Commit Graph

11788 Commits

Author SHA1 Message Date
John Ferlan
185478af09 vmware: Avoid NULL dereference for 'caps'
When virCapabilitiesNew() fails, caps will be NULL resulting in possible
core when deref'd in cpuDataFree() call.
2013-01-07 17:11:56 -07:00
John Ferlan
f36d83c441 xen: Avoid possible NULL dereference
Change calling sequence to only call xenUnifiedDomainSetVcpusFlags() when
'dom' is not NULL. Use the GET_PRIVATE() macro to reference privateData.
Just return -1 if dom is NULL.
2013-01-07 17:11:56 -07:00
Daniel P. Berrange
4f1f9d91ab Fix virLXCPrepareHostDevices method
The virLXCPrepareHostDevices method was returning success even
when it reported an error, and failed to handle several host
device types
2013-01-07 18:16:54 +00:00
Daniel P. Berrange
f0e4af91e4 Ensure we always setup a private mount namespace for LXC controller
The code for setting up a private /dev/pts for the containers
is also responsible for making the LXC controller have a
private mount namespace. Unfortunately the /dev/pts code is
not run if launching a container without a custom root. This
causes the LXC FUSE mount to leak into the host FS.
2013-01-07 18:14:34 +00:00
Daniel P. Berrange
198c992d26 Speed up fallback to legacy non-QMP probing
Since we daemonized QEMU for capabilities probing there is a long
time if QEMU fails to launch. This is because we're not passing in
any virDomainObjPtr instance and thus the monitor code can not
check to see if the PID is still alive.

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2013-01-07 18:13:54 +00:00
Daniel P. Berrange
038cffd831 Only initialize capabilities after setting dir permissions
The current code is initializing capabilities before setting
directory permissions. Thus the QEMU binaries being run may
not have the ability to create the UNIX monitor socket on
the first run of libvirtd.
2013-01-07 18:13:39 +00:00
Jim Fehlig
68e7bc4561 build: Add libxenctrl to LIBXL_LIBS
Commit dfa1e1dd removed libxenctrl from LIBXL_LIBS, but the libxl
driver uses a symbol from this library.  Explicitly link with
libxenctrl instead of relying on the build system to support
implicit DSO linking.
2013-01-07 10:15:56 -07:00
Eric Blake
a1fd56cb30 build: install libvirt sysctl file correctly
https://bugzilla.redhat.com/show_bug.cgi?id=887017 reports that
even though libvirt attempts to set fs.aio-max-nr via sysctl,
the file was installed with the wrong name and gets ignored by
sysctl.  Furthermore, 'man systcl.d' recommends that packages
install into hard-coded /usr/lib/sysctl.d (even when libdir is
/usr/lib64), so that sysadmins can use /etc/sysctl.d for overrides.

* daemon/Makefile.am (install-sysctl, uninstall-sysctl): Use
correct location.
* libvirt.spec.in (network_files): Reflect this.
2013-01-07 08:56:37 -07:00
Eric Blake
5ec4b22b77 build: .service files don't need to be executable
See also commit 66ff2dd, where we avoided installing these files
as executables.

* daemon/Makefile.am (libvirtd.service): Drop chmod.
* tools/Makefile.am (libvirt-guests.service): Likewise.
* src/Makefile.am (virtlockd.service, virtlockd.socket):
Likewise.
2013-01-07 08:56:36 -07:00
Eric Blake
cb854b8f61 build: properly substitute virtlockd.socket
virtlockd.service could be installed to a configurable root,
but virtlockd.socket was hardcoded to installation into a
distro.

* src/Makefile.am (virtlockd.service, virtlockd.socket): Drop
unused substitutions.
* src/locking/virtlockd.socket.in (ListenStream): Don't hard-code
/var.
2013-01-07 08:56:36 -07:00
Eric Blake
462a69621e build: use common .in replacement mechanism
We had several different styles of .in conversion in our Makefiles:
ALLCAPS, @ALLCAPS@, @lower@, ::lower::
Canonicalize on one form, to make it easier to copy and paste
between .in files.

