When qemuMonitorSetCapabilities() fails, there is no need to
call qemuMonitorClose(), because the caller will already see
the error code and tear down the entire VM. The extra call to
qemuMonitorClose resulted in a double-free due to it removing
a ref count prematurely.
* src/qemu/qemu_driver.c: Remove premature close of monitor
Regression in commit caa805ea let a lot of bad messages slip in.
* cfg.mk (msg_gen_function): Fix function name.
* src/qemu/qemu_cgroup.c (qemuRemoveCgroup): Fix fallout from
'make syntax-check'.
* src/qemu/qemu_driver.c (qemudDomainGetInfo)
(qemuDomainWaitForMigrationComplete, qemudStartVMDaemon)
(qemudDomainSaveFlag, qemudDomainAttachDevice)
(qemuDomainUpdateDeviceFlags): Likewise.
* src/qemu/qemu_hotplug.c (qemuDomainAttachHostUsbDevice)
(qemuDomainDetachPciDiskDevice, qemuDomainDetachSCSIDiskDevice):
Likewise.
When attaching device from a xml file and the device is mis-configured,
virsh gives mis-leading message "out of memory". This patch fixes this.
Signed-off-by: Eric Blake <eblake@redhat.com>
* src/qemu/qemu_command.c (qemuBuildChrChardevStr): Alter the
chardev alias.
(qemuBuildCommandLine): Output an id for the chardev counterpart.
* tests/qemuxml2argvdata/*: Update tests to match.
Reported by Daniel P. Berrange.
Steps to reproduce this bug:
1. service libvirtd start
2. virsh start <domain>
3. kill -STOP $(cat /var/run/libvirt/qemu/<domain>.pid)
4. service libvirtd restart
5. kill -9 $(cat /var/run/libvirt/qemu/<domain>.pid)
Then libvirtd will core dump or be in deadlock state.
Make sure that json is built into libvirt and the version
of qemu is newer than 0.13.0.
The reason of libvirtd cores dump is that:
We add vm->refs when we alloc the memory, and decrease it
in the function qemuHandleMonitorEOF() in other thread.
We add vm->refs in the function qemuConnectMonitor() and
decrease it when the vm is inactive.
The libvirtd will block in the function qemuMonitorSetCapabilities()
because the vm is stopped by signal SIGSTOP. Now the vm->refs is 2.
Then we kill the vm by signal SIGKILL. The function
qemuMonitorSetCapabilities() failed, and then we will decrease vm->refs
in the function qemuMonitorClose().
In another thread, mon->fd is broken and the function
qemuHandleMonitorEOF() is called.
If qemuHandleMonitorEOF() decreases vm->refs before qemuConnectMonitor()
returns, vm->refs will be decrease to 0 and the memory is freed.
We will call qemudShutdownVMDaemon() as qemuConnectMonitor() failed.
The memory has been freed, so qemudShutdownVMDaemon() is too dangerous.
We will reference NULL pointer in the function virDomainConfVMNWFilterTeardown():
=============
void
virDomainConfVMNWFilterTeardown(virDomainObjPtr vm) {
int i;
if (nwfilterDriver != NULL) {
for (i = 0; i < vm->def->nnets; i++)
virDomainConfNWFilterTeardown(vm->def->nets[i]);
}
}
============
vm->def->nnets is not 0 but vm->def->nets is NULL(We don't set vm->def->nnets
to 0 when we free vm).
We should add an extra reference of vm to avoid vm to be deleted if
qemuConnectMonitor() failed.
Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
If a guest image is saved in compressed format, and the restore fails
in some way after the intermediate process used to uncompress the
image has been started, but before qemu has been started to hook up to
the uncompressor, libvirt will endlessly wait for the uncompressor to
finish, but it never will because it's still waiting to have something
hooked up to drain its output.
The solution is to close the pipes on both sides of the uncompressor,
then send a SIGTERM before calling waitpid on it (only if the restore
has failed, of course).
This patch is a partial resolution to the following bug:
https://bugzilla.redhat.com/show_bug.cgi?id=667756
(to complete the fix, an updated selinux-policy package is required,
to add the policy that allows libvirt to set the context of a fifo,
which was previously not allowed).
Explanation : When an incoming migration is over a pipe (for example,
if the image was compressed and is being fed through gzip, or was on a
root-squash nfs server, so needed to be opened by a child process
running as a different uid), qemu cannot read it unless the selinux
context label for the pipe has been set properly.
The solution is to check the fd used as the source of the migration
just before passing it to qemu; if it's a fifo (implying that it's a
pipe), we call the newly added virSecurityManagerSetFDLabel() function
to set the context properly.
The problem was introduced by commit 4303c91, which removed the checking
of domain state, this patch is to fix it.
Otherwise, improper error will be thrown, e.g.
error: Failed to save domain rhel6 state
error: cannot resolve symlink /var/lib/libvirt/qemu/save/rhel6.save: No such
file or directory
In QEMU, the card itself is a PCI device, but it requires a codec
(either -device hda-output or -device hda-duplex) to actually output
sound. Specifying <sound model='ich6'/> gives us -device intel-hda
-device hda-duplex I think it's important that a simple <sound model='ich6'/>
sets up a useful codec, to have consistent behavior with all other sound cards.
This is basically Dan's proposal of
<sound model='ich6'>
<codec type='output' slot='0'/>
<codec type='duplex' slot='3'/>
</sound>
without the codec bits implemented.
The important thing is to keep a consistent API here, we don't want some
<sound> devs require tweaking codecs but not others. Steps I see to
accomplishing this:
- every <sound> device has a <codec type='default'/> (unless codecs are
manually specified)
- <codec type='none'/> is required to specify 'no codecs'
- new audio settings like mic=on|off could then be exposed in
<sound> or <codec> in a consistent manner for all sound models
v2:
Use model='ich6'
v3:
Use feature detection, from eblake
Set codec id, bus, and cad values
v4:
intel-hda isn't supported if -device isn't available
v5:
Comment spelling fixes
If vnc_auto_unix_socket is enabled, any VNC devices without a hardcoded
listen or socket value will be setup to serve over a unix socket in
/var/lib/libvirt/qemu/$vmname.vnc.
We store the generated socket path in the transient VM definition at
CLI build time.
QEMU supports serving VNC over a unix domain socket rather than traditional
TCP host/port. This is specified with:
<graphics type='vnc' socket='/foo/bar/baz'/>
This provides better security access control than VNC listening on
127.0.0.1, but will cause issues with tools that rely on the lax security
(virt-manager in fedora runs as regular user by default, and wouldn't be
able to access a socket owned by 'qemu' or 'root').
Also not currently supported by any clients, though I have patches for
virt-manager, and virt-viewer should be simple to update.
v2:
schema: Make listen vs. socket a <choice>
This will allow us to record transient runtime state in vm->def, like
default VNC parameters. Accomplish this by adding an extra 'live' parameter
to SetDefTransient, with similar semantics to the 'live' flag for
AssignDef.
When restoring a saved qemu instance via JSON monitor, the vm is
left in a paused state. Turns out the 'cont' cmd was failing with
"MigrationExpected" error class and "An incoming migration is
expected before this command can be executed" error description
due to migration (restore) not yet complete.
Detect if 'cont' cmd fails with "MigrationExpecte" error class and
retry 'cont' cmd.
V2: Fix potential double-free noted by Laine Stump
Report VIR_ERR_CONFIG_UNSUPPORTED instead of VIR_ERR_INTERNAL_ERROR,
as it's valid in our domain schema, just unsupported by hypervisor
here.
* src/qemu/qemu_command.c
The code which set VNC passwords correctly had fallback for
the set_password command, but was lacking it for the
expire_password command. This made it impossible to start
a guest. It also failed to check whether QEMU was still
running after the initial 'set_password' command completed
* src/qemu/qemu_hotplug.c: Fix error handling when
password expiry fails
* src/qemu/qemu_monitor_json.c, src/qemu/qemu_monitor_text.c: Fix
return code for missing expire_password command
Avoid overwriting the real error message with a generic
OOM failure message, when machine type probe fails
* src/qemu/qemu_driver.c: Don't overwrite error
The function virUnrefConnect() may call virReleaseConnect() to release
the dest connection, and the function virReleaseConnect() will call
conn->driver->close().
So the function virUnrefConnect() should be surrounded by
qemuDomainObjEnterRemoteWithDriver() and
qemuDomainObjExitRemoteWithDriver() to prevent possible deadlock between
two communicating libvirt daemons.
See commit f0c8e1cb37 for further details.
Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
In some circumstances, libvirtd would issue two STOPPED events after it
stopped a domain. This was because an EOF event can arrive after a qemu
process is killed but before qemuMonitorClose() is called.
qemuHandleMonitorEOF() should ignore EOF when the domain is not running.
I wasn't able to reproduce this bug directly, only after adding an
artificial sleep() into qemudShutdownVMDaemon().
Fixes regression introduced in commit 2211518, where all qemu 0.12.x
fails to start, as does qemu 0.13.x lacking the pci-assign device.
Prior to 2211518, the code was just ignoring a non-zero exit status
from the qemu child, but the virCommand code checked this to avoid
masking any other issues, which means the real bug of provoking
non-zero exit status has been latent for a longer time.
* src/qemu/qemu_capabilities.c (qemuCapsExtractVersionInfo): Check
for -device driver,? support.
(qemuCapsExtractDeviceStr): Avoid failure if all probed devices
are unsupported.
Reported by Ken Congyang.
https://bugzilla.redhat.com/show_bug.cgi?id=620363
When using -incoming stdio or -incoming exec:, qemu keeps the
stdin fd open long after the migration is complete. Not to
mention that exec:cat is horribly inefficient, by doubling the
I/O and going through a popen interface in qemu.
The new -incoming fd: of qemu 0.12.0 closes the fd after using
it, and allows us to bypass an intermediary cat process for
less I/O.
* src/qemu/qemu_command.h (qemuBuildCommandLine): Add parameter.
* src/qemu/qemu_command.c (qemuBuildCommandLine): Support
migration via fd: when possible. Consolidate migration handling
into one spot, now that it is more complex.
* src/qemu/qemu_driver.c (qemudStartVMDaemon): Update caller.
* tests/qemuxml2argvtest.c (mymain): Likewise.
* tests/qemuxml2argvdata/qemuxml2argv-restore-v2-fd.args: New file.
* tests/qemuxml2argvdata/qemuxml2argv-restore-v2-fd.xml: Likewise.
Display or set unlimited values for memory parameters. Unlimited is
represented by INT64_MAX in memory cgroup.
Signed-off-by: Nikunj A. Dadhania <nikunj@linux.vnet.ibm.com>
Reported-by: Justin Clift <jclift@redhat.com>
This is in response to a request in:
https://bugzilla.redhat.com/show_bug.cgi?id=665293
In short, under heavy load, it's possible for qemu's networking to
lock up due to the tap device's default 1MB sndbuf being
inadequate. adding "sndbuf=0" to the qemu commandline -netdevice
option will alleviate this problem (sndbuf=0 actually sets it to
0xffffffff).
Because we must be able to explicitly specify "0" as a value, the
standard practice of "0 means not specified" won't work here. Instead,
virDomainNetDef also has a sndbuf_specified, which defaults to 0, but
is set to 1 if some value was given.
The sndbuf value is put inside a <tune> element of each <interface> in
the domain. The intent is that further tunable settings will also be
placed inside this element.
<interface type='network'>
...
<tune>
<sndbuf>0</sndbuf>
...
</tune>
</interface>
This patch is in response to
https://bugzilla.redhat.com/show_bug.cgi?id=643050
The existing libvirt support for the vhost-net backend to the virtio
network driver happens automatically - if the vhost-net device is
available, it is always enabled, otherwise the standard userland
virtio backend is used.
This patch makes it possible to force whether or not vhost-net is used
with a bit of XML. Adding a <driver> element to the interface XML, eg:
<interface type="network">
<model type="virtio"/>
<driver name="vhost"/>
will force use of vhost-net (if it's not available, the domain will
fail to start). if driver name="qemu", vhost-net will not be used even
if it is available.
If there is no <driver name='xxx'/> in the config, libvirt will revert
to the pre-existing automatic behavior - use vhost-net if it's
available, and userland backend if vhost-net isn't available.
We try to use that command first when setting a VNC/SPICE password. If
that doesn't work we fallback to the legacy VNC only password
Allow an expiry time to be set, if that doesn't work, throw an error
if they try to use SPICE.
Change since v1:
- moved qemuInitGraphicsPasswords to qemu_hotplug, renamed
to qemuDomainChangeGraphicsPasswords.
- updated what looks like a typo (that appears to work anyway) in
initial patch from Daniel:
- ret = qemuInitGraphicsPasswords(driver, vm,
- VIR_DOMAIN_GRAPHICS_TYPE_SPICE,
- &vm->def->graphics[0]->data.vnc.auth,
- driver->vncPassword);
+ ret = qemuInitGraphicsPasswords(driver, vm,
+ VIR_DOMAIN_GRAPHICS_TYPE_SPICE,
+ &vm->def->graphics[0]->data.spice.auth,
+ driver->spicePassword);
Based on patch by Daniel P. Berrange <berrange@redhat.com>.
* src/qemu/qemu_capabilities.h (qemuCapsParseDeviceStr): New
prototype.
* src/qemu/qemu_capabilities.c (qemuCapsParsePCIDeviceStrs)
Rename and split...
(qemuCapsExtractDeviceStr, qemuCapsParseDeviceStr): ...to make it
easier to add and test device-specific checks.
(qemuCapsExtractVersionInfo): Update caller.
* tests/qemuhelptest.c (testHelpStrParsing): Also test parsing of
device-related flags.
(mymain): Update expected flags.
* tests/qemuhelpdata/qemu-0.12.1-device: New file.
* tests/qemuhelpdata/qemu-kvm-0.12.1.2-rhel60-device: New file.
* tests/qemuhelpdata/qemu-kvm-0.12.3-device: New file.
* tests/qemuhelpdata/qemu-kvm-0.13.0-device: New file.
* src/qemu/qemu_capabilities.c (qemuCapsProbeMachineTypes)
(qemuCapsProbeCPUModels, qemuCapsParsePCIDeviceStrs)
(qemuCapsExtractVersionInfo): Use virCommand rather than virExec.
If the emulator doesn't support SDL graphic, we should reject
the use of SDL graphic xml with error messages, but not ignore
it silently, and pretend things are fine.
"-sdl" flag was exposed explicitly by qemu since 0.10.0, more detail:
http://www.redhat.com/archives/libvir-list/2011-January/msg00442.html
And we already have capability flag "QEMUD_CMD_FLAG_0_10", which
could be used to prevent the patch affecting the older versions
of QEMU.
* src/qemu/qemu_command.c
Skip IB700 when assigning PCI slots.
Note: the I6300ESB watchdog _is_ a PCI device.
To test this: I applied this patch to libvirt-0.8.3-2.fc14 (rebasing
it slightly: qemu_command.c didn't exist in that version) and
installed this on my machine, then tested that I could successfully
add an ib700 watchdog device to a guest, start the guest, and the
ib700 was available to the guest. I also added an i6300esb (PCI)
watchdog to another guest, and verified that libvirt assigned a PCI
device to it, that the guest could be started, and that i6300esb was
present in the guest.
Note that if you previously had a domain with a ib700 watchdog, it
would have had an <address type='pci' .../> clause added to it in the
libvirt configuration. This patch does not attempt to remove this.
You cannot start such a domain -- qemu gives an error if you try.
With this patch you are able to remove the bogus address element
without libvirt adding it back.
Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
The current security driver usage requires horrible code like
if (driver->securityDriver &&
driver->securityDriver->domainSetSecurityHostdevLabel &&
driver->securityDriver->domainSetSecurityHostdevLabel(driver->securityDriver,
vm, hostdev) < 0)
This pair of checks for NULL clutters up the code, making the driver
calls 2 lines longer than they really need to be. The goal of the
patchset is to change the calling convention to simply
if (virSecurityManagerSetHostdevLabel(driver->securityDriver,
vm, hostdev) < 0)
The first check for 'driver->securityDriver' being NULL is removed
by introducing a 'no op' security driver that will always be present
if no real driver is enabled. This guarentees driver->securityDriver
!= NULL.
The second check for 'driver->securityDriver->domainSetSecurityHostdevLabel'
being non-NULL is hidden in a new abstraction called virSecurityManager.
This separates the driver callbacks, from main internal API. The addition
of a virSecurityManager object, that is separate from the virSecurityDriver
struct also allows for security drivers to carry state / configuration
information directly. Thus the DAC/Stack drivers from src/qemu which
used to pull config from 'struct qemud_driver' can now be moved into
the 'src/security' directory and store their config directly.
