commit f094aaac changed qemuPrepareHostdevPCIDevices() such that it
may modify the "backend" (vfio vs. legacy kvm) setting in the
virHostdevDef. However, qemuDomainAttachHostPciDevice() (used by
hotplug) copies the backend setting into a local *before* calling
qemuPrepareHostdevPCIDevices(), and then later makes a decision based
on that pre-change value.
The result is that, if the backend had been set to "default" (i.e. not
specified in the config) and was later updated to "VFIO" by
qemuPrepareHostdevPCIDevices(), the qemu process' MacMemLock is not
increased (as is required for VFIO device assignment).
This patch delays making the local copy of backend until after its
potential modification.
The previous patch fixed "forwardPlainNames" so that it really is
doing only what is intended, but left the default to be
"forwardPlainNames='no'". Discussion around the initial version of
that patch led to the decision that the default should instead be
"forwardPlainNames='yes'" (i.e. the original behavior before commit
f3886825). This patch makes that change to the default.
In commit f386825 we began adding the options
--domain-needed
--local=/$mydomain/
to all dnsmasq commandlines with the stated reason of preventing
forwarding of DNS queries for names that weren't fully qualified
domain names ("FQDN", i.e. a name that included some "."s and a domain
name). This was later changed to
domain-needed
local=/$mydomain/
when we moved the options from the dnsmasq commandline to a conf file.
The original patch on the list, and discussion about it, is here:
https://www.redhat.com/archives/libvir-list/2012-August/msg01594.html
When a domain name isn't specified (mydomain == ""), the addition of
"domain-needed local=//" will prevent forwarding of domain-less
requests to the virtualization host's DNS resolver, but if a domain
*is* specified, the addition of "local=/domain/" will prevent
forwarding of any requests for *qualified* names within that domain
that aren't resolvable by libvirt's dnsmasq itself.
An example of the problems this causes - let's say a network is
defined with:
<domain name='example.com'/>
<dhcp>
..
<host mac='52:54:00:11:22:33' ip='1.2.3.4' name='myguest'/>
</dhcp>
This results in "local=/example.com/" being added to the dnsmasq options.
If a guest requests "myguest" or "myguest.example.com", that will be
resolved by dnsmasq. If the guest asks for "www.example.com", dnsmasq
will not know the answer, but instead of forwarding it to the host, it
will return NOT FOUND to the guest. In most cases that isn't the
behavior an admin is looking for.
A later patch (commit 4f595ba) attempted to remedy this by adding a
"forwardPlainNames" attribute to the <dns> element. The idea was that
if forwardPlainNames='yes' (default is 'no'), we would allow
unresolved names to be forwarded. However, that patch was botched, in
that it only removed the "domain-needed" option when
forwardPlainNames='yes', and left the "local=/mydomain/".
Really we should have been just including the option "--domain-needed
--local=//" (note the lack of domain name) regardless of the
configured domain of the network, so that requests for names without a
domain would be treated as "local to dnsmasq" and not forwarded, but
all others (including those in the network's configured domain) would
be forwarded. We also shouldn't include *either* of those options if
forwardPlainNames='yes'. This patch makes those corrections.
This patch doesn't remedy the fact that default behavior was changed
by the addition of this feature. That will be handled in a subsequent
patch.
We support only one spicevmc channel name anyway and the code is
prepared to use the default one, there's only one check missing. It
is also mentioned in the documentation already and helps defining
domains with spice vdagent for people using virsh.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Coverity complains about default: label in libxl_driver.c not be able
to be reached. It's by design for the code and since it's not necessary
in the code nor does it elicit any compiler/make check warnings - just
remove it rather than adding a coverity[dead_error_begin] tag.
While I'm at it, lxc_driver.c and nodeinfo.c have the same design, so I
removed the default labels and the existing coverity tags.
The NWFilter code has as a deadlock race condition between
the virNWFilter{Define,Undefine} APIs and starting of guest
VMs due to mis-matched lock ordering.
In the virNWFilter{Define,Undefine} codepaths the lock ordering
is
1. nwfilter driver lock
2. virt driver lock
3. nwfilter update lock
4. domain object lock
In the VM guest startup paths the lock ordering is
1. virt driver lock
2. domain object lock
3. nwfilter update lock
As can be seen the domain object and nwfilter update locks are
not acquired in a consistent order.
The fix used is to push the nwfilter update lock upto the top
level resulting in a lock ordering for virNWFilter{Define,Undefine}
of
1. nwfilter driver lock
2. nwfilter update lock
3. virt driver lock
4. domain object lock
and VM start using
1. nwfilter update lock
2. virt driver lock
3. domain object lock
This has the effect of serializing VM startup once again, even if
no nwfilters are applied to the guest. There is also the possibility
of deadlock due to a call graph loop via virNWFilterInstantiate
and virNWFilterInstantiateFilterLate.
These two problems mean the lock must be turned into a read/write
lock instead of a plain mutex at the same time. The lock is used to
serialize changes to the "driver->nwfilters" hash, so the write lock
only needs to be held by the define/undefine methods. All other
methods can rely on a read lock which allows good concurrency.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
There are a number of pthreads impls available on Win32
these days, in particular the mingw64 project has a good
impl. Delete the native windows thread implementation and
rely on using pthreads everywhere.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
The check-augeas-lockd test depends on the file
locking/qemu-lockd.conf, so must be skipped when QEMU
is disabled.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Commit 10c9ceff6d intended to introduce new argument for the
testing purpose, but it missed the similar changing of the
device's sg_path. The problem was hidden since my laptop has
the /dev/sg0 and /dev/sg1. A later patch will modify the tests
accordingly.
Signed-off-by: Osier Yang <jyang@redhat.com>
Reported-by: Pavel Hrdina <phrdina@redhat.com>
To support passing the path of the test data to the utils, one
more argument is added to virSCSIDeviceGetSgName,
virSCSIDeviceGetDevName, and virSCSIDeviceNew, and the related
code is changed accordingly.
Later tests for the scsi utils will be based on this patch.
Signed-off-by: Osier Yang <jyang@redhat.com>
It doesn't make sense to fail if the SCSI host device is specified
as "shareable" explicitly between domains (NB, it works if and only
if the device is specified as "shareable" for *all* domains,
otherwise it fails).
To fix the problem, this patch introduces an array for virSCSIDevice
struct, which records all the names of domain which are using the
device (note that the recorded domains must specify the device as
shareable). And the change on the data struct brings on many
subsequent changes in the code.
Prior to this patch, the "shareable" tag didn't work as expected,
it actually work like "non-shareable". So this patch also added notes
in formatdomain.html to declare the fact.
* src/util/virscsi.h:
- Remove virSCSIDeviceGetUsedBy
- Change definition of virSCSIDeviceGetUsedBy and virSCSIDeviceListDel
- Add virSCSIDeviceIsAvailable
* src/util/virscsi.c:
- struct virSCSIDevice: Change "used_by" to be an array; Add
"n_used_by" as the array count
- virSCSIDeviceGetUsedBy: Removed
- virSCSIDeviceFree: frees the "used_by" array
- virSCSIDeviceSetUsedBy: Copy the domain name to avoid potential
memory corruption
- virSCSIDeviceIsAvailable: New
- virSCSIDeviceListDel: Change the logic, for device which is already
in the list, just remove the corresponding entry in "used_by". And
since it's only used in one place, we can safely removing the code
to find out the dev in the list first.
- Copyright updating
* src/libvirt_private.sys:
- virSCSIDeviceGetUsedBy: Remove
- virSCSIDeviceIsAvailable: New
* src/qemu/qemu_hostdev.c:
- qemuUpdateActiveScsiHostdevs: Check if the device existing before
adding it to the list;
- qemuPrepareHostdevSCSIDevices: Error out if the not all domains
use the device as "shareable"; Also don't try to add the device
to the activeScsiHostdevs list if it already there; And make
more sensible error w.r.t the current "shareable" value in
driver->activeScsiHostdevs.
- qemuDomainReAttachHostScsiDevices: Change the logic according
to the changes on helpers.
Signed-off-by: Osier Yang <jyang@redhat.com>
This reverts commit 2996e6be19
and some parts of 2636dc8c4d.
The former one tried to implement QoS setting on bridgeless networks.
However, as discussed upstream [1], the patch is far away from being
useful in even a single case. The whole idea of network QoS is to have
aggregated limits over several interfaces. This patch is doing
completely the opposite when merging two QoS settings (from the network
and the domain interface) into one which is then set at the domain
interface itself, not the network.
The latter one is the test for the previous one. Now none of them makes
sense.
1: https://www.redhat.com/archives/libvir-list/2014-January/msg01441.html
Conflicts:
tests/virnetdevbandwidthtest.c: New test has been introduced since
then.
There are some units within libvirt that utilize virCommand API to run
some commands and deserve own unit testing. These units are, however,
not desired to be rewritten to dig virCommand API usage out. As a great
example virNetDevBandwidth could be used. The problem with the bandwidth
unit is: it uses virCommand API heavily. Therefore we need a mechanism
to not really run a command, but rather see its string representation
after which we can decide if the unit construct the correct sequence of
commands or not.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Add support for specifying various types when doing snapshots. This will
later allow to do snapshots on network backed volumes. Disks of type
'volume' are not supported by snapshots (yet).
Also amend the test suite to check parsing of the various new disk
types that can now be specified.
Commit df36af58 broke parsing of http response from xend. The prior
use of atoi() would happily parse e.g. a string containing "200 OK\r\n",
whereas virStrToLong_i() will fail when called with a NULL end_ptr.
Change the calls to virStrToLong_i() to provide a non-NULL end_ptr.
https://bugzilla.redhat.com/show_bug.cgi?id=1049391
When all source CPU XMLs contain just a single CPU model (with a
possibly varying set of additional feature elements),
virConnectBaselineCPU will try to use this CPU model in the computed
guest CPU. Thus, when used on just a single CPU (useful with
VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES), the result will not use a
different CPU model.
If the computed CPU uses the source model, set fallback mode to 'forbid'
to make sure the guest CPU will always be as close as possible to the
source CPUs.
https://bugzilla.redhat.com/show_bug.cgi?id=1049391
VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES flag for virConnectBaselineCPU
did not work if the resulting guest CPU would disable some features
present in its base model. This patch makes sure we won't try to add
such features twice.
Implement virProcess{Get,Set}Affinity() using cpuset_getaffinity()
and cpuset_setaffinity() calls. Quick search showed that they are
only available on FreeBSD, so placed it inside existing #ifdef
blocks for FreeBSD instead of adding configure checks.