Also, we were using some non-portable sed constructs: \@ is an
undefined escape sequence (it happens to be @ itself in GNU sed,
but POSIX allows it to mean something else), as well as risky
behavior (failure to consistently quote things means a space
in $(sysconfdir) could throw things off; also, Autoconf recommends
using | rather than , or ! in the s||| operator, because | has to
be quoted in shell and is therefore less likely to appear in file
names than , or !).

Fix all of these uses to follow the same syntax.

* daemon/libvirtd.8.in: Switch to @var@.
* tools/virt-xml-validate.in: Likewise.
* tools/virt-pki-validate.in: Likewise.
* src/locking/virtlockd.init.in: Likewise.
* daemon/Makefile.am: Prefer | over ! in sed.
(libvirtd.8): Prefer consistent substitution.
(libvirtd.init, libvirtd.service): Avoid non-portable sed.
* tools/Makefile.am (libvirt-guests.sh, libvirt-guests.init)
(libvirt-guests.service): Likewise.
(virt-xml-validate, virt-pki-validate, virt-sanlock-cleanup):
Prefer consistent capitalization.
* src/Makefile.am (virtlockd.init, virtlockd.service)
(virtlockd.socket): Prefer consistent substitution.
2013-01-07 08:56:36 -07:00
Osier Yang
1279e421b2 qemu: Check if the shared disk's cdbfilter conflicts with others
This prevents domain starting and disk attaching if the shared disk's
setting conflicts with other active domain(s), E.g. A domain with
"sgio" set as "filtered", however, another active domain is using
it set as "unfiltered".
2013-01-07 21:39:20 +08:00
Osier Yang
278f87c4b5 qemu: set unpriv_sgio when starting domain and attaching disk
This ignores the default "filtered" if unpriv_sgio is not supported
by kernel, but for explicit request "filtered", it error out for
domain starting.
2013-01-07 21:39:06 +08:00
Osier Yang
535aed56a2 conf: Parse and format the new XML
Like "rawio", "sgio" is only allowed for block disk of device
type "lun".

It doesn't default disk->sgio to "filtered" when parsing, as
it won't be able to distinguish explicitly requested "filtered"
and a default "filtered" in driver then. We have to error out for
explicit request when the kernel doesn't support the new sysfs
knob "unpriv_sgio", however, for defaulted "filtered", we can
just ignore it if the kernel doesn't support "unpriv_sgio".
2013-01-07 21:38:43 +08:00
Osier Yang
b9c57e7b0d docs: Add docs and rng schema for new XML tag sgio
This introduces new XML tag "sgio" for disk, its valid values
are "filtered" and "unfiltered", setting it as "filtered" will
set the disk's unpriv_sgio to 0, and "unfiltered" to set it
as 1, which allows the unprivileged SG_IO commands.
2013-01-07 21:37:24 +08:00
Osier Yang
d7ead3e19a qemu: Add a hash table for the shared disks
This introduces a hash table for qemu driver, to store the shared
disk's info as (@major:minor, @ref_count). @ref_count is the number
of domains which shares the disk.

Since we only care about if the disk support unprivileged SG_IO
commands, and the SG_IO commands only make sense for block disk,
this patch only manages (add/remove hash entry) the shared disk for
block disk.

* src/qemu/qemu_conf.h: (Add member 'sharedDisks' of type
                         virHashTablePtr; Declare helpers
                         qemuGetSharedDiskKey, qemuAddSharedDisk
                         and qemuRemoveSharedDisk)
* src/qemu/qemu_conf.c (Implement the 3 helpers)
* src/qemu/qemu_process.c (Update 'sharedDisks' when domain
                           starting and shutdown)
* src/qemu/qemu_driver.c (Update 'sharedDisks' when attaching
                          or detaching disk).
2013-01-07 21:35:19 +08:00
Osier Yang
ba72cb12fa util: Prepare helpers for unpriv_sgio setting
"virGetDeviceID" could be used across the sources, but it doesn't
relate with this series, and could be done later.