* src/qemu/qemu_conf.h, src/qemu/qemu_driver.c: Update to
use new virSecurityManager APIs
* src/qemu/qemu_security_dac.c, src/qemu/qemu_security_dac.h
src/qemu/qemu_security_stacked.c, src/qemu/qemu_security_stacked.h:
Move into src/security directory
* src/security/security_stack.c, src/security/security_stack.h,
src/security/security_dac.c, src/security/security_dac.h: Generic
versions of previous QEMU specific drivers
* src/security/security_apparmor.c, src/security/security_apparmor.h,
src/security/security_driver.c, src/security/security_driver.h,
src/security/security_selinux.c, src/security/security_selinux.h:
Update to take virSecurityManagerPtr object as the first param
in all callbacks
* src/security/security_nop.c, src/security/security_nop.h: Stub
implementation of all security driver APIs.
* src/security/security_manager.h, src/security/security_manager.c:
New internal API for invoking security drivers
* src/libvirt.c: Add missing debug for security APIs
When dynamic_ownership=0, saved images must be owned by the same uid
as is used to run the qemu process, otherwise restore won't work. To
accomplish this, qemuSecurityDACRestoreSavedStateLabel() needs to
simply return when it's called.
This fix is in response to:
https://bugzilla.redhat.com/show_bug.cgi?id=661720
While doing some testing with Qemu and creating huge logfiles I encountered the case where the VM could not start anymore due to the lseek() to the end of the Qemu VM's log file failing. The patch below fixes the problem by replacing the previously used 'int' with 'off_t'.
To reproduce this error, you could do the following:
dd if=/dev/zero of=/var/log/libvirt/qemu/<name of VM>.log bs=1024 count=$((1024*2048))
and you should get an error like this:
error: Failed to start domain <name of VM>
error: Unable to seek to -2147482651 in /var/log/libvirt/qemu/<name of VM>.log: Success
This patch fixes https://bugzilla.redhat.com/show_bug.cgi?id=664406
If qemu is run as a different uid, it has been unable to access mode
0660 files that are owned by a different user, but with a group that
the qemu is a member of (aside from the one group listed in the passwd
file), because initgroups() is not being called prior to the
exec. initgroups will change the group membership of the process (and
its children) to match the new uid.
To make this happen, the setregid()/setreuid() code in
qemuSecurityDACSetProcessLabel has been replaced with a call to
virSetUIDGID(), which does both of those, plus calls initgroups.
Similar, but not identical, code in qemudOpenAsUID() has been replaced
with virSetUIDGID(). This not only consolidates the functionality to a
single location, but also potentially fixes some as-yet unreported
bugs.
Shorten qemuDomainSnapshotWriteSnapshotMetadata function name
and make it take a snapshot pointer instead of dealing with
the current snapshot. Update other functions accordingly.
Add a qemuDomainSnapshotReparentChildren hash iterator to
reparent the children of a snapshot that is being deleted. Use
qemuDomainSnapshotWriteMetadata to write updated metadata
to disk.
This fixes a problem where outdated parent information breaks
the snapshot tree and hinders the deletion of child snapshots.
Reported by Philipp Hahn.
Network disks are accessed by qemu directly, and have no
associated file on the host, so checking for file ownership etc.
is unnecessary.
Signed-off-by: Josh Durgin <joshd@hq.newdream.net>
The QEMU driver file is far too large. Move all the hotplug
helper code out into a separate file. No functional change.
* src/qemu/qemu_hotplug.c, src/qemu/qemu_hotplug.h,
src/Makefile.am: Add hotplug helper file
* src/qemu/qemu_driver.c: Delete hotplug code
To allow the APIs to be used from separate files, move the domain
lock / job helper code into qemu_domain.c
* src/qemu/qemu_domain.c, src/qemu/qemu_domain.h: Add domain lock
/ job code
* src/qemu/qemu_driver.c: Remove domain lock / job code
To allow their use from other source files, move qemuDriverLock
and qemuDriverUnlock to qemu_conf.h and make them non-static
* src/qemu/qemu_conf.c, src/qemu/qemu_conf.h: Add qemuDriverLock
qemuDriverUnlock
* src/qemu/qemu_driver.c: Remove qemuDriverLock and qemuDriverUnlock
The QEMU driver file is far too large. Move all the hostdev
helper code out into a separate file. No functional change.
* src/qemu/qemu_hostdev.c, src/qemu/qemu_hostdev.h,
src/Makefile.am: Add hostdev helper file
* src/qemu/qemu_driver.c: Delete hostdev code
The QEMU driver file is far too large. Move all the cgroup
helper code out into a separate file. No functional change.
* src/qemu/qemu_cgroup.c, src/qemu/qemu_cgroup.h,
src/Makefile.am: Add cgroup helper file
* src/qemu/qemu_driver.c: Delete cgroup code
The QEMU driver file is far too large. Move all the audit
helper code out into a separate file. No functional change.
* src/qemu/qemu_audit.c, src/qemu/qemu_audit.h,
src/Makefile.am: Add audit helper file
* src/qemu/qemu_driver.c: Delete audit code
Move the code for handling the QEMU virDomainObjPtr private
data, and custom XML namespace into a separate file
* src/qemu/qemu_domain.c, src/qemu/qemu_domain.h: New file
for private data & namespace code
* src/qemu/qemu_driver.c, src/qemu/qemu_driver.h: Remove
private data & namespace code
* src/qemu/qemu_driver.h, src/qemu/qemu_command.h: Update
includes
* src/Makefile.am: Add src/qemu/qemu_domain.c
The qemu_conf.c code is doing three jobs, driver config file
loading, QEMU capabilities management and QEMU command line
management. Move the command line code into its own file
* src/qemu/qemu_command.c, src/qemu/qemu_command.h: New
command line management code
* src/qemu/qemu_conf.c, src/qemu/qemu_conf.h: Delete command
line code
* src/qemu/qemu_conf.h, src/qemu_conf.c: Adapt for API renames
* src/Makefile.am: add src/qemu/qemu_command.c
* src/qemu/qemu_monitor_json.c, src/qemu/qemu_monitor_text.c: Add
import of qemu_command.h
The qemu_conf.c code is doing three jobs, driver config file
loading, QEMU capabilities management and QEMU command line
management. Move the capabilities code into its own file
* src/qemu/qemu_capabilities.c, src/qemu/qemu_capabilities.h: New
capabilities management code
* src/qemu/qemu_conf.c, src/qemu/qemu_conf.h: Delete capabilities
code
* src/qemu/qemu_conf.h: Adapt for API renames
* src/Makefile.am: add src/qemu/qemu_capabilities.c
Currently, all of domain "save/dump/managed save/migration"
use the same function "qemudDomainWaitForMigrationComplete"
to wait the job finished, but the error messages are all
about "migration", e.g. when a domain saving job is canceled
by user, "migration was cancled by client" will be throwed as
an error message, which will be confused for user.
As a solution, intoduce two new job types(QEMU_JOB_SAVE,
QEMU_JOB_DUMP), and set "priv->jobActive" to "QEMU_JOB_SAVE"
before saving, to "QEMU_JOB_DUMP" before dumping, so that we
could get the real job type in
"qemudDomainWaitForMigrationComplete", and give more clear
message further.
And as It's not important to figure out what's the exact job
is in the DEBUG and WARN log, also we don't need translated
string in logs, simply repace "migration" with "job" in some
statements.
* src/qemu/qemu_driver.c
Changes common to all network disks:
-Make source name optional in the domain schema, since NBD doesn't use it
-Add a hostName type to the domain schema, and use it instead of genericName, which doesn't include .
-Don't leak host names or ports
-Set the source protocol in qemuParseCommandline
Signed-off-by: Josh Durgin <joshd@hq.newdream.net>
This patch adds network disk support to libvirt/QEMU. The currently
supported protocols are nbd, rbd, and sheepdog. The XML syntax is like
this:
<disk type="network" device="disk">
<driver name="qemu" type="raw" />
<source protocol='rbd|sheepdog|nbd' name="...some image identifier...">
<host name="mon1.example.org" port="6000">
<host name="mon2.example.org" port="6000">
<host name="mon3.example.org" port="6000">
</source>
<target dev="vda" bus="virtio" />
</disk>
Signed-off-by: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
`dump' watchdog action lets libvirtd to dump the guest when receives a
watchdog event (which probably means a guest crash)
Currently only qemu is supported.
When we get an EOF event on monitor connection, it may be a result of
either crash or graceful shutdown. QEMU which supports async events
(i.e., we are talking to it using JSON monitor) emits SHUTDOWN event on
graceful shutdown. In case we don't get this event by the time monitor
connection is closed, we assume the associated domain crashed.