Creating a qemu VM with /dev/hwrng as backend RNG device throws the
following error - "Could not open '/dev/hwrng': Permission denied"
This patch fixes the issue
Signed-off-by: Pradipta Kr. Banerjee <bpradip@in.ibm.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1055484
Currently, libvirt's XML schema of network allows QoS to be defined for
every network even though it has no bridge. For instance:
<network>
<name>vdsm-no-bridge</name>
<forward mode='passthrough'>
<interface dev='em1.10'/>
</forward>
<bandwidth>
<inbound average='1000' peak='5000' burst='1024'/>
<outbound average='1000' burst='1024'/>
</bandwidth>
</network>
The bandwidth limitations can be, however, applied even on such
networks. In fact, they are going to be applied on the interface that
will be connected to the network on a domain startup. This approach,
however, has one limitation. With bridged networks, there are two points
where QoS can be set: bridge and domain interface. The lower limit of
the two is enforced then. For instance, if the interface has 10Mbps
average, but the network only 1Mbps, there's no way for interface to
transmit packets faster than the 1Mbps limit. With two points this is
enforced by kernel. With only one point, we must combine both QoS
settings into one which is set afterwards. Look at
virNetDevBandwidthMinimal() and you'll understand immediately what I
mean.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
This patch allows libvirt user to specify 'host-passthrough'
cpu mode while using qemu/kvm backend on aarch64.
It uses 'host' as a CPU model name instead of some other stub
(correct CPU detection is not implemented yet) to allow libvirt
user to specify 'host-model' cpu mode as well.
Signed-off-by: Oleg Strikov <oleg.strikov@canonical.com>
(crobinso: fix some indentation)
Currently the qemuDomainGetBlockInfo will return allocation == physical
for most backing stores. For a qcow2 block backed device it's possible
to return the highest lv extent allocated from qemu for an active guest.
That is a value where allocation != physical and one would hope be less.
However, if the guest is not running, then the code falls back to returning
allocation == physical. This turns out to be problematic for rhev which
monitors the size of the backing store. During a migration, before the
VM has been started on the target and while it is deemed inactive on the
source, there's a small window of time where the allocation is returned
as physical triggering the code to extend the file unnecessarily.
Since rhev uses transient domains and this is edge condition for a transient
domain, rather than returning good status and allocation == physical when
this "window of opportunity" exists, this patch will check for a transient
(or non persistent) domain and return a failure to the caller rather than
returning the defaults. For a persistent domain, the defaults will be
returned. The description for the virDomainGetBlockInfo has been updated
to describe the phenomena.
the array params is allocated by VIR_ALLOC_N in
remoteDispatchDomainGetCPUStats. it had been set
to zero. No need to reset it to zero again, and
this reset here is incorrect too, nparams * ncpus
is the array length not the size of params array.
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
Unlike the host devices of other types, SCSI host device XML supports
"shareable" tag. This patch introduces it for the virSCSIDevice struct
for a later patch use (to detect if the SCSI device is shareable when
preparing the SCSI host device in QEMU driver).
The "checkPool" is a bit different for pool with "fc_host"
type source adapter, since the vHBA it's based on might be
not created yet (it's created by "startPool", which is
involked after "checkPool" in storageDriverAutostart). So it
should not fail, otherwise the "autostart" of the pool will
fail either.
The problem is easy to reproduce:
* Enable "autostart" for the pool
* Restart libvirtd service
* Check the pool's state
94f8205 added a space to the string but didn't change the buffer size.
Signed-off-by: Bing Bu Cao <mars@linux.vnet.ibm.com>
Signed-off-by: Ján Tomko <jtomko@redhat.com>
For pool which relies on remote resources, such as a "iscsi" type
pool, since how long it takes to export the corresponding devices
to host's sysfs is really depended, it could depend on the network
connection, it also could depend on the host's udev procedures. So
it's likely that the volumes are not able to be detected during pool
starting process, polling the sysfs doesn't work, since we don't
know how much time is best for the polling, and even worse, the
volumes could still be not detected or partly not detected even after
the polling. So we end up with a documentation to prompt the fact,
in virsh manual.
And as a small improvement, let's explicitly say no LUNs found in
the debug log in that case.
There are 2 issues here: First we shouldn't add "1" to the return
value of numa_max_node(), since the semanteme of the error message
was changed, it's not saying about the number of total NUMA nodes
anymore. Second, the value of "bit" is the position of the first
bit which exceeds either numa_max_node() or NUMA_NUM_NODES, it can
be any number in the range, so saying "bigger than $bit" is quite
confused now. For example, assuming there is a NUMA machine which
has 10 NUMA nodes, and one specifies the "nodeset" as "0,5,88",
the error message will be like:
Nodeset is out of range, host cannot support NUMA node bigger than 88
It sounds like all NUMA node number less than 88 is fine, but
actually the maximum NUMA node number the machine supports is 9.
This patch fixes the issues by removing the addition with "1" and
simplifies the error message as "NUMA node $bit is out of range".
Also simplifies the comparision in the while loop by getting the
smaller one of numa_max_node() and NUMA_NUM_NODES up front.
I noticed that we allow virDomainGetVcpusFlags even for read-only
connections, but that with a flag, it can require guest agent
interaction. It is feasible that a malicious guest could
intentionally abuse the replies it sends over the guest agent
connection to possibly trigger a bug in libvirt's JSON parser,
or withhold an answer so as to prevent the use of the agent
in a later command such as a shutdown request. Although we
don't know of any such exploits now (and therefore don't mind
posting this patch publicly without trying to get a CVE assigned),
it is better to err on the side of caution and explicitly require
full access to any domain where the API requires guest interaction
to operate correctly.
I audited all commands that are marked as conditionally using a
guest agent. Note that at least virDomainFSTrim is documented
as needing a guest agent, but that such use is unconditional
depending on the hypervisor (so the existing domain:fs_trim ACL
should be sufficient there, rather than also requirng domain:write).
But when designing future APIs, such as the plans for obtaining
a domain's IP addresses, we should copy the approach of this patch
in making interaction with the guest be specified via a flag, and
use that flag to also require stricter access checks.
* src/libvirt.c (virDomainGetVcpusFlags): Forbid guest interaction
on read-only connection.
(virDomainShutdownFlags, virDomainReboot): Improve docs on agent
interaction.
* src/remote/remote_protocol.x
(REMOTE_PROC_DOMAIN_SNAPSHOT_CREATE_XML)
(REMOTE_PROC_DOMAIN_SET_VCPUS_FLAGS)
(REMOTE_PROC_DOMAIN_GET_VCPUS_FLAGS, REMOTE_PROC_DOMAIN_REBOOT)
(REMOTE_PROC_DOMAIN_SHUTDOWN_FLAGS): Require domain:write for any
conditional use of a guest agent.
* src/xen/xen_driver.c: Fix clients.
* src/libxl/libxl_driver.c: Likewise.
* src/uml/uml_driver.c: Likewise.
* src/qemu/qemu_driver.c: Likewise.
* src/lxc/lxc_driver.c: Likewise.
Signed-off-by: Eric Blake <eblake@redhat.com>
Bugs have been found in the VirtualBox API C bindings. These bugs have
been fixed in versions 4.2.20 and 4.3.4. However, the changes in the
C bindings are incompatible with the vbox_CAPI_v4_2.h and vbox_CAPI_v4_3.h
files which are bundled in libvirt source code.
This is why the following patch adds vbox_CAPI_v4_2_20.h and
vbox_CAPI_v4_3_4.h.
The actual underlying problem here is that until now,
libvirt assumed that VirtualBox API can only change between minor
versions (4.2 -> 4.3), but we have a case here where it changed
(or got fixed) between patch versions (4.2.18 -> 4.2.20).
This patch makes the VBOX_API_VERSION represent the full API
version number (i.e 4002 => 4002000) so there are specific version
numbers for Vbox 4.2.20 (4002020) and 4.3.4 (4003004)
Libvirtd would crash if a domain contained an empty cdrom drive of
type='volume' as the disk def->srcpool member would be dereferenced. Fix
it by checking if the source pool is present before dereferencing it.
Also alter tests to catch this issue in the future.
Reported by: Kevin Shanahan
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1056328
- Use $XDG_RUNTIME_DIR for re-exec state file when running unprivileged.
- argv[0] may not contain a full path to the binary, however it should
contain something that can be looked up in the PATH. Use execvp() to
do path lookup on re-exec.
- As per list discussion [1], ignore --daemon on re-exec.
[1] https://www.redhat.com/archives/libvir-list/2013-December/msg00514.html
Signed-off-by: Michael Chapman <mike@very.puzzling.org>
To retrieve node cpu statistics on Linux system, the
linuxNodeGetCPUstats function simply uses STRPREFIX() to match the cpuid
with the one read from /proc/stat. However, as the file is read line by
line it may happen, that some CPUs share the same prefix. So if user
requested stats for the first CPU, which is offline, then there's no
cpu1 in the stats file so the one that we match is cpu10. Which is
obviously wrong. Fortunately, the IDs are terminated by a space, so we
can utilize that.
Signed-off-by: Bing Bu Cao <mars@linux.vnet.ibm.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1034993
SCSI passthrough disks (<disk .. device="lun">) can't be used as backing
for snapshots. Currently with upstream qemu the vm crashes on such
attempt.
This patch adds a early check to catch an attempt to do such a snapshot
and rejects it right away. qemu will fix the issue but this will let us
control the error message.
I noticed this problem when adding systemd support to netcf, because I
setup the configure.ac to automatically prefer using systemd over
initscripts when possible - although I had copied the
install-data-local target from the example of libvirt's
"libvirt-guests" service more or less verbatim, "make distcheck" would
fail because it was trying to install the service file directly into
/lib/systemd/system rather than into
/home/user/some/unimportant/name/lib/systemd/system.
This is caused by the install/uninstall rules for the systemd unit
files relying on $(DESTDIR) pointing the installed files to the right
place, but in reality $(DESTDIR) is empty during this part of make
distcheck - it instead sets $(prefix) with the toplevel directory used
for its test build/install/uninstall cycle.
(This problem hasn't been seen when running "make distcheck" in
libvirt because libvirt will never build/install systemd support
unless explicitly told to do so on the configure commandline, and
"make distcheck" doesn't put the "--with-initscript=..." option on the
configure commandline.)
I verified that the same problem does exist in libvirt by modifying
libvirt's configure.ac to set:
init_systemd=yes
with_init_script=systemd+redhat
This forces a build/install of the systemd unit files during
distcheck, which yields an error like this:
/usr/bin/install -c -m 644 virtlockd.service \
/lib/systemd/system/
libtool: install: warning: relinking `libvirt-qemu.la'
/usr/bin/install: cannot remove '/lib/systemd/system/virtlockd.service': Permission denied
make[4]: *** [install-systemd] Error 1
After adding $(prefix) to all the definitions of SYSTEMD_UNIT_DIR,
make distcheck now completes successfully with the modified
configure.ac, and the above lines change to something like this:
/usr/bin/install -c -m 644 virtlockd.service \
/home/laine/devel/libvirt/libvirt-1.2.1/_inst/lib/systemd/system/
We shouldn't access the domain definition while we are in the monitor
section as the domain is unlocked. Additionally after we exit from the
monitor we need to check if the VM is still alive. Not doing so resulted
in a crash if qemu exits while attempting to do an external VM snapshot.
spice-server offers an API to disable file transfer messages
on the agent channel between the client and the guest.