* src/util/virutil.h: (Declare virGetDeviceID, and
                       vir{Get,Set}DeviceUnprivSGIO)
* src/util/virutil.c: (Implement virGetDeviceID and
                       vir{Get,Set}DeviceUnprivSGIO)
* src/libvirt_private.syms: Export private symbols of upper helpers
2013-01-07 21:34:01 +08:00
Peter Krempa
731a5a4df7 snapshot: qemu: Allow redefinition of external snapshots
A redefinition of an external inactive snapshot/checkpoint wasn't
possible without this change.
2013-01-05 08:40:01 +01:00
Peter Krempa
709b0f37c5 snapshot: qemu: Fix segfault and vanishing snapshots when redefining
When the disk alignment check done while redefining an existing snapshot
failed, the qemu driver attempted to free the existing snapshot. As in
the cleanup path the definition of the snapshot wasn't assigned, the
cleanup code dereferenced a NULL pointer.

This patch changes the behavior on error paths while redefining snapshot
in two ways:

1) On failure, modifications done on the snapshot definition object are
rolled back.

2) The previous definition of the data isn't freed until it's certain it
won't be needed any more.

This change avoids the segfault and additionally the snapshot doesn't
vanish if redefinition fails for some reason.
2013-01-05 08:40:01 +01:00
Peter Krempa
4494b11f8f snapshot: qemu: Separate logic blocks with newlines 2013-01-05 08:40:00 +01:00
Peter Krempa
51b12b8cca snapshot: conf: Make virDomainSnapshotIsExternal more reusable
Allow to use definition objects with this predicate function.
2013-01-05 08:40:00 +01:00
John Eckersberg
346e43ecfd qemu: Implement virDomainOpenChannel API 2013-01-04 19:03:32 -07:00
John Eckersberg
66a0664974 conf: Add unix socket support to virChrdevOpen
This also changes the function signature to take a
virDomainChrSourceDefPtr instead of just a path, since it needs to
differentiate behavior based on source->type.
2013-01-04 18:07:11 -07:00
John Eckersberg
3c971c675a conf: Rename console-specific identifiers to be more generic
The functionality provided in virchrdev.c (previously virconsole.c) is
applicable to other types of character devices besides consoles, such
as channels.  This patch is just code motion, renaming things such as
"console" or "pty", instead using more general terms such as
"character device" or "device path".
2013-01-04 17:43:21 -07:00
John Eckersberg
4c85421c6c conf: Rename virconsole.* to virchrdev.*
This is just code motion, in preparation to rename identifiers to be
less console-specific.
2013-01-04 17:26:30 -07:00
John Eckersberg
d52add46ed api: Add API to tunnel a guest channel via stream
This patch adds a new API, virDomainOpenChannel, that uses streams to
connect to a virtio channel on a guest.  This creates a secure
communication channel between a guest and a libvirt client.

This behaves the same as virDomainOpenConsole, except on channels
instead of console/serial/parallel devices.
2013-01-04 17:10:55 -07:00
Eric Blake
54df702ed0 build: fix mingw rpm build
Commit d13155c changed which files get installed for the
libvirt-guests service, but did not touch up the mingw spec
file.  As a result, rpmbuild complained:

RPM build errors:
    Installed (but unpackaged) file(s) found:
   /usr/i686-w64-mingw32/sys-root/mingw/libexec/libvirt-guests.sh
   /usr/x86_64-w64-mingw32/sys-root/mingw/libexec/libvirt-guests.sh

which in turn breaks ./autobuild.sh when cross-compilers are present.

* mingw-libvirt.spec.in: Remove files not relevant to mingw.
2013-01-04 17:06:39 -07:00
Eric Blake
f5b654e33e network: fix check for ambiguous lookup
gcc -O2 complained:
../../src/conf/network_conf.c: In function 'virNetworkDefUpdateDNSSrv':
../../src/conf/network_conf.c:3232: error: 'foundIdx' may be used uninitialized in this function [-Wuninitialized]

It turned out to be a spurious warning (we didn't use foundIdx
unless foundCt was non-zero).  But in investigating that, I noticed
a worse problem: we were using 'if (foundCt > 1)', but since foundCt
was bool, it could never be > 1.

