't have to be
changed when the struct changes.
Eric Blake [Tue, 18 Mar 2014 21:41:28 +0000 (15:41 -0600)]
vmware: fix parse of disk source
While writing disk source refactoring, I discovered that conversion
from XML to vmware modified the disk source in place; if the same
code is reached twice, the second call behaves differently because
the first call didn't clean up its mess.
I get the SIGSEGV when starting the domain. The thing is, when
starting a domain, we check for its disk presence. For some reason,
when determining the disk chain, we parse the <seclabel/> (don't ask
me why). However, there's no label attribute in the XML, so we end up
calling virParseOwnershipIds() over NULL string:
[Switching to Thread 0x7ffff10c4700 (LWP 30956)]
__strchr_sse42 () at ../sysdeps/x86_64/multiarch/strchr.S:136
136 ../sysdeps/x86_64/multiarch/strchr.S: No such file or directory.
(gdb) bt
#0 __strchr_sse42 () at ../sysdeps/x86_64/multiarch/strchr.S:136
#1 0x00007ffff749f800 in virParseOwnershipIds (label=0x0, uidPtr=uidPtr@entry=0x7ffff10c2df0, gidPtr=gidPtr@entry=0x7ffff10c2df4) at util/virutil.c:2115
#2 0x00007fffe929f006 in qemuDomainGetImageIds (gid=0x7ffff10c2df4, uid=0x7ffff10c2df0, disk=0x7fffe40cb000, vm=0x7fffe40a6410, cfg=0x7fffe409ae00) at qemu/qemu_domain.c:2385
#3 qemuDomainDetermineDiskChain (driver=driver@entry=0x7fffe40120e0, vm=vm@entry=0x7fffe40a6410, disk=disk@entry=0x7fffe40cb000, force=force@entry=false) at qemu/qemu_domain.c:2414
#4 0x00007fffe929f128 in qemuDomainCheckDiskPresence (driver=driver@entry=0x7fffe40120e0, vm=vm@entry=0x7fffe40a6410, cold_boot=cold_boot@entry=true) at qemu/qemu_domain.c:2250
#5 0x00007fffe92b6fc8 in qemuProcessStart (conn=conn@entry=0x7fffd4000b60, driver=driver@entry=0x7fffe40120e0, vm=vm@entry=0x7fffe40a6410, migrateFrom=migrateFrom@entry=0x0, stdin_fd=stdin_fd@entry=-1, stdin_path=stdin_path@entry=0x0, snapshot=snapshot@entry=0x0,
vmop=vmop@entry=VIR_NETDEV_VPORT_PROFILE_OP_CREATE, flags=flags@entry=1) at qemu/qemu_process.c:3813
#6 0x00007fffe93087e8 in qemuDomainObjStart (conn=0x7fffd4000b60, driver=driver@entry=0x7fffe40120e0, vm=vm@entry=0x7fffe40a6410, flags=flags@entry=0) at qemu/qemu_driver.c:6051
#7 0x00007fffe9308e32 in qemuDomainCreateWithFlags (dom=0x7fffcc000d50, flags=0) at qemu/qemu_driver.c:6105
#8 0x00007ffff753c5cc in virDomainCreate (domain=domain@entry=0x7fffcc000d50) at libvirt.c:8861
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Chegu Vinod [Thu, 6 Feb 2014 23:44:36 +0000 (15:44 -0800)]
libvirt support to force convergence of live guest migration
Busy enterprise workloads hosted on large sized VM's tend to dirty
memory faster than the transfer rate achieved via live guest migration.
Despite some good recent improvements (& using dedicated 10Gig NICs
between hosts) the live migration may NOT converge.
Recently support was added in qemu (version 1.6) to allow a user to
choose if they wish to force convergence of their migration via a
new migration capability : "auto-converge". This feature allows for qemu
to auto-detect lack of convergence and trigger a throttle-down of the
VCPUs.
This patch includes the libvirt support needed to trigger this
feature. (Testing is in progress)
Wang Yufei [Thu, 20 Mar 2014 07:14:01 +0000 (07:14 +0000)]
cgroup: Fix start VMs coincidently failed
When I start multi VMs coincidently and any of the cgroup directories
named machine doesn't exist. There's a chance that VM start failed because
of creating directory failed:
Unable to initialize /machine cgroup: File exists
When the errno returned by mkdir in virCgroupMakeGroup is EEXIST,
we should pass it through and continue to start the VM. Signed-off-by: Wang Yufei <james.wangyufei@huawei.com>
The caller may not want all DBus error conditions to be turned
into libvirt errors, so provide a way for the caller to get
back the full DBusError object. They can then check the errors
and only report those that they consider to be fatal.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Introduce alternate way to encode/decode arrays in DBus messages
Currently the DBus helper APIs require the values for an array
to be passed inline in the variadic argument list. This change
introduces support for passing arrays using a pointer to a plain
C array of the basic type. This is of particular benefit for
decoding messages when you don't know how many array elements
are being received.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
The dbus_connection_send_with_reply_and_block method will
automatically call dbus_set_error_from_message for us. We
mistakenly thought we had todo it because of a flaw in the
systemd unit test mock impl. The latter should have directly
set the error object, instead of creating an error message
object.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
The virDBusMessageRead method should not have side-effects on
the message parameter passed in, so unref'ing it is wrong.
The caller should unref only when they decided they are done
with it.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Add DBus helper methods for creating reply messages
The test suites often have to create DBus method reply messages
with payloads. Create two helpers for simplifying the process
of creating replies with payloads.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Zhou Yimin [Fri, 21 Mar 2014 06:52:40 +0000 (14:52 +0800)]
virlog: Modify virLogParseDefaultPriority's comment of return value
virLogParseDefaultPriority's successful return value is the same as
virLogSetDefaultPriority's successful return value. So it should be 0
rather than the parsed log level.