This is supported in qemu through the disable-agent-file-xfer option.
This patch exposes this option to libvirt.
Adds a new element 'filetransfer', with one property,
'enable', which accepts a boolean.
Default is enabled, for backward compatibility.
Depends on the capability exported in the first patch of the series.
Signed-off-by: Francesco Romani <fromani@redhat.com>
spice-server offers an API to disable file transfer messages
on the agent channel between the client and the guest.
This is supported in qemu through the disable-agent-file-xfer option.
This patch detects if QEMU supports this option, and add
a capability if does.
Signed-off-by: Francesco Romani <fromani@redhat.com>
With this patch,user can set throttle blkio cgroup for
lxc domain through virsh tool.
Signed-off-by: Guan Qiang <hzguanqiang@corp.netease.com>
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
This is useful in certain circumstances, for example when
libvirtd is being executed by FreeBSD rc script, it cannot find
dmidecode installed from FreeBSD ports because it doesn't have
/usr/local (default prefix for ports) in PATH.
https://bugzilla.redhat.com/show_bug.cgi?id=1046919
If none (KVM, VFIO) of the supported PCI passthrough methods is known to
work on a host, it's better to fail right away with a nice error message
rather than letting attachment fail with a more cryptic message such as
Failed to bind PCI device '0000:07:05.0' to vfio-pci: No such device
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1046919
Since commit v0.9.0-47-g4e8969e (released in 0.9.1) some failures during
device detach were reported to callers of virPCIDeviceBindToStub as
success. For example, even though a device seemed to be detached
virsh # nodedev-detach pci_0000_07_05_0 --driver vfio
Device pci_0000_07_05_0 detached
one could find similar message in libvirt logs:
Failed to bind PCI device '0000:07:05.0' to vfio-pci: No such device
This patch fixes these paths and also avoids overwriting real errors
with errors encountered during a cleanup phase.
https://bugzilla.redhat.com/show_bug.cgi?id=1046919
When a PCI device is not bound to any driver, reattach should just
trigger driver probe rather than failing with
Invalid device 0000:00:19.0 driver file
/sys/bus/pci/devices/0000:00:19.0/driver is not a symlink
While virPCIDeviceGetDriverPathAndName was documented to return success
and NULL driver and path when a device is not attached to any driver but
didn't do so. Thus callers could not distinguish unbound devices from
failures.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
With this patch, user can setup throttle blkio cgroup
through virsh for qemu domain.
Signed-off-by: Guan Qiang <hzguanqiang@corp.netease.com>
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
This patch introduces virCgroupSetBlkioDeviceReadIops,
virCgroupSetBlkioDeviceWriteIops,
virCgroupSetBlkioDeviceReadBps and
virCgroupSetBlkioDeviceWriteBps,
we can use these interfaces to set up throttle
blkio cgroup for domain.
This patch also adds the new throttle blkio cgroup
elements to the test xml.
Signed-off-by: Guan Qiang <hzguanqiang@corp.netease.com>
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
This patch introduces new xml elements under <blkiotune>,
we use these new elements to setup the throttle blkio
cgroup for domain. The new blkiotune node looks like this:
<blkiotune>
<device>
<path>/path/to/block</path>
<weight>1000</weight>
<read_iops_sec>10000</read_iops_sec>
<write_iops_sec>10000</write_iops_sec>
<read_bytes_sec>1000000</read_bytes_sec>
<write_bytes_sec>1000000</write_bytes_sec>
</device>
</blkiotune>
Signed-off-by: Guan Qiang <hzguanqiang@corp.netease.com>
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
https://bugzilla.redhat.com/show_bug.cgi?id=996543
When starting up a domain, the SELinux labeling is done depending on
current configuration. If the labeling fails we check for possible
causes, as not all labeling failures are fatal. For example, if the
labeled file is on NFS which lacks SELinux support, the file can still
be readable to qemu process. These cases are distinguished by the errno
code: NFS without SELinux support returns EOPNOTSUPP. However, we were
missing one scenario. In case there's a read-only disk on a read-only
NFS (and possibly any FS) and the labeling is just optional (not
explicitly requested in the XML) there's no need to make the labeling
error fatal. In other words, read-only file on read-only NFS can fail to
be labeled, but be readable at the same time.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Finish the cleanup of libvirt.c; all uses of virLib*Error have
now been converted to more canonical conventions.
* src/libvirt.c: Use virReportError in remaining errors.
(virLibConnError, virLibDomainError): Delete unused macros.
* cfg.mk (msg_gen_function): Drop unused names.
Signed-off-by: Eric Blake <eblake@redhat.com>
We had a lot of repetition of errors that would occur if we
ever register too many drivers; this is unlikely to occur
unless we start adding a lot of new hypervisor modules, but
if it does occur, it's better to have uniform handling of the
situation, so that a one-line change is all that would be
needed if we decide that an internal error is not the best.
* src/libvirt.c (virDriverCheckTabMaxReturn): New define.
(virRegister*Driver): Use it for less code duplication.
Signed-off-by: Eric Blake <eblake@redhat.com>
The choice of error message and category was not consistent
in the migration code; furthermore, the use of virLibConnError
is no longer necessary now that we have a generic virReportError.
* src/qemu/qemu_migration.c (virDomainMigrate*): Prefer
virReportError over virLibConnError.
Signed-off-by: Eric Blake <eblake@redhat.com>
While auditing the error reporting, I noticed that migration
had some issues. Some of the static helper functions tried
to call virDispatchError(), even though their caller will also
report the error. Also, if a migration is cancelled early
because a uri was not set, we did not guarantee that the finish
stage would not overwrite the first error message.
* src/qemu/qemu_migration.c (doPeer2PeerMigrate2)
(doPeer2PeerMigrate3): Preserve first error when cancelling.
* src/libvirt.c (virDomainMigrateVersion3Full): Likewise.
(virDomainMigrateVersion1, virDomainMigrateVersion2)
(virDomainMigrateDirect): Avoid redundant error dispatch.
(virDomainMigrateFinish2, virDomainMigrateFinish3)
(virDomainMigrateFinish3Params): Don't report error on cleanup
path.
Signed-off-by: Eric Blake <eblake@redhat.com>
The public virConnectRef and virConnectClose API are just thin
wrappers around virObjectRef/virObjectRef, with added object
validation and an error reset. Within our backend drivers, use
of the object validation is just an inefficiency since we always
pass valid objects. More important to think about is what
happens with the error reset; our uses of virConnectRef happened
to be safe (since we hadn't encountered any earlier errors), but
in several cases the use of virConnectClose could lose a real
error.
Ideally, we should also avoid calling virConnectOpen() from
within backend drivers - but that is a known situation that
needs much more design work.
* src/qemu/qemu_process.c (qemuProcessReconnectHelper)
(qemuProcessReconnect): Avoid nested public API call.
* src/qemu/qemu_driver.c (qemuAutostartDomains)
(qemuStateInitialize, qemuStateStop): Likewise.
* src/qemu/qemu_migration.c (doPeer2PeerMigrate): Likewise.
* src/storage/storage_driver.c (storageDriverAutostart):
Likewise.
* src/uml/uml_driver.c (umlAutostartConfigs): Likewise.
* src/lxc/lxc_process.c (virLXCProcessAutostartAll): Likewise.
(virLXCProcessReboot): Likewise, and avoid leaking conn on error.
Signed-off-by: Eric Blake <eblake@redhat.com>
Several APIs clear out a user input buffer before attempting to
populate it; but in a few cases we missed this memset if we
detect a reason for an early exit. Note that these APIs
check for non-NULL arguments, and exit early with an error
message when NULL is passed in; which means that we must be
careful to avoid a NULL deref in order to get to that error
message. Also, we were inconsistent on the use of
sizeof(virType) vs. sizeof(expression); the latter is more
robust if we ever change the type of the expression (although
such action is unlikely since these types are part of our
public API).
* src/libvirt.c (virDomainGetInfo, virDomainGetBlockInfo)
(virStoragePoolGetInfo, virStorageVolGetInfo)
(virDomainGetJobInfo, virDomainGetBlockJobInfo): Move memset
before any returns.
Signed-off-by: Eric Blake <eblake@redhat.com>
There is a number of reported issues when we fail starting a domain.
Turns out that, in some scenarios like high load, 3 second timeout is
not enough for qemu to start up to the phase where the socket is
created. Since there is no downside of waiting longer, raise the
timeout right to 30 seconds.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
To allow using the storage driver APIs to do operation on generic domain
disks we will need to introduce internal storage pools that will give is
a base to support this stuff even on files that weren't originally
defined as a part of the pool.
This patch introduces the 'internal' flag for a storage pool that will
prevent it from being listed along with the user defined storage pools.
Separate the steps to create libvirt's volume metadata from the actual
volume building process. This is already done for regular file based
pools to allow job support for storage APIs.
Currently, during XML parsing, when a call to a FromString() function to
get an enum value fails, the error which is reported is either
VIR_ERR_CONFIG_UNSUPPORTED, VIR_ERR_INTERNAL_ERROR or VIR_ERR_XML_ERROR.
This commit makes such conversion failures consistently return
VIR_ERR_CONFIG_UNSUPPORTED.
Ever since ACL filtering was added in commit 7639736 (v1.1.1), a
user could still use event registration to obtain access to a
domain that they could not normally access via virDomainLookup*
or virConnectListAllDomains and friends. We already have the
framework in the RPC generator for creating the filter, and
previous cleanup patches got us to the point that we can now
wire the filter through the entire object event stack.
Furthermore, whether or not domain:getattr is honored, use of
global events is a form of obtaining a list of networks, which
is covered by connect:search_domains added in a93cd08 (v1.1.0).
Ideally, we'd have a way to enforce connect:search_domains when
doing global registrations while omitting that check on a
per-domain registration. But this patch just unconditionally
requires connect:search_domains, even when no list could be
obtained, based on the following observations:
1. Administrators are unlikely to grant domain:getattr for one
or all domains while still denying connect:search_domains - a
user that is able to manage domains will want to be able to
manage them efficiently, but efficient management includes being
able to list the domains they can access. The idea of denying
connect:search_domains while still granting access to individual
domains is therefore not adding any real security, but just
serves as a layer of obscurity to annoy the end user.
2. In the current implementation, domain events are filtered
on the client; the server has no idea if a domain filter was
requested, and must therefore assume that all domain event
requests are global. Even if we fix the RPC protocol to
allow for server-side filtering for newer client/server combos,
making the connect:serach_domains ACL check conditional on
whether the domain argument was NULL won't benefit older clients.
Therefore, we choose to document that connect:search_domains
is a pre-requisite to any domain event management.