Currently libvirt doesn't confirm whether the guest has responded to the
disk removal request. In some cases this can leave the guest with
continued access to the device while the mgmt layer believes that it has
been removed. With a recent qemu monitor command[1] we can
deterministically revoke a guests access to the disk (on the QEMU side)
to ensure no futher access is permitted.
This patch adds support for the drive_del() command and introduces it
in the disk removal paths. If the guest is running in a QEMU without this
command we currently explicitly check for unknown command/CommandNotFound
and log the issue.
If QEMU supports the command we issue the drive_del command after we attempt
to remove the device. The guest may respond and remove the block device
before we get to attempt to call drive_del. In that case, we explicitly check
for 'Device not found' from the monitor indicating that the target drive
was auto-deleted upon guest responds to the device removal notification.
1. http://thread.gmane.org/gmane.comp.emulators.qemu/84745
Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
Currently libvirt doesn't confirm whether the guest has responded to the
disk removal request. In some cases this can leave the guest with
continued access to the device while the mgmt layer believes that it has
been removed. With a recent qemu monitor command[1] we can
deterministically revoke a guests access to the disk (on the QEMU side)
to ensure no futher access is permitted.
This patch adds support for the drive_unplug() command and introduces it
in the disk removal paths. There is some discussion to be had about how
to handle the case where the guest is running in a QEMU without this
command (and the fact that we currently don't have a way of detecting
what monitor commands are available).
Changes since v2:
- use VIR_ERROR to report when unplug command not found
Changes since v1:
- return > 0 when command isn't present, < 0 on command failure
- detect when drive_unplug command isn't present and log error
instead of failing entire command
Signed-off-by: Ryan Harper <ryanh@us.ibm.com>
- qemudDomainAttachPciControllerDevice: Don't build "devstr"
if "-device" of qemu is not available, as "devstr" will only
be used by "qemuMonitorAddDevice", which depends on "-device"
argument of qemu is supported.
- "qemudDomainSaveImageOpen": Fix indent problem.
* src/qemu/qemu_driver.c
Commit febc591683 introduced -vga none in
case no video card is included in domain XML. However, old qemu
versions do not support this and such domain cannot be successfully
started.
* src/conf/domain_conf.c (virDomainDefParseXML): Prefer sysinfo
uuid over generating one, and if both uuids are present, require
them to be identical.
* src/qemu/qemu_conf.c (qemuBuildSmbiosSystemStr): Allow skipping
the uuid.
(qemudBuildCommandLine): Adjust caller; <smbios mode=host/> must
not use host uuid in place of guest uuid.
The log lists things like -smbios type=1,vendor="Red Hat", which
is great for shell parsing, but not so great when you realize that
execve() then passes those literal "" on as part of the command
line argument, such that qemu sets SMBIOS with extra literal quotes.
The eventual addition of virCommand is needed before we have the API
to shell-quote a string representation of a command line, so that the
log can still be pasted into a shell, but without inserting extra
bytes into the execve() arguments.
* src/qemu/qemu_conf.c (qemuBuildSmbiosBiosStr)
(qemuBuildSmbiosSystemStr): Qemu doesn't like quotes around uuid
arguments, and the remaining quotes are passed literally to
smbios, making <smbios mode='host'/> inaccurate. Removing the
quotes makes the log harder to parse, but that can be fixed later
with virCommand improvements.
* tests/qemuxml2argvdata/qemuxml2argv-smbios.args: 'Fix' test; it
will need fixing again once virCommand learns how to shell-quote a
potential command line.
Humans consider January as month #1, while gmtime_r(3) calls it month #0.
While fixing it, render qemu's rtc parameter with leading zeros, as is more
commonplace.
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=660194
Two more calls to remote libvirtd have to be surrounded by
qemuDomainObjEnterRemoteWithDriver() and
qemuDomainObjExitRemoteWithDriver() to prevent possible deadlock between
two communicating libvirt daemons.
See commit f0c8e1cb37 for further details.
* src/qemu/qemu_conf.c (qemudExtractVersionInfo): Check for file
before executing it here, rather than in callers.
(qemudBuildCommandLine): Rewrite with virCommand.
* src/qemu/qemu_conf.h (qemudBuildCommandLine): Update signature.
* src/qemu/qemu_driver.c (qemuAssignPCIAddresses)
(qemudStartVMDaemon, qemuDomainXMLToNative): Adjust callers.
* src/qemu/qemu_driver.c (though MACROS QEMU_VNC_PORT_MAX, and
QEMU_VNC_PORT_MIN are defined at the beginning, numbers (65535, 5900)
are still used, replace them)
Use macvtap specific functions depending on WITH_MACVTAP.
Use #if instead of #ifdef to check for WITH_MACVTAP, because
WITH_MACVTAP is always defined with value 0 or 1.
Also export virVMOperationType{To|From}String unconditional,
because they are used unconditional in the domain config code.
When dumping a domain, it's reasonable to save dump-file in raw format
if dump format is misconfigured or the corresponding compress program
is not available rather then fail dumping.
This patch introduces the usage of the pre-associate state of the IEEE 802.1Qbg standard on incoming VM migration on the target host. It is in response to bugzilla entry 632750.
https://bugzilla.redhat.com/show_bug.cgi?id=632750
For being able to differentiate the exact reason as to why a macvtap device is being created, either due to a VM creation or an incoming VM migration, I needed to pass that reason as a parameter from wherever qemudStartVMDaemon is being called in order to determine whether to send an ASSOCIATE (VM creation) or a PRE-ASSOCIATE (incoming VM migration) towards lldpad.
I am also fixing a problem with the virsh domainxml-to-native call on the way.
Gerhard successfully tested the patch with a recent blade network 802.1Qbg-compliant switch.
The patch should not have any side-effects on the 802.1Qbh support in libvirt, but Roopa (cc'ed) may want to verify this.
When we set migrate_speed by json, we receive the following
error message:
libvirtError: internal error unable to execute QEMU command
'migrate_set_speed': Invalid parameter type, expected: number
The reason is that: the arguments of migrate_set_speed
by json is json number, not json string.
Signed-off-by: Wen Congyang <wency@cn.fujitsu.com>
When running non-root, the QEMU log file is usually opened with
truncation, since there is no logrotate for non-root usage.
This means that when libvirt logs the shutdown timestamp, the
log is accidentally truncated
* src/qemu/qemu_driver.c: Never truncate log file with shutdown
message
Do this by adding a helper function to get the persistent domain config. This
should be useful for other functions that may eventually want to alter
the persistent domain config (attach/detach device). Also make similar changes
to the test drivers setvcpus command.
A caveat is that the function will return the running config for a transient
domain, rather than error. This simplifies callers, as long as they use
other methods to ensure the guest is persistent.
Doing 'virsh setvcpus $vm --config 10' doesn't check the value against the
domains maxvcpus value. A larger value for example will prevent the guest
from starting.
Also make a similar change to the test driver.
The current semantics of non-persistent hotplug/update are confusing: the
changes will persist as long as the in memory domain definition isn't
overwritten. This means hotplug changes stay around until the domain is
redefined or libvirtd is restarted.
Call virDomainObjSetDefTransient at VM startup, so that we properly discard
hotplug changes when the VM is shutdown.
I am replacing the last instances of close() I found with VIR_CLOSE() / VIR_FORCE_CLOSE respectively.
The first part patches virsh, which I missed out on previously.
The 2nd patch I had left out intentionally to look at it more carefully:
The 'closed' variable could be easily removed since it wasn't used anywhere else. The possible race condition that could result from the filedescriptor being closed and not set to -1 (and possibly let us write into 'something' totally different if the fd was allocated by another thread) seems to be prevented by the qemuMonitorLock() already placed around the code that reads from or writes to the fd. So the change of this code as shown in the patch should not have any side-effects.
Similarly to deprecating close(), I am now deprecating fclose() and
introduce VIR_FORCE_FCLOSE() and VIR_FCLOSE(). Also, fdopen() is replaced with
VIR_FDOPEN().
Most of the files are opened in read-only mode, so usage of
VIR_FORCE_CLOSE() seemed appropriate. Others that are opened in write
mode already had the fclose()< 0 check and I converted those to
VIR_FCLOSE()< 0.
I did not find occurrences of possible double-closed files on the way.
Currently only support domain start and shutdown, for domain start,
record timestamp before the qemu command line, and for domain shutdown,
just say it's shutting down with timestamp.
