The current virDomainMigrateFinish3 method signature attempts to
distinguish two types of errors, by allowing return with ret== 0,
but ddomain == NULL, to indicate a failure to start the guest.
This is flawed, because when ret == 0, there is no way for the
virErrorPtr details to be sent back to the client.
Change the signature of virDomainMigrateFinish3 so it simply
returns a virDomainPtr, in the same way as virDomainMigrateFinish2
The disk locking code will protect against the only possible
failure mode this doesn't account for (loosing conenctivity to
libvirtd after Finish3 starts the CPUs, but before the client
sees the reply for Finish3).
* src/driver.h, src/libvirt.c, src/libvirt_internal.h: Change
virDomainMigrateFinish3 to return a virDomainPtr instead of int
* src/remote/remote_driver.c, src/remote/remote_protocol.x,
daemon/remote.c, src/qemu/qemu_driver.c, src/qemu/qemu_migration.c:
Update for API change
The virDomainMigratePerform3 currently has a single URI parameter
whose meaning varies. It is either
- A QEMU migration URI (normal migration)
- A libvirtd connection URI (peer2peer migration)
Unfortunately when using peer2peer migration, without also
using tunnelled migration, it is possible that both URIs are
required.
This adds a second URI parameter to the virDomainMigratePerform3
method, to cope with this scenario. Each parameter how has a fixed
meaning.
NB, there is no way to actually take advantage of this yet,
since virDomainMigrate/virDomainMigrateToURI do not have any
way to provide the 2 separate URIs
* daemon/remote.c, src/remote/remote_driver.c,
src/remote/remote_protocol.x, src/remote_protocol-structs: Add
the second URI parameter to perform3 message
* src/driver.h, src/libvirt.c, src/libvirt_internal.h: Add
the second URI parameter to Perform3 method
* src/libvirt_internal.h, src/qemu/qemu_migration.c,
src/qemu/qemu_migration.h: Update to handle URIs correctly
This extends the v3 migration protocol such that the
virDomainMigrateBegin3 and virDomainMigratePerform3
methods accept an application supplied XML config for
the target VM.
If the 'xmlin' parameter is NULL, then Begin3 uses the
current guest XML as normal. A driver implementing the
Begin3 method should either reject all non-NULL 'xmlin'
parameters, or strictly validate that the app supplied
XML does not change guest ABI.
The Perform3 method also needed the xmlin parameter to
cope with the Peer2Peer migration sequence.
NB it is not yet possible to use this capability since
neither of the public virDomainMigrate/virDomainMigrateToURI
methods have a way to pass in XML.
* daemon/remote.c, src/remote/remote_driver.c,
src/remote/remote_protocol.x, src/remote_protocol-structs:
Add 'remote_string xmlin' parameter to begin3/perform3
RPC messages
* src/libvirt.c, src/driver.h, src/libvirt_internal.h: Add
'const char *xmlin' parameter to Begin3/Perform3 methods
* src/qemu/qemu_driver.c, src/qemu/qemu_migration.c,
src/qemu/qemu_migration.h: Pass xmlin parameter around
migration methods
Migration just seems to go from bad to worse. We already had to
introduce a second migration protocol when adding the QEMU driver,
since the one from Xen was insufficiently flexible to cope with
passing the data the QEMU driver required.
It turns out that this protocol still has some flaws that we
need to address. The current sequence is
* Src: DumpXML
- Generate XML to pass to dst
* Dst: Prepare
- Get ready to accept incoming VM
- Generate optional cookie to pass to src
* Src: Perform
- Start migration and wait for send completion
- Kill off VM if successful, resume if failed
* Dst: Finish
- Wait for recv completion and check status
- Kill off VM if unsuccessful
The problems with this are:
- Since the first step is a generic 'DumpXML' call, we can't
add in other migration specific data. eg, we can't include
any VM lease data from lock manager plugins
- Since the first step is a generic 'DumpXML' call, we can't
emit any 'migration begin' event on the source, or have
any hook that runs right at the start of the process
- Since there is no final step on the source, if the Finish
method fails to receive all migration data & has to kill
the VM, then there's no way to resume the original VM
on the source
This patch attempts to introduce a version 3 that uses the
improved 5 step sequence
* Src: Begin
- Generate XML to pass to dst
- Generate optional cookie to pass to dst
* Dst: Prepare
- Get ready to accept incoming VM
- Generate optional cookie to pass to src
* Src: Perform
- Start migration and wait for send completion
- Generate optional cookie to pass to dst
* Dst: Finish
- Wait for recv completion and check status
- Kill off VM if failed, resume if success
- Generate optional cookie to pass to src
* Src: Confirm
- Kill off VM if success, resume if failed
The API is designed to allow both input and output cookies
in all methods where applicable. This lets us pass around
arbitrary extra driver specific data between src & dst during
migration. Combined with the extra 'Begin' method this lets
us pass lease information from source to dst at the start of
migration
Moving the killing of the source VM out of Perform and
into Confirm, means we can now recover if the dst host
can't successfully Finish receiving migration data.