Network events need the same treatment, with the obvious
change of using connect:search_networks and network:getattr.
* src/access/viraccessperm.h
(VIR_ACCESS_PERM_CONNECT_SEARCH_DOMAINS)
(VIR_ACCESS_PERM_CONNECT_SEARCH_NETWORKS): Document additional
effect of the permission.
* src/conf/domain_event.h (virDomainEventStateRegister)
(virDomainEventStateRegisterID): Add new parameter.
* src/conf/network_event.h (virNetworkEventStateRegisterID):
Likewise.
* src/conf/object_event_private.h (virObjectEventStateRegisterID):
Likewise.
* src/conf/object_event.c (_virObjectEventCallback): Track a filter.
(virObjectEventDispatchMatchCallback): Use filter.
(virObjectEventCallbackListAddID): Register filter.
* src/conf/domain_event.c (virDomainEventFilter): New function.
(virDomainEventStateRegister, virDomainEventStateRegisterID):
Adjust callers.
* src/conf/network_event.c (virNetworkEventFilter): New function.
(virNetworkEventStateRegisterID): Adjust caller.
* src/remote/remote_protocol.x
(REMOTE_PROC_CONNECT_DOMAIN_EVENT_REGISTER)
(REMOTE_PROC_CONNECT_DOMAIN_EVENT_REGISTER_ANY)
(REMOTE_PROC_CONNECT_NETWORK_EVENT_REGISTER_ANY): Generate a
filter, and require connect:search_domains instead of weaker
connect:read.
* src/test/test_driver.c (testConnectDomainEventRegister)
(testConnectDomainEventRegisterAny)
(testConnectNetworkEventRegisterAny): Update callers.
* src/remote/remote_driver.c (remoteConnectDomainEventRegister)
(remoteConnectDomainEventRegisterAny): Likewise.
* src/xen/xen_driver.c (xenUnifiedConnectDomainEventRegister)
(xenUnifiedConnectDomainEventRegisterAny): Likewise.
* src/vbox/vbox_tmpl.c (vboxDomainGetXMLDesc): Likewise.
* src/libxl/libxl_driver.c (libxlConnectDomainEventRegister)
(libxlConnectDomainEventRegisterAny): Likewise.
* src/qemu/qemu_driver.c (qemuConnectDomainEventRegister)
(qemuConnectDomainEventRegisterAny): Likewise.
* src/uml/uml_driver.c (umlConnectDomainEventRegister)
(umlConnectDomainEventRegisterAny): Likewise.
* src/network/bridge_driver.c
(networkConnectNetworkEventRegisterAny): Likewise.
* src/lxc/lxc_driver.c (lxcConnectDomainEventRegister)
(lxcConnectDomainEventRegisterAny): Likewise.
Signed-off-by: Eric Blake <eblake@redhat.com>
We haven't had a release with network events yet, so we are free
to fix the RPC so that it actually does what we want. Doing
client-side filtering of per-network events is inefficient if a
connection is only interested in events on a single network out
of hundreds available on the server. But to do server-side
per-network filtering, the server needs to know which network
to filter on - so we need to pass an optional network over on
registration. Furthermore, it is possible to have a client with
both a global and per-network filter; in the existing code, the
server sends only one event and the client replicates to both
callbacks. But with server-side filtering, the server will send
the event twice, so we need a way for the client to know which
callbackID is sending an event, to ensure that the client can
filter out events from a registration that does not match the
callbackID from the server. Likewise, the existing style of
deregistering by eventID alone is fine; but in the new style,
we have to remember which callbackID to delete.
This patch fixes the RPC wire definition to contain all the
needed pieces of information, and hooks into the server and
client side improvements of the previous patches, in order to
switch over to full server-side filtering of network events.
Also, since we fixed this in time, all released versions of
libvirtd that support network events also support per-network
filtering, so we can hard-code that assumption into
network_event.c.
Converting domain events to server-side filtering will require
the introduction of new RPC numbers, as well as a server
feature bit that the client can use to tell whether to use
old-style (server only supports global events) or new-style
(server supports filtered events), so that is deferred to a
later set of patches.
* src/conf/network_event.c (virNetworkEventStateRegisterClient):
Assume server-side filtering.
* src/remote/remote_protocol.x
(remote_connect_network_event_register_any_args): Add network
argument.
(remote_connect_network_event_register_any_ret): Return callbackID
instead of count.
(remote_connect_network_event_deregister_any_args): Pass
callbackID instead of eventID.
(remote_connect_network_event_deregister_any_ret): Drop unused
type.
(remote_network_event_lifecycle_msg): Add callbackID.
* daemon/remote.c
(remoteDispatchConnectNetworkEventDeregisterAny): Drop unused arg,
and deal with callbackID from client.
(remoteRelayNetworkEventLifecycle): Pass callbackID.
(remoteDispatchConnectNetworkEventRegisterAny): Likewise, and
recognize non-NULL network.
* src/remote/remote_driver.c
(remoteConnectNetworkEventRegisterAny): Pass network, and track
server side id.
(remoteConnectNetworkEventDeregisterAny): Deregister by callback id.
(remoteNetworkBuildEventLifecycle): Pass remote id to event queue.
* src/remote_protocol-structs: Regenerate.
Signed-off-by: Eric Blake <eblake@redhat.com>
In order to mirror a server with per-object filtering, the client
needs to track which server callbackID is servicing the client
callback. This patch introduces the notion of a serverID, as
well as the plumbing to use it for network events, although the
actual complexity of using per-object filtering in the remote
driver is deferred to a later patch.
* src/conf/object_event.h (virObjectEventStateEventID): Add parameter.
(virObjectEventStateQueueRemote, virObjectEventStateSetRemote):
New prototypes.
(virObjectEventStateRegisterID): Move...
* src/conf/object_event_private.h: ...here, and add parameter.
(_virObjectEvent): Add field.
* src/conf/network_event.h (virNetworkEventStateRegisterClient): New
prototype.
* src/conf/object_event.c (_virObjectEventCallback): Add field.
(virObjectEventStateSetRemote): New function.
(virObjectEventStateQueue): Make wrapper around...
(virObjectEventStateQueueRemote): New function.
(virObjectEventCallbackListCount): Tweak return count when remote
id matching is used.
(virObjectEventCallbackLookup, virObjectEventStateRegisterID):
Tweak registration when remote id matching will be used.
(virObjectEventNew): Default to no remote id.
(virObjectEventCallbackListAddID): Likewise, but set remote id
when one is available.
(virObjectEventCallbackListRemoveID)
(virObjectEventCallbackListMarkDeleteID): Adjust return value when
remote id was set.
(virObjectEventStateEventID): Query existing id.
(virObjectEventDispatchMatchCallback): Require matching event id.
(virObjectEventStateCallbackID): Adjust caller.
* src/conf/network_event.c (virNetworkEventStateRegisterClient): New
function.
(virNetworkEventStateRegisterID): Update caller.
* src/conf/domain_event.c (virDomainEventStateRegister)
(virDomainEventStateRegisterID): Update callers.
* src/remote/remote_driver.c
(remoteConnectNetworkEventRegisterAny)
(remoteConnectNetworkEventDeregisterAny)
(remoteConnectDomainEventDeregisterAny): Likewise.
(remoteEventQueue): Hoist earlier to avoid forward declaration,
and add parameter. Adjust all callers.
* src/libvirt_private.syms (conf/object_event.h): Drop function.
Signed-off-by: Eric Blake <eblake@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1047659
If a VM dies very early during an attempted connect to the guest agent
while the locks are down the domain monitor object will be freed. The
object is then accessed later as any failure during guest agent startup
isn't considered fatal.
In the current upstream version this doesn't lead to a crash as
virObjectLock called when entering the monitor in
qemuProcessDetectVcpuPIDs checks the pointer before attempting to
dereference (lock) it. The NULL pointer is then caught in the monitor
helper code.
Before the introduction of virObjectLockable - observed on 0.10.2 - the
pointer is locked directly via virMutexLock leading to a crash.
To avoid this problem we need to differentiate between the guest agent
not being present and the VM quitting when the locks were down. The fix
reorganizes the code in qemuConnectAgent to add the check and then adds
special handling to the callers.
A "xmlstr" string may not be assigned into a "doc" pointer and it
could cause memory leak. To fix it if the "doc" pointer is NULL and
the "xmlstr" string is not assigned we should free it.
This has been found by coverity.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
There could be a memory leak caused by "managed_system" string, if any
error occurs before "managed_system" is assigned into
"phyp_driver->managed_system". The "managed_system" string wouldn't be
freed at all. The better way is to free the "managed_system" instead
of the one assigned in the "phyp_driver".
This has been found by coverity.
Pointed out by John, that the "phyp_driver->xmlopt" needs to be
unreferenced as well.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Signed-off-by: John Ferlan <jferlan@redhat.com>
If there is no error while executing a function "openvzParseBarrierLimit"
a "str" string where is duplicate of a "value" string isn't freed and it
leads into memory leak.
This has been found by coverity.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1047577
When writing commit 173c291, I missed the fact virNetServerClientClose
unlocks the client object before actually clearing client->sock and thus
it is possible to hit a window when client->keepalive is NULL while
client->sock is not NULL. I was thinking client->sock == NULL was a
better check for a closed connection but apparently we have to go with
client->keepalive == NULL to actually fix the crash.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
On my Fedora 20 box with mingw cross-compiler, the build failed with:
../../src/rpc/virnetclient.c: In function 'virNetClientSetTLSSession':
../../src/rpc/virnetclient.c:745:14: error: unused variable 'oldmask' [-Werror=unused-variable]
sigset_t oldmask, blockedsigs;
^
I traced it to the fact that mingw64-winpthreads installs a header
that does #define pthread_sigmask(...) 0, which means any argument
only ever passed to pthread_sigmask is reported as unused. This
patch works around the compilation failure, with behavior no worse
than what mingw already gives us regarding the function being a
no-op.
* configure.ac (pthread_sigmask): Probe for broken mingw macro.
* src/util/virutil.h (pthread_sigmask): Rewrite to something that
avoids unused variables.
Signed-off-by: Eric Blake <eblake@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1047577
When a client closes its connection to libvirtd early during
virConnectOpen, more specifically just after making
REMOTE_PROC_CONNECT_SUPPORTS_FEATURE call to check if
VIR_DRV_FEATURE_PROGRAM_KEEPALIVE is supported without even waiting for
the result, libvirtd may crash due to a race in keep-alive
initialization. Once receiving the REMOTE_PROC_CONNECT_SUPPORTS_FEATURE
call, the daemon's event loop delegates it to a worker thread. In case
the event loop detects EOF on the connection and calls
virNetServerClientClose before the worker thread starts to handle
REMOTE_PROC_CONNECT_SUPPORTS_FEATURE call, client->keepalive will be
disposed by the time virNetServerClientStartKeepAlive gets called from
remoteDispatchConnectSupportsFeature. Because the flow is common for
both authenticated and read-only connections, even unprivileged clients
may cause the daemon to crash.