* src/qemu/qemu_driver.c (qemudStartVMDaemon, qemudShutdownVMDaemon
introduced two macros - START_POSTFIX, SHUTDOWN_POSTFIX)
This provides an implementation of the virDomainOpenConsole
API with the QEMU driver. For the streams code, this reuses
most of the code previously added for the tunnelled migration
streams since it is generic.
* src/qemu/qemu_driver.c: Support virDomainOpenConsole
To avoid the need for duplicating implementations of virStream
drivers, provide a generic implementation that can handle any
FD based stream. This code is copied from the existing impl
in the QEMU driver, with the locking moved into the stream
impl, and addition of a read callback
The FD stream code will refuse to operate on regular files or
block devices, since those can't report EAGAIN properly when
they would block on I/O
* include/libvirt/virterror.h, include/libvirt/virterror.h: Add
VIR_FROM_STREAM error domain
* src/qemu/qemu_driver.c: Remove code obsoleted by the new
generic streams driver.
* src/fdstream.h, src/fdstream.c, src/fdstream.c,
src/libvirt_private.syms: Generic reusable FD based streams
To enable virsh console (or equivalent) to be used remotely
it is necessary to provide remote access to the /dev/pts/XXX
pseudo-TTY associated with the console/serial/parallel device
in the guest. The virStream API provide a bi-directional I/O
stream capability that can be used for this purpose. This
patch thus introduces a virDomainOpenConsole API that uses
the stream APIs.
* src/libvirt.c, src/libvirt_public.syms,
include/libvirt/libvirt.h.in, src/driver.h: Define the
new virDomainOpenConsole API
* src/esx/esx_driver.c, src/lxc/lxc_driver.c,
src/opennebula/one_driver.c, src/openvz/openvz_driver.c,
src/phyp/phyp_driver.c, src/qemu/qemu_driver.c,
src/remote/remote_driver.c, src/test/test_driver.c,
src/uml/uml_driver.c, src/vbox/vbox_tmpl.c,
src/xen/xen_driver.c, src/xenapi/xenapi_driver.c: Stub
API entry point
QEMU allows forcing a CDROM eject even if the guest has locked the device.
Expose this via a new UpdateDevice flag, VIR_DOMAIN_DEVICE_MODIFY_FORCE.
This has been requested for RHEV:
https://bugzilla.redhat.com/show_bug.cgi?id=626305
v2: Change flag name, bool cleanups
Using automated replacement with sed and editing I have now replaced all
occurrences of close() with VIR_(FORCE_)CLOSE() except for one, of
course. Some replacements were straight forward, others I needed to pay
attention. I hope I payed attention in all the right places... Please
have a look. This should have at least solved one more double-close
error.
This extends the SPICE XML to allow channel security options
<graphics type='spice' port='-1' tlsPort='-1' autoport='yes'>
<channel name='main' mode='secure'/>
<channel name='record' mode='insecure'/>
</graphics>
Any non-specified channel uses the default, which allows both
secure & insecure usage
* src/conf/domain_conf.c, src/conf/domain_conf.h,
src/libvirt_private.syms: Add XML syntax for specifying per
channel security options for spice.
* src/qemu/qemu_conf.c: Configure channel security with spice
QEMU crashes & burns if you try multiple Cirrus video cards, but
QXL copes fine. Adapt QEMU config code to allow multiple QXL
video cards
* src/qemu/qemu_conf.c: Support multiple QXL video cards
This extends the XML syntax for <graphics> to allow a password
expiry time to be set
eg
<graphics type='vnc' port='5900' autoport='yes' keymap='en-us' passwd='12345' passwdValidTo='2010-04-09T15:51:00'/>
The timestamp is in UTC.
* src/conf/domain_conf.h: Pull passwd out into separate struct
virDomainGraphicsAuthDef to allow sharing between VNC & SPICE
* src/conf/domain_conf.c: Add parsing/formatting of new passwdValidTo
argument
* src/opennebula/one_conf.c, src/qemu/qemu_conf.c, src/qemu/qemu_driver.c,
src/xen/xend_internal.c, src/xen/xm_internal.c: Update for changed
struct containing VNC password
In common with VNC, the QEMU driver configuration file is used
specify the host level TLS certificate location and a default
password / listen address
* src/qemu/qemu.conf: Add spice_listen, spice_tls,
spice_tls_x509_cert_dir & spice_password config params
* src/qemu/qemu_conf.c, src/qemu/qemu_conf.h: Parsing of
spice config parameters and updating -spice arg generation
to use them
* tests/qemuxml2argvdata/qemuxml2argv-graphics-spice-rhel6.args,
tests/qemuxml2argvtest.c: Expand test case to cover driver
level configuration
This supports the -spice argument posted for review against
the latest upstream QEMU/KVM. This supports the bare minimum
config with port, TLS port & listen address. The x509 bits are
added in a later patch.
* src/qemu_conf.c, src/qemu_conf.h: Add SPICE flag. Check for
-spice availability. Format -spice arg for command line
* qemuhelptest.c: Add SPICE flag
* qemuxml2argvdata/qemuxml2argv-graphics-spice.args: Add <graphics>
for spice
* qemuxml2argvdata/qemuxml2argv-graphics-spice.xml: Add -spice arg
* qemuxml2argvtest.c: Add SPICE flag
This supports the '-vga qxl' parameter in upstream QEMU/KVM
which has SPICE support added. This isn't particularly useful
until you get the next patch for -spice support. Also note that
while the libvirt XML supports multiple video devices, this
patch only supports a single one. A later patch can add support
for 2nd, 3rd, etc PCI devices for QXL
* src/qemu/qemu_conf.h: Flag for QXL support
* src/qemu/qemu_conf.c: Probe for '-vga qxl' support and implement it
* tests/qemuxml2argvtest.c, tests/qemuxml2xmltest.c,
tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.args,
tests/qemuxml2argvdata/qemuxml2argv-graphics-spice.xml: Test
case for generating spice args with RHEL6 kvm
This adds an element
<graphics type='spice' port='5903' tlsPort='5904' autoport='yes' listen='127.0.0.1'/>
This is the bare minimum that should be exposed in the guest
config for SPICE. Other parameters are better handled as per
host level configuration tunables
* docs/schemas/domain.rng: Define the SPICE <graphics> schema
* src/domain_conf.h, src/domain_conf.c: Add parsing and formatting
for SPICE graphics config
* src/qemu_conf.c: Complain about unsupported graphics types
* src/qemu_conf.c: Add dummy entry in enumeration
* docs/schemas/domain.rng: Add 'qxl' as a type for the <video> tag
* src/domain_conf.c, src/domain_conf.h: Add QXL to video type
enumerations
The patch is based on the possiblity in the QEmu command line to
add -smbios options allowing to override the default values picked
by QEmu. We need to detect this first from QEmu help output.
If the domain is defined with smbios to be inherited from host
then we pass the values coming from the Host own SMBIOS, but
if the domain is defined with smbios to come from sysinfo, we
use the ones coming from the domain definition.
* src/qemu/qemu_conf.h: add the QEMUD_CMD_FLAG_SMBIOS_TYPE enum
value
* src/qemu/qemu_conf.c: scan the help output for the smbios support,
and if available add support based on the domain definitions,
and host data
* tests/qemuhelptest.c: add the new enum in the outputs
A couple of places in the text monitor were overwriting the
'ret' variable with a >= 0 value before success was actually
determined. So later error paths would not correctly return
the -1 value. The drive_add code was not checking for errors
like missing command
* src/qemu/qemu_monitor_text.c: Misc error handling fixes
Commit 06f81c63eb attempted to make
QEMU driver ignore the failure to relabel 'stdin_path' if it was
on NFS. The actual result was that it ignores *all* failures to
label any aspect of the VM, unless stdin_path is non-NULL and
is not on NFS.
* src/qemu/qemu_driver.c: Treat all relabel failures as terminal
Add dump_image_format[] to qemu.conf and support compressed dump
at virsh dump. coredump compression is important for saving disk space
in an environment where multiple guests run.
In general, "disk space for dump" is specially allocated and will be
a dead space in the system. It's used only at emergency. So, it's better
to have both of save_image_format and dump_image_format. "save" is done
in scheduled manner with enough calculated disk space for it.
This code reuses some of save_image_format[] and supports the same format.
Changelog:
- modified libvirtd_qemu.aug
- modified test_libvirtd_qemu.aug
- fixed error handling of qemudSaveCompressionTypeFromString()
When we mount any cgroup without "-o devices", we will fail to start vms:
error: Failed to start domain vm1
error: Unable to deny all devices for vm1: No such file or directory
When we mount any cgroup without "-o cpu", we will fail to get schedinfo:
Scheduler : posix
error: unable to get cpu shares tunable: No such file or directory
We should only use the cgroup controllers which are mounted on host.
