[libvirt] "virsh destroy" deletes extra VMs
by Shi Jin
Hi there,
I am running Ubuntu 9.04 X64 server with KVM/libvirt. I recently found that the "virsh destroy" command actually delete VMs it should not delete.
For example, please see the following screen copy:
root@node1:/srv/cloud/ImgRep/WinXP# virsh list
Connecting to uri: qemu:///system
Id Name State
----------------------------------
27 one-80 running
28 one-81 running
29 one-83 running
30 one-82 running
31 one-79 running
root@node1:/srv/cloud/ImgRep/WinXP# virsh destroy 29
Connecting to uri: qemu:///system
Domain 29 destroyed
root@node1:/srv/cloud/ImgRep/WinXP# virsh list
Connecting to uri: qemu:///system
Id Name State
----------------------------------
27 one-80 running
28 one-81 running
31 one-79 running
Please note that VM 30 was deleted when we tried to delete VM 29. The /var/log/messages showed the following information:
ep 30 08:58:59 node1 kernel: [238101.304619] br1: port 5(vnet4) entering disabled state
Sep 30 08:58:59 node1 kernel: [238101.342851] device vnet4 left promiscuous mode
Sep 30 08:58:59 node1 kernel: [238101.342854] br1: port 5(vnet4) entering disabled state
Sep 30 08:58:59 node1 kernel: [238101.424851] br1: port 4(vnet3) entering disabled state
Sep 30 08:59:00 node1 kernel: [238101.463031] device vnet3 left promiscuous mode
Sep 30 08:59:00 node1 kernel: [238101.463035] br1: port 4(vnet3) entering disabled state
Sep 30 09:10:53 node1 -- MARK --
Sep 30 09:30:53 node1 -- MARK --
Sep 30 09:50:53 node1 -- MARK --
Sep 30 09:51:31 node1 kernel: [241253.266955] device vnet3 entered promiscuous mode
Sep 30 09:51:31 node1 kernel: [241253.268857] br1: port 4(vnet3) entering learning state
Sep 30 09:51:40 node1 kernel: [241262.260147] br1: topology change detected, propagating
Sep 30 09:51:40 node1 kernel: [241262.260152] br1: port 4(vnet3) entering forwarding state
Sep 30 09:53:59 node1 kernel: [241401.204570] br1: port 4(vnet3) entering disabled state
Sep 30 09:53:59 node1 kernel: [241401.232395] device vnet3 left promiscuous mode
Sep 30 09:53:59 node1 kernel: [241401.232399] br1: port 4(vnet3) entering disabled state
Sep 30 09:54:20 node1 kernel: [241421.694803] br1: port 2(vnet0) entering disabled state
Sep 30 09:54:20 node1 kernel: [241421.733095] device vnet0 left promiscuous mode
Sep 30 09:54:20 node1 kernel: [241421.733099] br1: port 2(vnet0) entering disabled state
Sep 30 09:54:20 node1 kernel: [241421.815209] br2: port 2(vnet1) entering disabled state
Sep 30 09:54:20 node1 kernel: [241421.853728] device vnet1 left promiscuous mode
Sep 30 09:54:20 node1 kernel: [241421.853732] br2: port 2(vnet1) entering disabled state
Sep 30 09:54:20 node1 kernel: [241421.935254] br1: port 3(vnet2) entering disabled state
Sep 30 09:54:20 node1 kernel: [241421.963527] device vnet2 left promiscuous mode
Sep 30 09:54:20 node1 kernel: [241421.963531] br1: port 3(vnet2) entering disabled state
Sep 30 09:55:29 node1 kernel: [241490.644204] br1: port 6(vnet5) entering disabled state
Sep 30 09:55:29 node1 kernel: [241490.673195] device vnet5 left promiscuous mode
Sep 30 09:55:29 node1 kernel: [241490.673199] br1: port 6(vnet5) entering disabled state
Sep 30 10:10:53 node1 -- MARK --
Sep 30 10:11:49 node1 kernel: [242470.955142] device vnet0 entered promiscuous mode
Sep 30 10:11:49 node1 kernel: [242470.957051] br1: port 2(vnet0) entering learning state
Sep 30 10:11:52 node1 kernel: [242474.180036] device vnet1 entered promiscuous mode
Sep 30 10:11:52 node1 kernel: [242474.181943] br1: port 3(vnet1) entering learning state
Sep 30 10:11:55 node1 kernel: [242477.431693] device vnet2 entered promiscuous mode