To avoid the crash, virNetServerClientStartKeepAlive needs to check if
the connection is still open before starting keep-alive protocol.
Every libvirt release since 0.9.8 is affected by this bug.
Any test suite which involves a virDomainDefPtr should
call virDomainDefCheckABIStability with itself just as
a basic sanity check that the identity-comparison always
succeeds. This would have caught the recent NULL pointer
access crash.
Make sure we cope with def->name being NULL since the
VMWare config parser produces NULL names.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Ever since commit 61ac8ce, Coverity complained about
remoteNetworkBuildEventLifecycle not checking for NULL failure
to build an event, compared to other calls in the code base.
But the problem is latent from copy and paste; all 17 of our
remote*BuildEvent* functions in remote_driver.c have the same
issue - if an OOM causes an event to not be built, we happily
pass NULL to remoteEventQueue(), but that function has marked
event as a nonnull parameter. We were getting lucky (the
event queue's first use of the event happened to be a call to
virIsObjectClass(), which acts gracefully on NULL, so there
was no way to crash); but this is a latent bug waiting to bite
us due to the disregard for the nonnull attribute, as well as
a waste of resources in the event queue. Better is to just
refuse to queue NULL. The discard is silent, since the problem
only happens on OOM, and since events are already best effort -
if we fail to get an event, it's not like we have any memory
left to report the issue, nor any idea of who would benefit
from knowing we couldn't create or queue the event.
* src/remote/remote_driver.c (remoteEventQueue): Ignore NULL event.
Signed-off-by: Eric Blake <eblake@redhat.com>
Our fixes for CVE-2013-4400 were so effective at "fixing" bugs
in virt-login-shell that we ended up fixing it into a useless
do-nothing program.
Commit 3e2f27e1 picked the name LIBVIRT_SETUID_RPC_CLIENT for
the witness macro when we are doing secure compilation. But
commit 9cd6a57d checked whether the name IN_VIRT_LOGIN_SHELL,
from an earlier version of the patch series, was defined; with
the net result that virt-login-shell invariably detected that
it was setuid and failed virInitialize.
Commit b7fcc799 closed all fds larger than stderr, but in the
wrong place. Looking at the larger context, we mistakenly did
the close in between obtaining the set of namespace fds, then
actually using those fds to switch namespace, which means that
virt-login-shell will ALWAYS fail.
This is the minimal patch to fix the regressions, although
further patches are also worth having to clean up poor
semantics of the resulting program (for example, it is rude to
not pass on the exit status of the wrapped program back to the
invoking shell).
* tools/virt-login-shell.c (main): Don't close fds until after
namespace swap.
* src/libvirt.c (virGlobalInit): Use correct macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
The existing check of domain snapshots validated that they
point to a domain, but did not validate that the domain
points to a connection, even though any errors blindly assume
the connection is valid. On the other hand, as mentioned in
commit 6e130ddc, any valid domain is already tied to a valid
connection, and VIR_IS_SNAPSHOT vs. VIR_IS_DOMAIN_SNAPSHOT
makes no real difference; it's best to just validate the chain
of all three. For consistency with previous patches, continue
the trend of using a common macro. For now, we don't need
virCheckDomainSnapshotGoto().
* src/datatypes.h (virCheckDomainSnapshotReturn): New macro.
(VIR_IS_SNAPSHOT, VIR_IS_DOMAIN_SNAPSHOT):
Drop unused macros.
* src/libvirt.c: Use macro throughout.
(virLibDomainSnapshotError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
While all errors related to invalid nwfilters appeared to be
consistent, we might as well continue the trend of using a
common macro. As in commit 6e130ddc, the difference between
VIR_IS_NWFILTER and VIR_IS_CONNECTED_NWFILTER is moot, since
reference counting means any valid nwfilter is also tied to
a valid connection. For now, we don't need virCheckNWFilterGoto().
* src/datatypes.h (virCheckNWFilterReturn): New macro.
(VIR_IS_NWFILTER, VIR_IS_CONNECTED_NWFILTER): Drop unused macros.
* src/libvirt.c: Use macro throughout.
(virLibNWFilterError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
For streams validation, we weren't consistent on whether to
use VIR_FROM_NONE or VIR_FROM_STREAMS. Furthermore, in many
API, we want to ensure that a stream is tied to the same
connection as the other object we are operating on; while
other API failed to validate the stream at all. And the
difference between VIR_IS_STREAM and VIR_IS_CONNECTED_STREAM
is moot; as in commit 6e130ddc, we know that reference
counting means a valid stream will always be tied to a valid
connection. Similar to previous patches, use a common macro
to make it nicer.
* src/datatypes.h (virCheckStreamReturn, virCheckStreamGoto):
New macros.
(VIR_IS_STREAM, VIR_IS_CONNECTED_STREAM): Drop unused macros.
* src/libvirt.c: Use macro throughout.
(virLibStreamError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
While all errors related to invalid secrets appeared to be
consistent, we might as well continue the trend of using a
common macro. Just as in commit 6e130ddc, the difference
between VIR_IS_SECRET and VIR_IS_CONNECTED_SECRET is moot
(due to reference counting, any valid secret must be tied to
a valid domain). For now, we don't need virCheckSecretGoto().
* src/datatypes.h (virCheckSecretReturn): New macro.
(VIR_IS_SECRET, VIR_IS_CONNECTED_SECRET): Drop unused macros.
* src/libvirt.c: Use macro throughout.
(virLibSecretError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
While all errors related to invalid node device appeared to be
consistent, we might as well continue the trend of using a
common macro. For now, we don't need virCheckNodeDeviceGoto().
* src/datatypes.h (virCheckNodeDeviceReturn): New macro.
(VIR_IS_NODE_DEVICE, VIR_IS_CONNECTED_NODE_DEVICE): Drop
unused macros.
* src/libvirt.c: Use macro throughout.
(virLibNodeDeviceError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
The commit cad3cf9a95 introduced a crash
due to wrong order of parameters being passed to the function. When
deleting an element, the function decreased the iterator instead of
count and if listing volumes after that (or undefining the pool, NULL
was being dereferenced.
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
For storage volume validation, we weren't consistent on
whether to use VIR_FROM_NONE or VIR_FROM_STORAGE. Similar
to previous patches, use a common macro to make it nicer.
Furthermore, just as in commit 6e130ddc, the difference
between VIR_IS_STORAGE_VOL and VIR_IS_CONNECTED_STORAGE_VOL
is moot (due to reference counting, any valid volume must
be tied to a valid connection).
virStorageVolCreateXMLFrom allows cross-connection cloning,
where the error is reported against the connection of the
destination pool.
* src/datatypes.h (virCheckStorageVolReturn)
(virCheckStorageVolGoto): New macros.
(VIR_IS_STORAGE_VOL, VIR_IS_CONNECTED_STORAGE_VOL): Drop
unused macros.
* src/libvirt.c: Use macro throughout.
(virLibStorageVolError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
This basically reverts commit ba64b97134
"libxl: Allow libxl to set NIC devid". However assigning devid's
before calling libxlMakeNic does not work as that is calling
libxl_device_nic_init which sets it back to -1.
Right now auto-assignment only works in the hotplug case. But even if
that would be fixed at some point (if that is possible at all), this
would add a weird dependency between Xen and libvirt versions.
The change here should accept any auto-assignment that makes it into
libxl_device_nic_init. My understanding is that a caller always is
allowed to make the devid choice itself. And assuming libxlMakeNicList
is only used on domain creation, a sequential numbering should be ok.
Signed-off-by: Stefan Bader <stefan.bader@canonical.com>
virStoragePoolBuild reported an invalid pool as if it were an
invalid network. Likewise, we weren't consistent on whether to
use VIR_FROM_NONE or VIR_FROM_STORAGE. Similar to previous
patches, use a common macro to make it nicer. Furthermore, just
as in commit 6e130ddc, the difference between VIR_IS_STORAGE_POOL
and VIR_IS_CONNECTED_STORAGE_POOL is moot (due to reference
counting, any valid pool must be tied to a valid connection).
For now, we don't need virCheckStoragePoolGoto().
* src/datatypes.h (virCheckStoragePoolReturn): New macro.
(VIR_IS_STORAGE_POOL, VIR_IS_CONNECTED_STORAGE_POOL): Drop
unused macros.
* src/libvirt.c: Use macro throughout.
(virLibStoragePoolError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
There is no easy way to test authentication against libvirt. This
commit modifies the test driver to allow simple username/password
authentication.
You modify the test XML by adding:
<node>
...
<auth>
<user password="123456">rich</user>
<user>jane</user>
</auth>
</node>
If there are any /node/auth/user elements, then authentication is
required by the test driver (if none are present, then the test driver
will work as before and not require authentication).
In the example above, two phony users are added:
rich password: 123456
jane no password required
The test driver will demand a username. If the password attribute is
present (or if the username entered is wrong), then the password is
also asked for and checked:
$ virsh -c test://$(pwd)/testnode.xml list
Enter username for localhost: rich
Enter rich's password for localhost: ***
Id Name State
----------------------------------------------------
1 fv0 running
2 fc4 running
Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
When checking for a valid interface, we weren't consistent on
whether we reported as VIR_FROM_NONE or VIR_FROM_INTERFACE.
Similar to previous patches, use a common macro to make it nicer.
Furthermore, just as in commit 6e130ddc, the difference between
VIR_IS_INTERFACE and VIR_IS_CONNECTED_INTERFACE is moot (due to
reference counting, any valid interface must be tied to a valid
connection). For now, we don't need virCheckInterfaceGoto().
* src/datatypes.h (virCheckInterfaceReturn): New macro.
(VIR_IS_INTERFACE, VIR_IS_CONNECTED_INTERFACE): Drop unused
macros.
* src/libvirt.c: Use macro throughout.
(virLibInterfaceError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
Commit cfd62c1 was incomplete; I found more cases where error
messages were being overwritten, and where the code between
the three registration/deregistration APIs was not consistent.
Since it is fairly easy to trigger an attempt to deregister an
unregistered object through public API, I also changed the error
message from VIR_ERR_INTERNAL_ERROR to VIR_ERR_INVALID_ARG.
* src/conf/object_event.c (virObjectEventCallbackListEventID):
Inline...
(virObjectEventStateEventID): ...into lone caller, and report
error on failure.
(virObjectEventCallbackListAddID, virObjectEventStateCallbackID)
(virObjectEventCallbackListRemoveID)
(virObjectEventCallbackListMarkDeleteID): Tweak error category.
* src/remote/remote_driver.c (remoteConnectDomainEventRegister):
Don't leak registration on failure.
(remoteConnectDomainEventDeregisterAny)
(remoteConnectNetworkEventDeregisterAny): Don't overwrite error.