Introduces several new public API options for migration
- VIR_MIGRATE_PEER2PEER: With this flag the client only
invokes the virDomainMigratePerform method, expecting
the source host driver to do whatever is required to
complete the entire migration process.
- VIR_MIGRATE_TUNNELLED: With this flag the actual data
for migration will be tunnelled over the libvirtd RPC
channel. This requires that VIR_MIGRATE_PEER2PEER is
also set.
- virDomainMigrateToURI: This is variant of the existing
virDomainMigrate method which does not require any
virConnectPtr for the destination host. Given suitable
driver support, this allows for all the same modes as
virDomainMigrate()
The URI for VIR_MIGRATE_PEER2PEER must be a valid libvirt
URI. For non-p2p migration a hypervisor specific migration
URI is used.
virDomainMigrateToURI without a PEER2PEER flag is only
support for Xen currently, and it involves XenD talking
directly to XenD, no libvirtd involved at all.
* include/libvirt/libvirt.h.in: Add VIR_MIGRATE_PEER2PEER
flag for migration
* src/libvirt_internal.h: Add feature flags for peer to
peer migration (VIR_FEATURE_MIGRATE_P2P) and direct
migration (VIR_MIGRATE_PEER2PEER mode)
* src/libvirt.c: Implement support for VIR_MIGRATE_PEER2PEER
and virDomainMigrateToURI APIs.
* src/xen/xen_driver.c: Advertise support for DIRECT migration
* src/xen/xend_internal.c: Add TODO item for p2p migration
* src/libvirt_public.syms: Export virDomainMigrateToURI
method
* src/qemu/qemu_driver.c: Add support for PEER2PEER and
migration, and adapt TUNNELLED migration.
* tools/virsh.c: Add --p2p and --direct args and use the
new virDomainMigrateToURI method where possible.
Since virMigratePrepareTunnel() is used for migration over the
native libvirt connection, there is never any need to pass the
target URI to this method.
* daemon/remote.c, src/driver.h, src/libvirt.c, src/libvirt_internal.h,
src/qemu/qemu_driver.c, src/remote/remote_driver.c,
src/remote/remote_protocol.c, src/remote/remote_protocol.h,
src/remote/remote_protocol.x: Remove 'uri_in' parameter from
virMigratePrepareTunnel() method
Move the VIR_DRV_FEATURE* constants into libvirt_internal.h
since these flags are indicating whether APIs in the
libvirt_internal.h file are supported by a driver
* src/driver.h: Remove VIR_DRV_FEATURE* constants
* src/libvirt_internal.h: Add VIR_DRV_FEATURE* constants, using
an enum instead of #define
* src/internal.h: pull in libvirt_internal.h
Implementation of tunnelled migration, using a Unix Domain Socket
on the qemu backend. Note that this requires very new versions of
qemu (0.10.7 at least) in order to get the appropriate bugfixes.
Signed-off-by: Chris Lalancette <clalance@redhat.com>
Add a VIR_SECRET_GET_VALUE_INTERNAL_CALL flag value, replacing the
originally separate libvirt_internal_call parameter. The flag is used
to differentiate external virSecretGetValue() calls from internal calls
by libvirt drivers that need to use the secret even if it is private.
* src/libvirt_internal.h Remove VIR_DOMAIN_XML_FLAGS_MASK
* src/driver.h Add VIR_SECRET_GET_VALUE_FLAGS_MASK constant and
VIR_SECRET_GET_VALUE_INTERNAL_CALL. Re-add the
VIR_DOMAIN_XML_FLAGS_MASK constant
* src/libvirt.c (virSecretGetValue): Don't allow the user to specify
internal flags.
We need to store things like device names and PCI slot numbers in the
qemu domain state file so that we don't lose that information on
libvirtd restart. Add a flag to indicate that this information should
be parsed or formatted.
Make bit 16 and above of the flags bitmask for internal use only and
consume the first bit for this new status flag.
* include/libvirt/libvirt.h: add VIR_DOMAIN_XML_FLAGS_MASK
* src/libvirt.c: reject private flags in virDomainGetXMLDesc()
* src/domain_conf.h: add VIR_DOMAIN_XML_INTERNAL_STATUS
* src/domain_conf.c: pass the flag from virDomainObjParseXML() and
virDomainSaveStatus
* qemud/remote.c qemud/remote_dispatch_localvars.h
qemud/remote_dispatch_proc_switch.h qemud/remote_dispatch_prototypes.h
qemud/remote_protocol.c qemud/remote_protocol.h
qemud/remote_protocol.x src/driver.h src/libvirt.c
src/libvirt_internal.h src/libvirt_sym.version src/lxc_driver.c
src/openvz_driver.c src/qemu_conf.h src/qemu_driver.c
src/remote_internal.c src/test.c src/virsh.c: large patch to
add migration support for KVM/QEmu, based on the work of Rich Jones
and Chris Lalancette. This introduce a new version of the prepare
and finish steps of the migration, so changes the driver API and
a lot of code.
Daniel