* src/conf/network_conf.c (virNetworkDefUpdateDNSHost): Use
correct type.
(virNetworkDefUpdateDNSSrv): Likewise, and silence compiler
warning.
2013-01-04 15:08:25 -07:00
John Ferlan
b667125065 Check and handle select() errors from waitsocket(). 2013-01-04 10:57:25 -07:00
John Ferlan
7ac12d84d0 tests: check and handle error for virAsprintf() calls 2013-01-04 10:57:20 -07:00
John Ferlan
0c6e95fe78 Check return on mkdir for LOCKSPACE_DIR 2013-01-04 10:57:09 -07:00
John Ferlan
0527d6bd41 vmware: silence Coverity warning
Ignore the return status check for vmwareUpdateVMStatus in convenience
routine vmwareDomainObjListUpdateDomain
2013-01-04 10:56:37 -07:00
Michal Privoznik
632c60edde qemu: Detect VGA_QXL capability correctly
Since 4c993d8a we failed to set this important capability, which
allows starting a domain with QXL video card. We set DEVICE_QXL
capability bit instead, which is not necessary wrong. Anyway, if
qemu supports the new '-device qxl' it supports older '-vga qxl'
as well. The latter is used for the primary (the first) qxl video
card, the former for other video cards.
2013-01-04 15:37:09 +01:00
Guido Günther
66ff2ddc29 Install virtlockd.{socket,service} non executable
since they're not scripts but systemd service files.
2013-01-04 11:48:05 +01:00
Ján Tomko
b7a443fcbb qemu: fix a segfault in qemuProcessWaitForMonitor
Commit b3f2b4ca5c left buf unallocated in
the case of QMP capability probing being used, leading to a segfault in
strlen in the cleanup path.

This patch opens the log and allocates the buffer if QMP probing was
used, so we can display the helpful error message.
2013-01-04 11:00:43 +01:00
John Ferlan
718af45895 parallels: check and handle error for virAsprintf() calls
Ignore the return in parallelsMakePoolName() since subsequent check
validates name was allocated.
2013-01-03 14:37:40 -07:00
John Ferlan
62cc7b702b udev: check and handle error for virAsprintf() calls 2013-01-03 14:36:15 -07:00
Eric Blake
35562e01b4 build: fix build with optimization enabled
Build failure detected by Jenkins autobuilder:
http://honk.sigxcpu.org:8001/job/libvirt-build/472/changes

* .gnulib: Update to latest, for stdio.h rpl_fwrite fix.
2013-01-03 10:12:47 -07:00
Michal Privoznik
b3f2b4ca5c qemu: Don't parse log output when starting up a domain
Despite our great effort we still parsed qemu log output.
We wouldn't notice unless upcoming qemu 1.4 changed the
format of the logs slightly. Anyway, now we should gather
all interesting knobs like pty paths from monitor. Moreover,
since for historical reasons the first console can be just
an alias to the first serial port, we need to check this and
copy the pty path if that's the case to the first console.
2013-01-03 09:56:51 +01:00
Eric Blake
83aca30f1e build: use autobuild module to make build logs nicer
A recent build failure made me realize that we could usefully add
a bit more information to configure output, for aid in analysis of
failed builds.  Pulling in the autobuild module merely adds these
four lines to configure output:

configure: autobuild project... libvirt
configure: autobuild revision... v1.0.1-113-g7a74eea
configure: autobuild hostname... myhost
configure: autobuild timestamp... 20130102T233543Z

which can be useful even if not using the Autobuild project to parse
build logs.

* bootstrap.conf (gnulib_modules): Add autobuild.
* configure.ac: Favor git version over release version, when available.
2013-01-02 16:43:58 -07:00
Eric Blake
7a74eeaf06 maint: update to latest gnulib
'make syntax-check' warned that gnulib's copyright is now out of date.