Signed-off-by: Eric Blake <eblake@redhat.com>
When checking for a valid network, we weren't consistent on
whether we reported an invalid network or a connection. Similar
to previous patches such as commit 6e130ddc, the difference
between VIR_IS_NETWORK and VIR_IS_CONNECTED_NETWORK is moot (due
to reference counting, any valid network must be tied to a valid
connection). Use a common macro to make the error reporting
for invalid networks nicer.
* src/datatypes.h (virCheckNetworkReturn, virCheckNetworkGoto): New
macros.
(VIR_IS_NETWORK, VIR_IS_CONNECTED_NETWORK): Drop unused macros.
* src/libvirt.c: Use macro throughout.
(virLibNetworkError): Drop unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
Like commit 94a26c7e from Eric Blake, the old fuzzy code should
be replaced by the new array management macros now.
And the type of scsi->count should be changed into "size_t", and
thus virSCSIDeviceListCount should return size_t instead, similar
for vir{PCI,USB}DeviceListCount.
the unix socket /var/run/libvirt/lxc/domain.sock is not created
under the selinux context which configured by <seclabel>.
If we try to connect the domain.sock under the selinux context
of domain in virtLXCProcessConnectMonitor,selinux will deny
this connect operation.
type=AVC msg=audit(1387953696.067:662): avc: denied { connectto } for pid=21206 comm="libvirtd" path="/usr/local/var/run/libvirt/lxc/systemd.sock" scontext=unconfined_u:system_r:svirt_lxc_net_t:s0:c770,c848 tcontext=unconfined_u:system_r:unconfined_t:s0-s0:c0.c1023 tclass=unix_stream_socket
fix this problem by creating socket under selinux context of domain.
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
The function checks for @conn to be valid and locks its mutex. Then, it
checks if callee is unregistering the same callback that he registered
previously. If this fails an error is reported and the control jumps to
'error' label. Here, if @conn has some errors (and it certainly does -
the one that's been just reported) the conn->mutex is locked again -
without any previous unlock:
Thread 1 (Thread 0x7fb500ef1800 (LWP 18982)):
#0 __lll_lock_wait () at ../nptl/sysdeps/unix/sysv/linux/x86_64/lowlevellock.S:135
#1 0x00007fb4fd99ce56 in _L_lock_918 () from /lib64/libpthread.so.0
#2 0x00007fb4fd99ccaa in __GI___pthread_mutex_lock (mutex=0x7fb50153b670) at pthread_mutex_lock.c:64
#3 0x00007fb5007e574d in virMutexLock (m=m@entry=0x7fb50153b670) at util/virthreadpthread.c:85
#4 0x00007fb5007b198e in virDispatchError (conn=conn@entry=0x7fb50153b5e0) at util/virerror.c:594
#5 0x00007fb5008a3735 in virConnectUnregisterCloseCallback (conn=0x7fb50153b5e0, cb=cb@entry=0x7fb500f588e0 <vshCatchDisconnect>) at libvirt.c:21025
#6 0x00007fb500f5d690 in vshReconnect (ctl=ctl@entry=0x7fffff60e710) at virsh.c:328
#7 0x00007fb500f5dc50 in vshCommandRun (ctl=ctl@entry=0x7fffff60e710, cmd=0x7fb50152ca80) at virsh.c:1755
#8 0x00007fb500f5861b in main (argc=<optimized out>, argv=<optimized out>) at virsh.c:3393
And since the conn's mutex is not recursive, the virDispatchError will
never ever lock it successfully.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Cleanup after a previous patch, commit 6e130dd. In particular,
note that xenDomainUsedCpus can only be reached from
xenUnifiedDomainGetXMLDesc, which in turn is only reached from
public API that already validated the domain.
* src/xen/xen_driver.c (xenDomainUsedCpus): Drop redundant check.
* src/datatypes.h (VIR_IS_DOMAIN, VIR_IS_CONNECTED_DOMAIN):
Delete, and inline into all callers, since no other file uses it
any more.
Signed-off-by: Eric Blake <eblake@redhat.com>
In datatype.c, virGetDomainSnapshot could result in the message:
error: invalid domain pointer in bad domain
Furthermore, while there are a few functions in libvirt.c that
only care about a virDomainPtr without regards to the connection
(such as virDomainGetName), most functions also require a valid
connection. Yet several functions were blindly dereferencing
the conn member without checking it for validity first (such as
virDomainOpenConsole). Rather than try and correct all usage
of VIR_IS_DOMAIN vs. VIR_IS_CONNECTED_DOMAIN, it is easier to
just blindly require that a valid domain object always has a
valid connection object (which should be true anyways, since
every domain object holds a reference to its connection, so the
connection will not be closed until all domain objects have
also been closed to release their reference).
After this patch, all places that validate a domain consistently
report:
error: invalid domain pointer in someFunc
* src/datatypes.h (virCheckDomainReturn, virCheckDomainGoto): New
macros.
* src/datatypes.c (virGetDomainSnapshot): Use new macro.
(virLibConnError): Delete unused macro.
Signed-off-by: Eric Blake <eblake@redhat.com>
While comparing network and domain events, I noticed that the
test driver had to do a cast in one place and not the other.
For consistency, we should hide the necessary casting as low
as possible in the stack, with everything else using saner
types.
* src/conf/network_event.h (virNetworkEventStateRegisterID): Alter
type.
* src/conf/network_event.c (virNetworkEventStateRegisterID): Hoist
cast here.
* src/test/test_driver.c (testConnectNetworkEventRegisterAny):
Simplify callers.
* src/remote/remote_driver.c
(remoteConnectNetworkEventRegisterAny): Likewise.
* src/network/bridge_driver.c
(networkConnectNetworkEventRegisterAny): Likewise.
Signed-off-by: Eric Blake <eblake@redhat.com>
If a user registers for a domain event filtered to a particular
domain, but the persistent domain is offline at the time, then
the code silently failed to set up the filter. As a result,
the event fires for all domains, rather than being filtered.
Network events were immune, since they always passed an id
0 argument.
The key to this patch is realizing that
virObjectEventDispatchMatchCallback() only cared about uuid;
so refusing to create a meta for a negative id is pointless,
and in fact, malloc'ing meta at all was overkill; instead,
just directly store a uuid and a flag of whether to filter.
Note that virObjectEventPtr still needs all fields of meta,
because this is how we reconstruct a virDomainPtr inside the
dispatch handler before calling the end user's callback
pointer with the correct object, even though only the uuid
portion of meta is used in deciding whether a callback
matches the given event. So while uuid is optional for
callbacks, it is mandatory for events.
The change to testDomainCreateXMLMixed is merely on the setup
scenario (as you can't register for a domain unless it is either
running or persistent). I actually first wrote that test for
this patch, then rebased it to also cover a prior patch (commit
4221d64), but had to adjust it for that patch to use Create
instead of Define for setting up the domain long enough to
register the event in order to work around this bug. But while
the setup is changed, the main body of the test is still about
whether creation events fire as expected.
* src/conf/object_event_private.h (_virObjectEventCallback):
Replace meta with uuid and flag.
(virObjectEventCallbackListAddID): Update signature.
* src/conf/object_event.h (virObjectEventStateRegisterID):
Likewise.
* src/conf/object_event_private.h (virObjectEventNew): Document
use of name and uuid in events.
* src/conf/object_event.c (virObjectEventCallbackListAddID): Drop
arguments that don't affect filtering.
(virObjectEventCallbackListRemoveID)
(virObjectEventDispatchMatchCallback)
(virObjectEventStateRegisterID): Update clients.
* src/conf/domain_event.c (virDomainEventCallbackListAdd)
(virDomainEventStateRegisterID): Likewise.
* src/conf/network_event.c (virNetworkEventStateRegisterID):
Likewise.
* tests/objecteventtest.c (testDomainCreateXMLMixed): Enhance test.
Signed-off-by: Eric Blake <eblake@redhat.com>
Consider these two calls, in either order:
id1 = virConnectDomainEventRegisterAny(conn, NULL,
VIR_DOMAIN_EVENT_ID_LIFECYCLE,
VIR_DOMAIN_EVENT_CALLBACK(callback), NULL, NULL);
virConnectDomainEventRegister(conn, callback, NULL, NULL);
Right now, the second call fails, because under the hood, the
old-style function registration is tightly coupled to the
new style lifecycle eventID, and the two calls both try
to register the same global eventID callback representation.
We've alreay documented that users should avoid old-style
registration and deregistration, so anyone heeding the advice
won't run into this situation. But it would be even nicer if
we pretend the two interfaces are completely separate, and
disallow any cross-linking. That is, a call to old-style
deregister should never remove a new-style callback even if it
is the same function pointer, and a call to new-style callback
using only callbackIDs obtained legitimately should never
remove an old-style callback (of course, since our callback
IDs are sequential, and there is still coupling under the
hood, you can easily guess the callbackID of an old style
registration and use new-style deregistration to nuke it - but
that starts to be blatantly bad coding on your part rather
than a surprising result on what looks like reasonable
stand-alone API).
With this patch, you can now register a global lifecycle event
handler twice, by using both old and new APIs; if such an event
occurs, your callback will be entered twice. But that is not a
problem in practice, since it is already possible to use the
new API to register both a global and per-domain event handler
using the same function, which will likewise fire your callback
twice for that domain. Duplicates are still prevented when
using the same API with same parameters twice (old-style twice,
new-style global twice, or new-style per-domain with same domain
twice), and things are still bounded (it is not possible to
register a single function pointer more than N+2 times per event
id, where N is the number of domains available on the connection).
Besides, it has always been possible to register as many
separate function pointers on the same event id as desired,
through either old or new style API, where the bound there is
the physical limitation of writing a program with enough
distinct function pointers.
Adding another event registration in the testsuite is sufficient
to cover this, where the test fails without the rest of the patch.
* src/conf/object_event.c (_virObjectEventCallback): Add field.
(virObjectEventCallbackLookup): Add argument.
(virObjectEventCallbackListAddID, virObjectEventStateCallbackID):
Adjust callers.
* tests/objecteventtest.c (testDomainCreateXMLMixed): Enhance test.
Signed-off-by: Eric Blake <eblake@redhat.com>
On the surface, this sequence of API calls should succeed:
id1 = virConnectDomainEventRegisterAny(..., VIR_DOMAIN_EVENT_ID_LIFECYCLE,...);
id2 = virConnectDomainEventRegisterAny(..., VIR_DOMAIN_EVENT_ID_RTC_CHANGE,...);
virConnectDomainEventDeregisterAny(id1);
id1 = virConnectDomainEventRegisterAny(..., VIR_DOMAIN_EVENT_ID_LIFECYCLE,...);
And for test:///default, it does. But for qemu:///system, it fails:
libvirt: XML-RPC error : internal error: domain event 0 already registered
Looking closer, the bug is caused by miscommunication between
the object event engine and the client side of the remote driver.