So I add virCgroupMounted() for qemuCgroupControllerActive()
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
https://bugzilla.redhat.com/show_bug.cgi?id=638285 - when migrating
a guest, it was very easy to provoke a race where an application
could query block information on a VM that had just been migrated
away. Any time qemu code obtains a job lock, it must also check
that the VM was not taken down in the time where it was waiting
for the lock.
* src/qemu/qemu_driver.c (qemudDomainSetMemory)
(qemudDomainGetInfo, qemuDomainGetBlockInfo): Check that vm still
exists after obtaining job lock, before starting monitor action.
Add auditing of all initial disk/net assignments to QEMU guests
at startup. Add auditing for all hotplug & unplug events and
disk media changes.
* src/qemu/qemu_driver.c: Add disk/net resource auditing
Revert most of commit a8b5f9bd27.
The audit hooks will be re-added directly in the QEMU driver code
in a future commit
* daemon/remote.c: Remove all audit logging hooks
* src/qemu/qemu_driver.c: Remove all audit logging hooks
When using 0-prefixed numbers, QEmu will interpret them as octal numbers
(as C convention says); this means that if you attach a device that has
addr > 10 (decimal) you're going to attach a different device.
Older dash mistakenly truncates regular files when using <> redirection;
this kills our use of double dd to reduce storage overhead when
saving qemu images. But qemu insists on running a command through
/bin/sh, so we work around it by having qemu run $sh -c 'real command'
when we have a replacement $sh in mind.
* configure.ac (VIR_WRAPPER_SHELL): Define to a replacement shell,
if /bin/sh is broken on <> redirection.
* src/qemu/qemu_monitor.h (VIR_WRAPPER_SHELL_PREFIX)
(VIR_WRAPPER_SHELL_SUFFIX): New macros.
* src/qemu/qemu_monitor_text.c (qemuMonitorTextMigrateToFile): Use
them.
* src/qemu/qemu_monitor_json.c (qemuMonitorJSONMigrateToFile):
Likewise.
So far, readonly=on option is used when qemu supports -device. However,
there are qemu versions which support readonly option with -drive
although they don't have support for -device.
The getnameinfo() function is more flexible than inet_ntop()
avoiding the need to if/else the code based on socket family.
Also make it support UNIX socket addrs and allow inclusion
of a port (service) address. Finally do proper error reporting
via normal APIs.
* src/conf/domain_conf.c, src/nwfilter/nwfilter_ebiptables_driver.c,
src/qemu/qemu_conf.c: Fix error handling with virSocketFormat
* src/util/network.c: Rewrite virSocketFormat to use getnameinfo
and cope with UNIX socket addrs.
The QEMU 0.13 release is finally out and from testing in RHEL-6
we know that its JSON and netdev features are now good enough
for us to use by default.
* src/qemu/qemu_conf.c: Enable JSON + netdev for QEMU >= 0.13
There is no point in trying to fill params beyond the first error,
because when qemuDomainGetMemoryParameters returns -1 then the caller
cannot detect which values in params are valid.
This sets the process name to the same value as the Windows title,
but since the name is limited to 16 chars only this is kept as a
configuration option and turned off by default
* src/qemu/qemu.conf src/qemu/qemu_conf.[ch]: hceck for support in the
QEmu help output, add the option in qemu conf file and augment
qemudBuildCommandLine to add it if switched on
* src/qemu/libvirtd_qemu.aug src/qemu/test_libvirtd_qemu.aug: augment
the augeas lenses accordingly
* tests/qemuhelptest.c: cope with the extra flag being detected now
Most operations are audited at the libvirtd level; auditing in
src/libvirt.c would result in two audit entries per operation (one in
the client, one in libvirtd).
The only exception is a domain stopping of its own will (e.g. because
the user clicks on "shutdown" inside the interface). There can often be
no client connected at the time the domain stops, so libvirtd does not
have any virConnectPtr object on which to attach an event watch. This
patch therefore adds auditing directly inside the qemu driver (other
drivers are not supported).
* src/qemu/qemu_conf.c (qemuParseCommandLineSmp): Distinguish
between vcpus and maxvcpus, for new enough qemu.
* tests/qemuargv2xmltest.c (mymain): Add new test.
* tests/qemuxml2argvtest.c (mymain): Likewise.
* tests/qemuxml2xmltest.c (mymain): Likewise.
* tests/qemuxml2argvdata/qemuxml2argv-smp.args: New file.
Although this patch adds a distinction between maximum vcpus and
current vcpus in the XML, the values should be identical for all
drivers at this point. Only in subsequent per-driver patches will
a distinction be made.
In general, virDomainGetInfo should prefer the current vcpus.
* src/conf/domain_conf.h (_virDomainDef): Adjust vcpus to unsigned
short, to match virDomainGetInfo limit. Add maxvcpus member.
* src/conf/domain_conf.c (virDomainDefParseXML)
(virDomainDefFormat): parse and print out vcpu details.
* src/xen/xend_internal.c (xenDaemonParseSxpr)
(xenDaemonFormatSxpr): Manage both vcpu numbers, and require them
to be equal for now.
* src/xen/xm_internal.c (xenXMDomainConfigParse)
(xenXMDomainConfigFormat): Likewise.
* src/phyp/phyp_driver.c (phypDomainDumpXML): Likewise.
* src/openvz/openvz_conf.c (openvzLoadDomains): Likewise.
* src/openvz/openvz_driver.c (openvzDomainDefineXML)
(openvzDomainCreateXML, openvzDomainSetVcpusInternal): Likewise.
* src/vbox/vbox_tmpl.c (vboxDomainDumpXML, vboxDomainDefineXML):
Likewise.
* src/xenapi/xenapi_driver.c (xenapiDomainDumpXML): Likewise.
* src/xenapi/xenapi_utils.c (createVMRecordFromXml): Likewise.
* src/esx/esx_vmx.c (esxVMX_ParseConfig, esxVMX_FormatConfig):
Likewise.
* src/qemu/qemu_conf.c (qemuBuildSmpArgStr)
(qemuParseCommandLineSmp, qemuParseCommandLine): Likewise.
* src/qemu/qemu_driver.c (qemudDomainHotplugVcpus): Likewise.
* src/opennebula/one_conf.c (xmlOneTemplate): Likewise.
Note - this wrapping is completely mechanical; the old API will
function identically, since the new API validates that the exact
same flags are provided by the old API. On a per-driver basis,
it may make sense to have the old API pass a different set of flags,
but that should be done in the per-driver patch that implements
the full range of flag support in the new API.
* src/esx/esx_driver.c (esxDomainSetVcpus, escDomainGetMaxVpcus):
Move guts...
(esxDomainSetVcpusFlags, esxDomainGetVcpusFlags): ...to new
functions.
(esxDriver): Trivially support the new API.
* src/openvz/openvz_driver.c (openvzDomainSetVcpus)
(openvzDomainSetVcpusFlags, openvzDomainGetMaxVcpus)
(openvzDomainGetVcpusFlags, openvzDriver): Likewise.
* src/phyp/phyp_driver.c (phypDomainSetCPU)
(phypDomainSetVcpusFlags, phypGetLparCPUMAX)
(phypDomainGetVcpusFlags, phypDriver): Likewise.
* src/qemu/qemu_driver.c (qemudDomainSetVcpus)
(qemudDomainSetVcpusFlags, qemudDomainGetMaxVcpus)
(qemudDomainGetVcpusFlags, qemuDriver): Likewise.
* src/test/test_driver.c (testSetVcpus, testDomainSetVcpusFlags)
(testDomainGetMaxVcpus, testDomainGetVcpusFlags, testDriver):
Likewise.
* src/vbox/vbox_tmpl.c (vboxDomainSetVcpus)
(vboxDomainSetVcpusFlags, virDomainGetMaxVcpus)
(virDomainGetVcpusFlags, virDriver): Likewise.
* src/xen/xen_driver.c (xenUnifiedDomainSetVcpus)
(xenUnifiedDomainSetVcpusFlags, xenUnifiedDomainGetMaxVcpus)
(xenUnifiedDomainGetVcpusFlags, xenUnifiedDriver): Likewise.
* src/xenapi/xenapi_driver.c (xenapiDomainSetVcpus)
(xenapiDomainSetVcpusFlags, xenapiDomainGetMaxVcpus)
(xenapiDomainGetVcpusFlags, xenapiDriver): Likewise.
(xenapiError): New helper macro.
