Below is backtraces of two deadlocked threads:
thread #1:
virDomainConfVMNWFilterTeardown
virNWFilterTeardownFilter
lock updateMutex <------------
_virNWFilterTeardownFilter
try to lock interface <----------
thread #2:
learnIPAddressThread
lock interface <-------
virNWFilterInstantiateFilterLate
try to lock updateMutex <----------
The problem is fixed by unlocking interface before calling
virNWFilterInstantiateFilterLate to avoid updateMutex and interface ordering
deadlocks. Otherwise we are going to instantiate the filter while holding
interface lock, which will try to lock updateMutex, and if some other thread
instantiating a filter in parallel is holding updateMutex and is trying to
lock interface, both will deadlock.
Also it is safe to unlock interface before virNWFilterInstantiateFilterLate
because learnIPAddressThread stopped capturing packets and applied necessary
rules on the interface, while instantiating a new filter doesn't require a
locked interface.
Signed-off-by: Maxim Nestratov <mnestratov@virtuozzo.com>
We need to append GNUTLS_CFLAGS while building utils because virtcrypto
is using it. This fixes build on freebsd where gnutuls is in
/usr/local/include.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
If try to stop VM or container which is already stopped than
Virtuozzo 7 returns code PRL_ERR_INVALID_ACTION_REQUESTED.
Error code PRL_ERR_DISP_VM_IS_NOT_STARTED is used in Virtuozzo 6
Commit 5e54361c added virStoragePoolObjClearVols before refreshPool
to prevent duplicate volume entries.
However it is not needed here because we're not refreshing the pool yet,
just checking for the existence of the refresh callback.
The actual refresh is done via virStorageVolFDStreamCloseCb
in virStorageVolPoolRefreshThread, which already calls
virStoragePoolObjClearVols.
Rather than only assigning a PCI address when no address is given at
all, also do it when the config says that the address type is 'pci',
but it gives no address (virDeviceInfoPCIAddressWanted()).
There are also several places after parsing but prior to address
assignment where code previously expected that any info with address
type='pci' would have a *valid* PCI address, which isn't always the
case - now we check not only for type='pci', but also for a valid
address (virDeviceInfoPCIAddressPresent()).
The test case added in this patch was directly copied from Cole's patch titled:
qemu: Wire up address type=pci auto_allocate
Rather than only assigning a PCI address when no address is given at
all, also do it when the config says that the address type is 'pci',
but it gives no address.
Prior to this, <address type='pci'/> wasn't allowed when parsing
(domain+bus+slot+function needed to be a "valid" PCI address, meaning
that at least one of domain/bus/slot had to be non-0), the RNG
required bus to be specified, and if type was set to PCI when
formatting, domain+bus+slot+function would always be output.
This makes all the address attributes optional during parse and RNG
validation, and suppresses domain+bus+slot+function if domain+bus+slot
are all 0 (NB: if d+b+s are all 0, any value for function is
nonsensical as that will never happen in the real world, and after
the next patch we will always assign a real working address to any
empty PCI address before it is ever output to anywhere).
Note that explicitly setting all attributes to 0 is equivalent to
setting none of them, which is okay, since 0000:00:00 is reserved in
any PCI bus setup, and can't be used anyway.
In order to allow <address type='pci'/> with no other attributes to
mean "I want a PCI address, but any PCI address will do" (just as
having no <address> at all usually indicates), we will need to change
several places in the code from a simple "info->type == (or !=)
VIR_DOMAIN_DEVICE_ADDRESS_TYPE_(PCI|NONE)" into something slightly
more complex, this patch adds to new functions that take a
virDomainDeviceInfoPtr and return true/false depending on 1) whether
the current state of the info indicates that we "want" a PCI address
for this device (virDeviceInfoPCIAddressWanted()) and 2) whether this
device already has a valid PCI address
(virDeviceInfoPCIAddressPresent()).
Both of these functions required the simpler check for whether a pci
address is "empty" (i.e. all of its attributes are 0, which can never
happen in a real PCI address, since slot 0 of bus 0 of domain 0 is
always reserved), so that function is also added.
Also moves all the subordinate structs. This is necessary due to a new
inline function that will be defined in device_conf.h, and also makes
sense, because it is the *device* info that's in the struct. (Actually
a lot more stuff from domain_conf.h could move to this newer file, but
I didn't want to disturb any more than necessary).
https://bugzilla.redhat.com/show_bug.cgi?id=1182074
If they're available and we need to pass secrets to qemu, then use the
qemu domain secret object in order to pass the secrets for RBD volumes
instead of passing the base64 encoded secret on the command line.
The goal is to make AES secrets the default and have no user interaction
required in order to allow using the AES mechanism. If the mechanism
is not available, then fall back to the current plain mechanism using
a base64 encoded secret.
New APIs:
qemu_domain.c:
qemuDomainGetSecretAESAlias:
Generate/return the secret object alias for an AES Secret Info type.