In our implementation, we set up a single server-side event per
eventID, then the client side replicates that one event to all
callbacks that have been registered client side. To know when
to turn the server side eventID on or off, the client side must
track how many events for the same eventID have been registered.
But while our code was filtering by eventID on event registration,
it did not filter on event deregistration. So the above API calls
resulted in the deregister returning 1 instead of 0, so no RPC
deregister was issued, and the final register detects on the
server side that the server is already handling eventID 0.
Unfortunately, since the problem is only observable on remote
connections, it's not possible to enhance objecteventtest to
expose the semantics using only public API entry points.
* src/conf/object_event.c (virObjectEventCallbackListCount): New
function.
(virObjectEventCallbackListAddID)
(virObjectEventCallbackListRemoveID)
(virObjectEventCallbackListMarkDeleteID): Use it.
Signed-off-by: Eric Blake <eblake@redhat.com>
When the host is configured with very restrictive firewall (default policy
is DROP for all chains, including OUTPUT), the bridge driver for Linux
adds netfilter entries to allow DHCP and DNS requests to go from the VM
to the dnsmasq of the host.
The issue that this commit fixes is the fact that a DROP policy on the OUTPUT
chain blocks the DHCP replies from the host’s dnsmasq to the VM.
As DHCP replies are sent in UDP, they are not caught by any --ctstate ESTABLISHED
rule and so, need to be explicitly allowed.
Signed-off-by: Lénaïc Huard <lenaic@lhuard.fr.eu.org>
When determining if a device is behind a PCI bridge, the PCI device
class is checked by reading the config space. However, there are some
devices which have the wrong class on the config space, but the class is
initialized by Linux correctly as a PCI BRIDGE. This class can be read
by the sysfs file '/sys/bus/pci/devices/xxxx:xx:xx.x/class'.
One example of such bridge is IBM PCI Bridge 1014:03b9, which is
identified as a Host Bridge when reading the config space.
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@linux.vnet.ibm.com>
Tighten up scope after the previous patch avoided using
internals. This will also make it easier to change
internal implementation without having to chase down quite
as many impacted callers or worrying about two files getting
implementations out of sync.
* src/conf/object_event_private.h
(virObjectEventCallbackListAddID, virObjectEventQueueClear)
(virObjectEventStateLock, virObjectEventStateUnlock)
(virObjectEventTimer): Drop prototype.
(_virObjectEventCallbackList, _virObjectEventState)
(_virObjectEventCallback): Move...
* src/conf/object_event.c: ...here.
(virObjectEventCallbackListAddID, virObjectEventQueueClear)
(virObjectEventStateLock, virObjectEventStateUnlock)
(virObjectEventTimer): Mark private.
Signed-off-by: Eric Blake <eblake@redhat.com>
Right now, the older virConnectDomainEventRegister (takes a
function pointer, returns 0 on success) and the newer
virConnectDomainEventRegisterID (takes an eventID, returns a
callbackID) share the underlying implementation (the older
API ends up consuming a callbackID for eventID 0 under the
hood). We implemented that by a lot of copy and pasted
code between object_event.c and domain_event.c, according to
whether we are dealing with a function pointer or an eventID.
However, our copy and paste is not symmetric. Consider this
sequence:
id1 = virConnectDomainEventRegisterAny(conn, dom,
VIR_DOMAIN_EVENT_ID_LIFECYCLE,
VIR_DOMAIN_EVENT_CALLBACK(callback), NULL, NULL);
virConnectDomainEventRegister(conn, callback, NULL, NULL);
virConnectDomainEventDeregister(conn, callback);
virConnectDomainEventDeregsiterAny(conn, id1);
the first three calls would succeed, but the third call ended
up nuking the id1 callbackID (the per-domain new-style handler),
then the fourth call failed with an error about an unknown
callbackID, leaving us with the global handler (old-style) still
live and receiving events. It required another old-style
deregister to clean up the mess. Root cause was that
virDomainEventCallbackList{Remove,MarkDelete} were only
checking for function pointer match, rather than also checking
for whether the registration was global.
Rather than playing with the guts of object_event ourselves
in domain_event, it is nicer to add a mapping function for the
internal callback id, then share common code for event removal.
For now, the function-to-id mapping is used only internally;
I thought about whether a new public API to let a user learn
the callback would be useful, but decided exposing this to the
user is probably a disservice, since we already publicly
document that they should avoid the old style, and since this
patch already demonstrates that older libvirt versions have
weird behavior when mixing old and new styles.
And like all good bug fix patches, I enhanced the testsuite,
validating that the changes in tests/ expose the failure
without the rest of the patch.
* src/conf/object_event.c (virObjectEventCallbackLookup)
(virObjectEventStateCallbackID): New functions.
(virObjectEventCallbackLookup): Use helper function.
* src/conf/object_event_private.h (virObjectEventStateCallbackID):
Declare new function.
* src/conf/domain_event.c (virDomainEventStateRegister)
(virDomainEventStateDeregister): Let common code handle the
complexity.
(virDomainEventCallbackListRemove)
(virDomainEventCallbackListMarkDelete)
(virDomainEventCallbackListAdd): Drop unused functions.
* tests/objecteventtest.c (testDomainCreateXMLMixed): New test.
Signed-off-by: Eric Blake <eblake@redhat.com>
Since the introduction of network events, any driver that uses
a single event state object to track both domain and network
events should not include 'domain' in the name of that object.
* src/test/test_driver.c (_testConn):
s/domainEventState/eventState/, and fix all callers.
* src/remote/remote_driver.c (private_data): Likewise.
(remoteDomainEventQueue): Rename to remoteEventQueue.
(remoteDomainEvents): Rename to remoteEvents.
Signed-off-by: Eric Blake <eblake@redhat.com>
Prior to this patch, every test:/// URI has its own event manager,
which means that registering for an event can only ever receive
events from the connection where it issued the API that triggered
the event. But the whole idea of events is to be able to learn
about something where an API call did NOT trigger the action.
In order to actually test asynchronous events, I wanted to be able
to tie multiple test connections to the same state. Use of a file
in a test URI is still per-connection state, but now parallel
connections to test:///default (from the same binary, of course)
now share common state and can affect one another.
The updated testsuite fails without the rest of this patch.
Valgrind didn't report any leaks.
* src/test/test_driver.c (testConnectOpen): Move per-connection
state initialization...
(testOpenFromFile): ...here.
(defaultConn, defaultConnections, defaultLock, testOnceInit): New
shared state.
(testOpenDefault): Only initialize on first connection.
(testConnectClose): Don't clobber state if still shared.
* tests/objecteventtest.c (testDomainStartStopEvent): Enhance to
cover this.
(timeout, mymain): Ensure test fails rather than blocks.
Signed-off-by: Eric Blake <eblake@redhat.com>
The @name variable is VIR_STRDUP()-ed into, but never freed. In fact,
there's no need to duplicate a command line argument since all places
where @name is used expect const char.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
CVE-2013-6458
Generally, every API that is going to begin a job should do that before
fetching data from vm->def. However, qemuDomainGetBlockInfo does not
know whether it will have to start a job or not before checking vm->def.
To avoid using disk alias that might have been freed while we were
waiting for a job, we use its copy. In case the disk was removed in the
meantime, we will fail with "cannot find statistics for device '...'"
error message.
CVE-2013-6458
https://bugzilla.redhat.com/show_bug.cgi?id=1043069
When virDomainDetachDeviceFlags is called concurrently to
virDomainBlockStats: libvirtd may crash because qemuDomainBlockStats
finds a disk in vm->def before getting a job on a domain and uses the
disk pointer after getting the job. However, the domain in unlocked
while waiting on a job condition and thus data behind the disk pointer
may disappear. This happens when thread 1 runs
virDomainDetachDeviceFlags and enters monitor to actually remove the
disk. Then another thread starts running virDomainBlockStats, finds the
disk in vm->def, and while it's waiting on the job condition (owned by
the first thread), the first thread finishes the disk removal. When the
second thread gets the job, the memory pointed to be the disk pointer is
already gone.
That said, every API that is going to begin a job should do that before
fetching data from vm->def.
This patch fixes a segmentation fault when creating new virtual machines using QEMU.
The segmentation fault is caused by commit f41830680e
and commit cbb6ec42e2.
In virQEMUCapsProbeQMPMachineTypes, when copying machines to qemuCaps, "none" is skipped.
Therefore, the value of i and "qemuCaps->nmachineTypes - 1" do not always match.
However, defIdx value (used to call virQEMUCapsSetDefaultMachine) is set using the value in i
when the array elements are in qemuCaps->nmachineTypes - 1.
So, when libvirt tries to create virtual machines using the default machine type,
qemuCaps->machineTypes[defIdx] is accessed and since the defIdx is NULL, it results in segmentation fault.
Signed-off-by: Yudai Yamagishi <yummy@sfc.wide.ad.jp>
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Cleanup after commit db3dd08 removed all clients outside of
the .h file.
* src/datatypes.h (VIR_IS_CONNECT): Delete, and inline into all
callers, since no other file uses it any more.
Signed-off-by: Eric Blake <eblake@redhat.com>
Ever since their introduction (commit 1509b80 in v0.5.0 for
virConnectDomainEventRegister, commit 4445723 in v0.8.0 for
virConnectDomainEventDeregisterAny), the event deregistration
functions have been documented as returning 0 on success;
likewise for older registration (only the newer RegisterAny
must return a non-zero callbackID). And now that we are
adding virConnectNetworkEventDeregisterAny for v1.2.1, it
should have the same semantics.
Fortunately, all of the stateful drivers have been obeying
the docs and returning 0, thanks to the way the remote_driver
tracks things (in fact, the RPC wire protocol is unable to
send a return value for DomainEventRegisterAny, at least not
without adding a new RPC number). Well, except for vbox,
which was always failing deregistration, due to failure to
set the return value to anything besides its initial -1.
But for local drivers, such as test:///default, we've been
returning non-zero numbers; worse, the non-zero numbers have
differed over time. For example, in Fedora 12 (libvirt 0.8.2),
calling Register twice would return 0 and 1 [the callbackID
generated under the hood]; while in Fedora 20 (libvirt 1.1.3),
it returns 1 and 2 [the number of callbacks registered for
that event type]. Since we have changed the behavior over
time, and since it differs by local vs. remote, we can safely
argue that no one could have been reasonably relying on any
particular behavior, so we might as well obey the docs, as well
as prepare callers that might deal with older clients to not be
surprised if the docs are not strictly followed.
For consistency, this patch fixes the code for all drivers,
even though it only makes an impact for vbox and for local
drivers. By fixing all drivers, future copy and paste from
a remote driver to a local driver is less likely to
reintroduce the bug.
Finally, update the testsuite to gain some coverage of the
issue for local drivers, including the first test of old-style
domain event registration via function pointer instead of
event id.
* src/libvirt.c (virConnectDomainEventRegister)
(virConnectDomainEventDeregister)
(virConnectDomainEventDeregisterAny): Clarify docs.