Over root-squashing nfs, when virFileOperation() is called as uid==0,
it may fail with EACCES, but also with EPERM, due to
virFileOperationNoFork()'s failed attemp to chown a writable file.
qemudDomainSaveFlag() should expect this case, too.
qemudOpenAsUID is intended to open a file with the credentials of a
specified uid. Current implementation fails if the file is accessible to
one of uid's groups but not owned by uid.
This patch replaces the supplementary group list that the child process
inherited from libvirtd with the default group list of uid.
This introduces new attribute to filesystem element
to support customizable access mode for mount type.
Valid accessmode are: passthrough, mapped and squash.
Usage:
<filesystem type='mount' accessmode='passthrough'>
<source dir='/export/to/guest'/>
<target dir='mount_tag'/>
</filesystem>
passthrough is the default model if not specified, that's
also the current behaviour.
Explicitly raising a nice error in the case user tries to migrate a
guest with assigned host devices is much better than waiting for a
mysterious error with no clue for the reason.
This enables support for nested SVM using the regular CPU
model/features block. If the CPU model or features include
'svm', then the '-enable-nesting' flag will be added to the
QEMU command line. Latest out of tree patches for nested
'vmx', no longer require the '-enable-nesting' flag. They
instead just look at the cpu features. Several of the models
already include svm support, but QEMU was just masking out
the svm bit silently. So this will enable SVM on such
models
* src/qemu/qemu_conf.h: flag for -enable-nesting
* src/qemu/qemu_conf.c: Use -enable-nesting if VMX or SVM are in
the CPUID
* src/cpu/cpu.h, src/cpu/cpu.c: API to check for a named feature
* src/cpu/cpu_x86.c: x86 impl of feature check
* src/libvirt_private.syms: Add cpuHasFeature
* src/qemuhelptest.c: Add nesting flag where required
This is from a bug report and conversation on IRC where Soren reported that while a filter update is occurring on one or more VMs (due to a rule having been edited for example), a deadlock can occur when a VM referencing a filter is started.
The problem is caused by the two locking sequences of
qemu driver, qemu domain, filter # for the VM start operation
filter, qemu_driver, qemu_domain # for the filter update operation
that obviously don't lock in the same order. The problem is the 2nd lock sequence. Here the qemu_driver lock is being grabbed in qemu_driver:qemudVMFilterRebuild()
The following solution is based on the idea of trying to re-arrange the 2nd sequence of locks as follows:
qemu_driver, filter, qemu_driver, qemu_domain
and making the qemu driver recursively lockable so that a second lock can occur, this would then lead to the following net-locking sequence
qemu_driver, filter, qemu_domain
where the 2nd qemu_driver lock has been ( logically ) eliminated.
The 2nd part of the idea is that the sequence of locks (filter, qemu_domain) and (qemu_domain, filter) becomes interchangeable if all code paths where filter AND qemu_domain are locked have a preceding qemu_domain lock that basically blocks their concurrent execution
So, the following code paths exist towards qemu_driver:qemudVMFilterRebuild where we now want to put a qemu_driver lock in front of the filter lock.
-> nwfilterUndefine() [ locks the filter ]
-> virNWFilterTestUnassignDef()
-> virNWFilterTriggerVMFilterRebuild()
-> qemudVMFilterRebuild()
-> nwfilterDefine()
-> virNWFilterPoolAssignDef() [ locks the filter ]
-> virNWFilterTriggerVMFilterRebuild()
-> qemudVMFilterRebuild()
-> nwfilterDriverReload()
-> virNWFilterPoolLoadAllConfigs()
->virNWFilterPoolObjLoad()
-> virNWFilterPoolAssignDef() [ locks the filter ]
-> virNWFilterTriggerVMFilterRebuild()
-> qemudVMFilterRebuild()
-> nwfilterDriverStartup()
-> virNWFilterPoolLoadAllConfigs()
->virNWFilterPoolObjLoad()
-> virNWFilterPoolAssignDef() [ locks the filter ]
-> virNWFilterTriggerVMFilterRebuild()
-> qemudVMFilterRebuild()
Qemu is not the only driver using the nwfilter driver, but also the UML driver calls into it. Therefore qemuVMFilterRebuild() can be exchanged with umlVMFilterRebuild() along with the driver lock of qemu_driver that can now be a uml_driver. Further, since UML and Qemu domains can be running on the same machine, the triggering of a rebuild of the filter can touch both types of drivers and their domains.
In the patch below I am now extending each nwfilter callback driver with functions for locking and unlocking the (VM) driver (UML, QEMU) and introduce new functions for locking all registered callback drivers and unlocking them. Then I am distributing the lock-all-cbdrivers/unlock-all-cbdrivers call into the above call paths. The last shown callpath starting with nwfilterDriverStart() is problematic since it is initialize before the Qemu and UML drives are and thus a lock in the path would result in a NULL pointer attempted to be locked -- the call to virNWFilterTriggerVMFilterRebuild() is never called, so we never lock either the qemu_driver or the uml_driver in that path. Therefore, only the first 3 paths now receive calls to lock and unlock all callback drivers. Now that the locks are distributed where it matters I can remove the qemu_driver and uml_driver lock from qemudVMFilterRebuild() and umlVMFilterRebuild() and not requiring the recursive locks.
For now I want to put this out as an RFC patch. I have tested it by 'stretching' the critical section after the define/undefine functions each lock the filter so I can (easily) concurrently execute another VM operation (suspend,start). That code is in this patch and if you want you can de-activate it. It seems to work ok and operations are being blocked while the update is being done.
I still also want to verify the other assumption above that locking filter and qemu_domain always has a preceding qemu_driver lock.
Make use of the existing <filesystem> element to support plan9fs
filesystem passthrough in the QEMU driver
<filesystem type='mount'>
<source dir='/export/to/guest'/>
<target dir='/import/from/host'/>
</filesystem>
NB, the target is not actually a directory, it is merely a arbitrary
string tag that is exported to the guest as a hint for where to mount
it.
Adding parsing code for memory tunables in the domain xml file
also change the internal define structures used for domain memory
informations
Adds a new specific test
Public api to set/get memory tunables supported by the hypervisors.
dv:
* some cleanups in libvirt.c
* adding extra checks in libvirt.c new entry points
v4:
* Move exporting public API to this patch
* Add unsigned int flags to the public api for future extensions
v3:
* Add domainGetMemoryParamters and NULL in all the driver interface
v2:
* Initialize domainSetMemoryParameters to NULL in all the driver
interface structure.
A QEMU guest can have upto VIR_DOMAIN_BOOT_LAST boot entries
defined. When building the QEMU arg, each entry takes a
single byte. This means the array must be declared to be
VIR_DOMAIN_BOOT_LAST+1 bytes in length to allow for the
trailing null
* src/qemu/qemu_conf.c: Fix off-by-1 boot arg array size
QMP in QEMU 0.13 has been fixed to enforce type correctness,
this means that boolean types must be true or false, not
integers.
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
QMP in QEMU 0.13 has been fixed to enforce type correctness,
this means that boolean types must be true or false, not
integers.
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Other drivers will need this same functionality, so move it to up to
conf/domain_conf.c and give it a more general name.
Signed-off-by: Soren Hansen <soren@linux2go.dk>
Previously QEMU enabled KQEMU by default and had -no-kqemu.
0.11.x switched to requiring -enable-kqemu. 0.12.x dropped
kqemu entirely. This patch adds support for -enable-kqemu
so 0.11.x works. It replaces a huge set of if() with a
switch() to make the code a bit more readable.
* src/qemu/qemu_conf.c, src/qemu/qemu_conf.h: Support
-enable-kqemu
We already filled the PCI address structure when we checked whether it's
free or not, so let's just use the structure here instead of filling it
again.
The current version of the qemu managed save implementation
is subject to a race where the domain shuts down between
the time that we start the command and the time that we
actually try to do the save. Close this race by making
qemuDomainSaveFlags() expect both the driver and the passed-in
vm object to be locked before executing.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
When reconnecting to existing VMs, we re-reserved only those PCI
addresses which were explicitly mentioned in domain XML. Since some
addresses are always reserved (e.g., 0:0:0 and 0:0:1), we need to handle
those too.
Also all this should only be done if device flag is supported by qemu.
In this patch I am extending and fixing the nwfilter module's reload support to stop all ongoing threads (for learning IP addresses of interfaces) and rebuild the filtering rules of all interfaces of all VMs when libvirt is started. Now libvirtd rebuilds the filters upon the SIGHUP signal and libvirtd restart.