This will be called from qemuDomainSecretAESSetup.
qemuDomainSecretAESSetup: (private)
This API handles the details of the generation of the AES secret
and saves the pieces that need to be passed to qemu in order for
the secret to be decrypted. The encrypted secret based upon the
domain master key, an initialization vector (16 byte random value),
and the stored secret. Finally, the requirement from qemu is the IV
and encrypted secret are to be base64 encoded.
qemu_command.c:
qemuBuildSecretInfoProps: (private)
Generate/return a JSON properties object for the AES secret to
be used by both the command building and eventually the hotplug
code in order to add the secret object. Code was designed so that
in the future perhaps hotplug could use it if it made sense.
qemuBuildObjectSecretCommandLine (private)
Generate and add to the command line the -object secret for the
secret. This will be required for the subsequent RBD reference
to the object.
qemuBuildDiskSecinfoCommandLine (private)
Handle adding the AES secret object.
Adjustments:
qemu_domain.c:
The qemuDomainSecretSetup was altered to call either the AES or Plain
Setup functions based upon whether AES secrets are possible (we have
the encryption API) or not, we have secrets, and of course if the
protocol source is RBD.
qemu_command.c:
Adjust the qemuBuildRBDSecinfoURI API's in order to generate the
specific command options for an AES secret, such as:
-object secret,id=$alias,keyid=$masterKey,data=$base64encodedencrypted,
format=base64
-drive file=rbd:pool/image:id=myname:auth_supported=cephx\;none:\
mon_host=mon1.example.org\:6321,password-secret=$alias,...
where the 'id=' value is the secret object alias generated by
concatenating the disk alias and "-aesKey0". The 'keyid= $masterKey'
is the master key shared with qemu, and the -drive syntax will
reference that alias as the 'password-secret'. For the -drive
syntax, the 'id=myname' is kept to define the username, while the
'key=$base64 encoded secret' is removed.
While according to the syntax described for qemu commit '60390a21'
or as seen in the email archive:
https://lists.gnu.org/archive/html/qemu-devel/2016-01/msg04083.html
it is possible to pass a plaintext password via a file, the qemu
commit 'ac1d8878' describes the more feature rich 'keyid=' option
based upon the shared masterKey.
Add tests for checking/comparing output.
NB: For hotplug, since the hotplug code doesn't add command line
arguments, passing the encoded secret directly to the monitor
will suffice.
Move the logic from qemuDomainGenerateRandomKey into this new
function, altering the comments, variable names, and error messages
to keep things more generic.
NB: Although perhaps more reasonable to add soemthing to virrandom.c.
The virrandom.c was included in the setuid_rpc_client, so I chose
placement in vircrypto.
Introduce virCryptoHaveCipher and virCryptoEncryptData to handle
performing encryption.
virCryptoHaveCipher:
Boolean function to determine whether the requested cipher algorithm
is available. It's expected this API will be called prior to
virCryptoEncryptdata. It will return true/false.
virCryptoEncryptData:
Based on the requested cipher type, call the specific encryption
API to encrypt the data.
Currently the only algorithm support is the AES 256 CBC encryption.
Adjust tests for the API's
According to QEMU docs, the '-m' option for specifying RAM is by default
in MiB, and a suffix of "M" or "G" may be passed for values in MiB and
GiB respectively. This commit adds support and a test for the same.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=812295
Signed-off-by: Nishith Shah <nishithshah.2211@gmail.com>
Both VNC and SPICE requires the same code to resolve address for listen
type network. Remove code duplication and create a new function that
will be used in qemuProcessSetupGraphics().
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Move code that decide whether we print the 'listen' attribute or not
into virDomainGraphicsListenDefFormatAddr() function.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
All callers of cpuGetModels expect @models to be NULL-terminated. Once
both x86GetModels and ppc64GetModels were fixed to meet this
expectation, we can explicitly document it.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
The architecture specific loaders are now called with a list of all
elements of a given type (rather than a single element at a time). This
avoids the need to reallocate the arrays in CPU maps for each element.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
There's no reason for keeping the models in a linked list. Especially
when we know upfront the total number of models we are loading.
As a nice side effect, this fixes ppc64GetModels to always return a
NULL-terminated list of models.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
There's no reason for keeping the vendors in a linked list. Especially
when we know upfront the total number of models we are loading.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
There's no reason for keeping the features in a linked list. Especially
when we know upfront the total number of features we are loading.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
There's no reason for keeping the vendors in a linked list. Especially
when we know upfront the total number of models we are loading.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
There's no reason for keeping the models in a linked list. Especially
when we know upfront the total number of models we are loading.
As a nice side effect, this fixes x86GetModels to always return a
NULL-terminated list of models.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
For some disk types (SD), we want to emit the syntax
we used for disks before -device was available even
if QEMU supports -device.
Use the qemuDiskBusNeedsDeviceArg helper to figure out
whether to use the old or new syntax.