* src/libxl/libxl_driver.c (libxlConnectDomainEventRegister)
(libxlConnectDomainEventDeregister)
(libxlConnectDomainEventDeregisterAny): Match documentation.
* src/lxc/lxc_driver.c (lxcConnectDomainEventRegister)
(lxcConnectDomainEventDeregister)
(lxcConnectDomainEventDeregisterAny): Likewise.
* src/test/test_driver.c (testConnectDomainEventRegister)
(testConnectDomainEventDeregister)
(testConnectDomainEventDeregisterAny)
(testConnectNetworkEventDeregisterAny): Likewise.
* src/uml/uml_driver.c (umlConnectDomainEventRegister)
(umlConnectDomainEventDeregister)
(umlConnectDomainEventDeregisterAny): Likewise.
* src/vbox/vbox_tmpl.c (vboxConnectDomainEventRegister)
(vboxConnectDomainEventDeregister)
(vboxConnectDomainEventDeregisterAny): Likewise.
* src/xen/xen_driver.c (xenUnifiedConnectDomainEventRegister)
(xenUnifiedConnectDomainEventDeregister)
(xenUnifiedConnectDomainEventDeregisterAny): Likewise.
* src/network/bridge_driver.c
(networkConnectNetworkEventDeregisterAny): Likewise.
* tests/objecteventtest.c (testDomainCreateXMLOld): New test.
(mymain): Run it.
(testDomainCreateXML): Check return values.
Signed-off-by: Eric Blake <eblake@redhat.com>
Currently, the qemuProcessStop tries to open the domain log file
and saves the original error afterwards. Then all the cleanup is
done after which the error is restored back. This has however one
flaw: if opening of the log file fails an error is reported,
which results in previous error being overwritten (the useful
one, e.g. "PCI device XXXX:XXXX could not be found"). Hence, user
sees something like:
error: failed to create logfile /var/log/libvirt/qemu/ovirt_usb.log: No such file or directory
instead of:
error: internal error: Did not find USB device 8644:8003
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reported-by: Zhou Yimin <zhouyimin@huawei.com>
@listenAddress and @cookiein arguments, should be exchanged,
because the order of the caller and the callee does not match.
This results in the listen address being ignored for peer-to-peer
migration and the cookie being ignored for v2 migration.
Introduced by c4ac7ef (v1.1.4-rc1~141).
https://bugzilla.redhat.com/show_bug.cgi?id=1049338
Signed-off-by: Minoru Usui <usui@mxm.nes.nec.co.jp>
Signed-off-by: Ján Tomko <jtomko@redhat.com>
The libvirt_internal.h header was included by the internal.h header.
This made it painful to add new stuff to the header file that would
require some more specific types. Remove inclusion by internal.h and add
it to appropriate places manually.
The datatype.c object checks could result in a message like:
error: invalid connection pointer in no connection
This consolidates all clients of this message to have uniform contents:
error: invalid connection pointer in someFunc
Note that virCheckConnectReturn raises an error immediately; in
datatypes.c, where we don't need to raise the error (but instead
just leave it in the thread-local setting), we use
virCheckConnectGoto and the cleanup label instead. Then, for
consistency in that file, all subsequent error messages are
touched to also use the cleanup error label.
* src/datatypes.h (virCheckConnectReturn)
(virCheckConnectGoto): New macros.
* src/datatypes.c: Use new macro.
* src/libvirt-qemu.c (virDomainQemuAttach): Likewise.
(virLibConnError): Delete unused macro.
* src/libvirt-lxc.c (virLibConnError): Likewise.
* src/libvirt.c: Use new macro throughout.
* docs/api_extension.html.in: Modernize documentation.
Signed-off-by: Eric Blake <eblake@redhat.com>
As pointed out by the Xen folks [1], HVM nics should always be set
to type LIBXL_NIC_TYPE_VIF_IOEMU unless the user explicity requests
LIBXL_NIC_TYPE_VIF via model='netfront'. The current logic in
libxlMakeNic() only sets the nictype to LIBXL_NIC_TYPE_VIF_IOEMU if
a model is specified that is not 'netfront', which breaks PXE booting
configurations where no model is specified (i.e. use the hypervisor
default).
Reported-by: Stefan Bader <stefan.bader@canonical.com>
[1] https://www.redhat.com/archives/libvir-list/2013-December/msg01156.html
This reverts commit aa4619337c.
This patch was accidentally pushed prematurely, and has incorrect
logic for which shutdown methods to attempt.
Signed-off-by: Eric Blake <eblake@redhat.com>
AArch64 qemu has similar behavior as armv7l, like use of mmio etc.
This patch adds similar bypass checks what we have for armv7l to aarch64.
E.g. we are enabling mmio transport for Nicdev.
Making addDefaultUSB and addDefaultMemballoon to false etc.
V3:
- Adding missing domain rng schema for aarcg64 and test case in
testutilsqemu.c which was causing test suite failure
while running make check.
V2:
- Added testcase to qemuxml2argvtest as suggested
during review comments of V1.
V1:
- Initial patch.
Signed-off-by: Anup Patel <anup.patel@linaro.org>
Signed-off-by: Pranavkumar Sawargaonkar <pranavkumar@linaro.org>
Some of our operation denied messages are outright stupid; for
example, if virIdentitySetAttr fails:
error: operation Identity attribute is already set forbidden for read only access
This patch fixes things to a saner:
error: operation forbidden: Identity attribute is already set
It also consolidates the most common usage pattern for operation
denied errors: read-only connections preventing a public API. In
this case, 'virsh -r -c test:///default destroy test' changes from:
error: operation virDomainDestroy forbidden for read only access
to:
error: operation forbidden: read only access prevents virDomainDestroy
Note that we were previously inconsistent on which APIs used
VIR_FROM_DOM (such as virDomainDestroy) vs. VIR_FROM_NONE (such as
virDomainPMSuspendForDuration). After this patch, all uses
consistently use VIR_FROM_NONE, on the grounds that it is unlikely
that a caller learning that a call is denied can do anything in
particular with extra knowledge which error domain the call belongs
to (similar to what we did in commit baa7244).
* src/util/virerror.c (virErrorMsg): Rework OPERATION_DENIED error
message.
* src/internal.h (virCheckReadOnlyGoto): New macro.
* src/util/virerror.h (virReportRestrictedError): New macro.
* src/libvirt-lxc.c: Use new macros.
* src/libvirt-qemu.c: Likewise.
* src/libvirt.c: Likewise.
* src/locking/lock_daemon.c (virLockDaemonClientNew): Likewise.
Signed-off-by: Eric Blake <eblake@redhat.com>
https://bugzilla.redhat.com/show_bug.cgi?id=1047234
Add a range check for supported numa memory placement modes provided by
the user before setting them in the domain definition. Without the check
the user is able to provide a (yet) unknown mode which is then stored in
the domain definition. This potentially causes a NULL dereference when
the defintion is formatted into the XML.
To reproduce run:
virsh numatune DOMNAME --mode 6 --nodeset 0
The XML will then contain:
<numatune>
<memory mode='(null)' nodeset='0'/>
</numatune>
With this fix, the command fails:
error: Unable to change numa parameters
error: invalid argument: unsupported numa_mode: '6'
Add whitespace to separate logical code blocks, reformat error messages
and clean up code flow.
This patch changes error handling in some cases where the the loop would
be continued to jump to cleanup instead and error out rather than modify
the domain any further.
We might as well take advantage of viralloc.h instead of open-coding
array management ourselves. While at it, I simplified several
places that were doing repetitive pointer chasing to use an
intermediate variable for legibility (some other places remain,
but they will disapper in later refactoring patches).
* src/conf/object_event_private.h (_virObjectEventCallbackList):
Use size_t for count.
* src/conf/object_event.c (_virObjectEventQueue): Likewise.
(virObjectEventCallbackListRemoveID): Use VIR_DELETE_ELEMENT.
(virObjectEventQueuePush, virObjectEventCallbackListAddID): Use
VIR_APPEND_ELEMENT.
(virObjectEventCallbackListEventID)
(virObjectEventStateDispatchCallbacks): Simplify code.
Signed-off-by: Eric Blake <eblake@redhat.com>
No need to use an int that only ever stores 0 and 1.
* src/conf/object_event_private.h (_virObjectEventCallback):
Change deleted to bool.
* src/conf/object_event.c (virObjectEventDispatchMatchCallback):
Switch return type to bool.
(virObjectEventCallbackListMarkDeleteID): Update client.
* src/conf/domain_event.c (virDomainEventCallbackListMarkDelete):
Likewise.
Do not leave the PCI address of the primary video card set
to the legacy default (0000:00:02.0) if we're doing two-pass
allocation.
Since QEMU 1.6 (QEMU_CAPS_VIDEO_PRIMARY) we allow the primary
video card to be on other slots than 0000:00:02.0 (as we use
-device instead of -vga).
However we fail to assign it an address if:
* another device explicitly uses 0000:00:02.0 and
* the primary video device has no address specified
On the first pass, we have set the address to default, then checked
if it's available, leaving it set even if it wasn't. This address
got picked up by the second pass, resulting in a conflict:
XML error: Attempted double use of PCI slot 0000:00:02.0
(may need "multifunction='on'" for device on function 0)
Also fix the test that was supposed to catch this.
We weren't very consistent in our use of VIR_ERR_NO_SUPPORT; many
users just passed __FUNCTION__ on, while others passed "%s" to
silence over-eager compilers that warn about __FUNCTION__ not
containing any %. It's nicer to route all these uses through
a single macro, so that if we ever need to change the reporting,
we can do it in one place.
I verified that 'virsh -c test:///default qemu-monitor-command test foo'
gives the same error message before and after this patch:
error: this function is not supported by the connection driver: virDomainQemuMonitorCommand
Note that in libvirt.c, we were inconsistent on whether virDomain*
API used virLibConnError() (with VIR_FROM_NONE) or virLibDomainError()
(with VIR_FROM_DOMAIN); this patch unifies these errors to all use
VIR_FROM_NONE, on the grounds that it is unlikely that a caller
learning that a call is unimplemented can do anything in particular
with extra knowledge of which error domain it belongs to.
One particular change to note is virDomainOpenGraphics which was
trying to fail with VIR_ERR_NO_SUPPORT after a failed
VIR_DRV_SUPPORTS_FEATURE check; all other places that fail a
feature check report VIR_ERR_ARGUMENT_UNSUPPORTED.
* src/util/virerror.h (virReportUnsupportedError): New macro.
* src/libvirt-qemu.c: Use new macro.
* src/libvirt-lxc.c: Likewise.
* src/lxc/lxc_driver.c: Likewise.
* src/security/security_manager.c: Likewise.
* src/util/virinitctl.c: Likewise.
* src/libvirt.c: Likewise.
(virDomainOpenGraphics): Use correct error for unsupported feature.
Signed-off-by: Eric Blake <eblake@redhat.com>