About the patch: The nwfilter functions require a virConnectPtr. Therefore I am opening a connection in qemudStartup, which later on needs to be closed outside where the driver lock is held since otherwise it ends up in a deadlock due to virConnectClose() trying to lock the driver as well.
I have tested this now for a while with several machines running and needing the IP address learner thread(s). The rebuilding of the firewall rules seems to work fine following libvirtd restart or a SIGHUP. Also the termination of libvirtd worked fine.
Since the qemu process is running as qemu:qemu, it can't actually
look at the unix socket in /var/run/libvirt/qemu which is owned by
root and has permission 700. Move the unix socket to
/var/lib/libvirt/qemu, which is already owned by qemu:qemu.
Thanks to Justin Clift for test this out for me.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
The problem is that on the source of the migration, libvirtd
is responsible for creating the unix socket over which the data
will flow. Since libvirtd is running as root, this file will
be created as root. When the qemu process running as qemu:qemu
goes to access the unix file to write data to it, it will get
permission denied and fail. Make sure to change the owner
of the unix file to qemu:qemu.
Thanks to Justin Clift for testing this patch out for me.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Basically a followup of the previous patch about balloon desactivation
if desactivated, to not ask for balloon information to qemu as we will
just get an error back.
This can make a huge difference in the time needed for domain
information or list when a machine is loaded, and balloon has been
desactivated in the guests.
* src/qemu/qemu_driver.c: do not get the balloon info if the balloon
suppor is disabled
The balloon device is automatically added to qemu guests if supported,
but it may be useful to desactivate it. The simplest to not change the
existing behaviour is to allow
<memballoon type="none"/>
as an extra option to desactivate it (it is automatically added if the
memballoon construct is missing for the domain).
The following simple patch just adds the extra option and does not
change the default behaviour but avoid creating a balloon device if
type="none" is used.
* docs/schemas/domain.rng: add the extra type attribute value
* src/conf/domain_conf.c src/conf/domain_conf.h: add the extra enum
value
* src/qemu/qemu_conf.c: if enum is NONE, don't activate the device,
i.e. don't pass the args to qemu/kvm
device_del command is not synchronous for PCI devices, it merely asks
the guest to release the device and returns. If the host wants to use
that device before the guest actually releases it, we are in big
trouble. To avoid this, we already added a loop which waits up to 10
seconds until the device is actually released before we do anything else
with that device. But we only added this loop for managed PCI devices
before we try reattach them back to the host.
However, we need to wait even for non-managed devices. We don't reattach
them automatically, but we still want to prevent the host from using it.
This was revealed thanks to sVirt: when we relabel sysfs files
corresponding to the PCI device before the guest finished releasing the
device, qemu is no longer allowed to access those files and if it wants
(as a result of guest's request) to write anything to them, it just
exits, which kills the guest.
This is not a proper fix and needs some further work both on libvirt and
qemu side in the future.
Fix the error checking to use the return value from brAddTap() instead
of checking the current errno value which might have been changed by
clean up calls inside of brAddTap().
Signed-off-by: Doug Goldstein <cardoe@gentoo.org>
Added a more detailed error message when adding a tap devices fails and
the kernel is missing tun support.
Signed-off-by: Doug Goldstein <cardoe@gentoo.org>
the followup on the boot=on problem, basically it's not needed to
specify it when booting out of IDE devices when using KVM
* src/qemu/qemu_conf.c: do not use boot=on for IDE devices
* tests/qemuxml2argvdata/qemuxml2argv*.args: this changes the output
for 5 of the tests
Patch version revamped by Eric Blake <eblake@redhat.com> of Jiri
Denemark <jdenemar@redhat.com> original patch
When attaching a PCI device which doesn't explicitly set its PCI
address, libvirt allocates the address automatically. The problem is
that when checking which PCI address is unused, we only check for those
with slot number higher than the highest slot number ever used.
Thus attaching/detaching such device several times in a row (31 is the
theoretical limit, less then 30 tries are enough in practise) makes any
further device attachment fail. Furthermore, attaching a device with
predefined PCI address to 0:0:31 immediately forbids attachment of any
PCI device without explicit address.
This patch changes the logic so that we always check all PCI addresses
before we say there is no PCI address available.
Modifications from v1: revert back to remembering the last slot
reserved, but allow wraparound to not be limited by the end.
In this way, slots are still assigned in the same order as
before the patch, rather than filling in the gaps closest to
0 and risking making windows guests mad.
* src/qemu/qemu_conf.c: fix pci reservation code to do a round-robbin
check of all available PCI splot availability before failing.
Basically the 'boot=on' boot selection device is something present in
KVM but not in upstream QEmu, as a result if we boot a QEmu domain
without KVM acceleration we must disable boot=on ... even if the front
end kvm binary expose that capability in the help page.
* src/qemu/qemu_conf.c: in qemudBuildCommandLine if -no-kvm
is passed, then deactivate QEMUD_CMD_FLAG_DRIVE_BOOT
ADD_ARG_LIT should only be used for literal arguments,
since it duplicates the memory. Since virBufferContentAndReset
is already allocating memory, we should only use ADD_ARG.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
If detecting the FLR flag of a pci device fails, then we
could run into the situation of trying to close a file
descriptor twice, once in pciInitDevice() and once in pciFreeDevice().
Fix that by removing the pciCloseConfig() in pciInitDevice() and
just letting pciFreeDevice() handle it.
Thanks to Chris Wright for pointing out this problem.
While we are at it, fix an error check. While it would actually
work as-is (since success returns 0), it's still more clear to
check for < 0 (as the rest of the code does).
Signed-off-by: Chris Lalancette <clalance@redhat.com>
All <console> devices now export a <target> type attribute. QEMU defaults
to 'serial', UML defaults to 'uml, xen can be either 'serial' or 'xen'
depending on fullvirt. Understandably there is lots of test fallout.
This will be used to differentiate between a serial vs. virtio console for
QEMU.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
targetType only tracks the actual <target> format we are parsing. Currently
we only fill abide this value for channel devices.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
There is actually a difference between the character device type (serial,
parallel, channel, ...) and the target type (virtio, guestfwd). Currently
they are awkwardly conflated.
Start to pull them apart by renaming targetType -> deviceType. This is
an entirely mechanical change.
Signed-off-by: Cole Robinson <crobinso@redhat.com>
QEMU has had two different syntax for disk cache options
Old: on|off
New: writeback|writethrough|none
QEMU recently added another 'unsafe' option which broke the
libvirt check. We can avoid this & future breakage, if we
do a negative check for the old syntax, instead of a positive
check for the new syntax
* src/qemu/qemu_conf.c: Invert cache option check
Add a new element to the <os> block:
<bootmenu enable="yes|no"/>
Which maps to -boot,menu=on|off on the QEMU command line.
I decided to use an explicit 'enable' attribute rather than just make the
bootmenu element boolean. This allows us to treat lack of a bootmenu element
as 'use hypervisor default'.
When doing a PCI secondary bus reset, we must be sure that there are no
active devices on the same bus segment. The active device tracking is
designed to only track host devices that are active in use by guests.
This ignores host devices that are actively in use by the host. So the
current logic will reset host devices.
Switch this logic around and allow sbus reset when we are assigning all
devices behind a bridge to the same guest at guest startup or as a result
of a single attach-device command.
* src/util/pci.h: change signature of pciResetDevice to add an
inactive devices list
* src/qemu/qemu_driver.c src/xen/xen_driver.c: use (or not) the new
functionality of pciResetDevice() depending on the place of use
* src/util/pci.c: implement the interface and logic changes
- src/qemu/qemu_driver.c: Eliminate code duplication by using the new
helpers qemuPrepareHostdevPCIDevices and qemuDomainReAttachHostdevDevices.
This reduces the number of open coded calls to pciResetDevice.
- src/qemu/qemu_driver.c: These new helpers take hostdev list and count
directly rather than getting them indirectly from domain definition.
This will allow reuse for the attach-device case.
- src/qemu/qemu_driver.c: Update qemuGetPciHostDeviceList to take a
hostdev list and count directly, rather than getting this indirectly
from domain definition. This will allow reuse for the attach-device case.
Thanks to DV for knocking together the Relax-NG changes
quickly for me.
Changes since v1:
- Change the domain.rng to correspond to the new schema
- Don't allocate caps->ns in testQemuCapsInit since it is a static table
Changes since v2:
- Change domain.rng to add restrictions on allowed environment names
Changes since v3:
- Remove a bogus comment in the tests
Signed-off-by: Chris Lalancette <clalance@redhat.com>