Sep 30 10:11:55 node1 kernel: [242477.433558] br1: port 4(vnet2) entering learning state
Sep 30 10:11:58 node1 kernel: [242479.950082] br1: topology change detected, propagating
Sep 30 10:11:58 node1 kernel: [242479.950086] br1: port 2(vnet0) entering forwarding state
Sep 30 10:12:00 node1 kernel: [242482.059146] device vnet3 entered promiscuous mode
Sep 30 10:12:00 node1 kernel: [242482.060330] br1: port 5(vnet3) entering learning state
Sep 30 10:12:01 node1 kernel: [242483.180170] br1: topology change detected, propagating
Sep 30 10:12:01 node1 kernel: [242483.180173] br1: port 3(vnet1) entering forwarding state
Sep 30 10:12:04 node1 kernel: [242485.710948] device vnet4 entered promiscuous mode
Sep 30 10:12:04 node1 kernel: [242485.712109] br1: port 6(vnet4) entering learning state
Sep 30 10:12:04 node1 kernel: [242486.430098] br1: topology change detected, propagating
Sep 30 10:12:04 node1 kernel: [242486.430104] br1: port 4(vnet2) entering forwarding state
Sep 30 10:12:09 node1 kernel: [242491.051857] br1: topology change detected, propagating
Sep 30 10:12:09 node1 kernel: [242491.051862] br1: port 5(vnet3) entering forwarding state
Sep 30 10:12:13 node1 kernel: [242494.711792] br1: topology change detected, propagating
Sep 30 10:12:13 node1 kernel: [242494.711797] br1: port 6(vnet4) entering forwarding state
Sep 30 10:22:00 node1 kernel: [243082.274008] br1: port 4(vnet2) entering disabled state
Sep 30 10:22:00 node1 kernel: [243082.312108] device vnet2 left promiscuous mode
Sep 30 10:22:00 node1 kernel: [243082.312112] br1: port 4(vnet2) entering disabled state
Sep 30 10:22:00 node1 kernel: [243082.365331] br1: port 5(vnet3) entering disabled state
Sep 30 10:22:00 node1 kernel: [243082.393430] device vnet3 left promiscuous mode
Sep 30 10:22:00 node1 kernel: [243082.393433] br1: port 5(vnet3) entering disabled state
I have 4 NICs and the ifconfig shows
root@node1:/srv/cloud/ImgRep/WinXP# ifconfig
bond0 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57
inet6 addr: fe80::224:e8ff:fe6b:d57/64 Scope:Link
UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
RX packets:7799931 errors:0 dropped:0 overruns:0 frame:0
TX packets:6879357 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:4360166425 (4.3 GB) TX bytes:4350243659 (4.3 GB)
br0 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57
inet addr:192.168.1.128 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::224:e8ff:fe6b:d57/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:7641913 errors:0 dropped:0 overruns:0 frame:0
TX packets:6412260 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:4169876739 (4.1 GB) TX bytes:4291182908 (4.2 GB)
br1 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5b
inet6 addr: fe80::224:e8ff:fe6b:d5b/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:152606 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:11276026 (11.2 MB) TX bytes:468 (468.0 B)
br2 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5d
inet6 addr: fe80::224:e8ff:fe6b:d5d/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:152596 errors:0 dropped:0 overruns:0 frame:0
TX packets:6 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:11286132 (11.2 MB) TX bytes:468 (468.0 B)
eth0 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:3622205 errors:0 dropped:0 overruns:0 frame:0
TX packets:3443103 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:1758147112 (1.7 GB) TX bytes:2174389369 (2.1 GB)
Interrupt:36 Memory:d6000000-d6012700
eth1 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:57
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:4177726 errors:0 dropped:0 overruns:0 frame:0