* .gnulib: Update to latest, for copyright year bump.
* gnulib/local/m4/ssize_t.m4.diff: Regenerate.
* bootstrap: Synchronize from upstream.
2013-01-02 09:38:30 -07:00
Osier Yang
91b749a0ac Fix the comment grammar 2013-01-02 23:29:20 +08:00
Osier Yang
c898c7b3d8 Remove the inexistent function in comments
virNodeDeviceLookupByKey is inexistent.
2013-01-02 22:42:44 +08:00
Michal Privoznik
fe915278c1 Revert "qemu: Adapt to new log format"
This reverts commit 28224c4d2a
which shouldn't be needed at all because with current qemu
we obtain all paths from 'query-chardev' output. We ought
not parse log output at all anymore.
2013-01-02 11:52:18 +01:00
Michal Privoznik
28224c4d2a qemu: Adapt to new log format
Since 586502189edf9fd0f89a83de96717a2ea826fdb0 qemu commit, the log
lines reporting chardev's path has changed from:

$ ./x86_64-softmmu/qemu-system-x86_64 -serial pty -serial pty -monitor pty
char device redirected to /dev/pts/5
char device redirected to /dev/pts/6
char device redirected to /dev/pts/7

to:

$ ./x86_64-softmmu/qemu-system-x86_64 -serial pty -serial pty -monitor pty
char device compat_monitor0 redirected to /dev/pts/5
char device serial0 redirected to /dev/pts/6
char device serial1 redirected to /dev/pts/7

However, with current code we are not prepared for such change, which
results in us being unable to start any domain.
2012-12-30 12:12:21 +01:00
Michal Privoznik
ce753ec5d6 sanlock: Chown lease files as well
Since sanlock doesn't run under root:root, we have chown()'ed the
__LIBVIRT__DISKS__ lease file to the user:group defined in the
sanlock config. However, when writing the patch I've forgot about
lease files for each disk (this is the
/var/lib/libvirt/sanlock/<md5>) file.
2012-12-28 16:42:38 +01:00
Michal Privoznik
e2d7e7c61a python: Adapt to virevent rename
With our recent renames under src/util/* we forgot to adapt
python wrapper code generator. This results in some methods being
not exposed:

$ python examples/domain-events/events-python/event-test.py
Using uri:qemu:///system
Traceback (most recent call last):
  File "examples/domain-events/events-python/event-test.py", line 585, in <module>
    main()
  File "examples/domain-events/events-python/event-test.py", line 543, in main
    virEventLoopPureStart()
  File "examples/domain-events/events-python/event-test.py", line 416, in virEventLoopPureStart
    virEventLoopPureRegister()
  File "examples/domain-events/events-python/event-test.py", line 397, in virEventLoopPureRegister
    libvirt.virEventRegisterImpl(virEventAddHandleImpl,
AttributeError: 'module' object has no attribute 'virEventRegisterImpl'
2012-12-28 16:22:09 +01:00
Michal Privoznik
a14768c9d3 qemu: Convert some APIs to use qemuDomObjFromDomain
Many internal qemu APIs must find domain object from passed
virDomainPtr. And with function Peter's introduced, we can use it
instead of copying multiple lines among code.
2012-12-24 09:34:13 +01:00
Michal Privoznik
8c5d2bad12 qemu: Relax locking in DomainHasManagedSaveImage and DomainMonitorCommand
There is no need to hold qemu lock during the whole execution
of these two APIs.
2012-12-24 09:34:13 +01:00
Viktor Mihajlovski
fec9822eeb S390: Re-enable capability probing for virtio devices.
Since we switched to QMP probing, the object types are spelled out
explicitly, i.e. virtio-net-pci. This has effectively disabled
the capability detection of s390 virtio devices. The trivial fix
is to add the s390 virtio types explicitly to qemuCapsObjectProps.

Signed-off-by: Viktor Mihajlovski <mihajlov@linux.vnet.ibm.com>
2012-12-21 14:20:28 -07:00