TX packets:3436254 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:2602019313 (2.6 GB) TX bytes:2175854290 (2.1 GB)
Interrupt:48 Memory:d8000000-d8012700
eth2 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5b
inet6 addr: fe80::224:e8ff:fe6b:d5b/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:256014 errors:0 dropped:0 overruns:0 frame:0
TX packets:91549 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:20680139 (20.6 MB) TX bytes:17317087 (17.3 MB)
Interrupt:32 Memory:da000000-da012700
eth3 Link encap:Ethernet HWaddr 00:24:e8:6b:0d:5d
inet6 addr: fe80::224:e8ff:fe6b:d5d/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:245189 errors:0 dropped:0 overruns:0 frame:0
TX packets:32669 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:20866257 (20.8 MB) TX bytes:8265329 (8.2 MB)
Interrupt:42 Memory:dc000000-dc012700
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:324787 errors:0 dropped:0 overruns:0 frame:0
TX packets:324787 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:571707504 (571.7 MB) TX bytes:571707504 (571.7 MB)
vnet0 Link encap:Ethernet HWaddr 52:c3:a3:d0:09:84
inet6 addr: fe80::50c3:a3ff:fed0:984/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:72 errors:0 dropped:0 overruns:0 frame:0
TX packets:533 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:500
RX bytes:10410 (10.4 KB) TX bytes:49306 (49.3 KB)
vnet1 Link encap:Ethernet HWaddr 02:0d:4c:1e:5c:82
inet6 addr: fe80::d:4cff:fe1e:5c82/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:36 errors:0 dropped:0 overruns:0 frame:0
TX packets:563 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:500
RX bytes:6023 (6.0 KB) TX bytes:53245 (53.2 KB)
vnet4 Link encap:Ethernet HWaddr a6:58:67:2b:37:8a
inet6 addr: fe80::a458:67ff:fe2b:378a/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:42 errors:0 dropped:0 overruns:0 frame:0
TX packets:552 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:500
RX bytes:6733 (6.7 KB) TX bytes:52189 (52.1 KB)
Note that there is no IP address assigned to the br1 device which is used here. I don't think that should be a problem. Also the first 2 NICs are bonded first and then created as a bridge br0 and I don't think it is a problem either.
root@node1:/srv/cloud/ImgRep/WinXP# virsh --version
Connecting to uri: qemu:///system
0.6.1
root@node1:/srv/cloud/ImgRep/WinXP# kvm -h
QEMU PC emulator version 0.9.1 (kvm-84), Copyright (c) 2003-2008 Fabrice Bellard
Has anyone had similar problem before? How can I fix it?
Thanks a lot.
Shi
--
Shi Jin, PhD
14 years, 7 months
[libvirt] [PATCH] Fix emission of domain events messages
by Daniel P. Berrange
The code which updated the message length after writing the
payload wrote the updated length word in the wrong place since
the XDR object was given a buffer pointing to the start of the
header payload, rather than message start.
* daemon/remote.c: Fix updating of event message length so that
we actually send the payload, not just the header
---
daemon/remote.c | 14 ++++++++------
1 files changed, 8 insertions(+), 6 deletions(-)
diff --git a/daemon/remote.c b/daemon/remote.c
index 6d4cc1c..2b58045 100644
--- a/daemon/remote.c
+++ b/daemon/remote.c
@@ -4559,17 +4559,19 @@ remoteDispatchDomainEventSend (struct qemud_client *client,
/* Serialise the return header and event. */
xdrmem_create (&xdr,
- msg->buffer + msg->bufferOffset,
- msg->bufferLength - msg->bufferOffset,
+ msg->buffer,
+ msg->bufferLength,
XDR_ENCODE);
- if (!xdr_remote_domain_event_msg(&xdr, data))
+ /* Skip over the header we just wrote */
+ if (xdr_setpos (&xdr, msg->bufferOffset) == 0)
goto xdr_error;
+ if (!xdr_remote_domain_event_msg(&xdr, data))
+ goto xdr_error;
- /* Update length word */
- msg->bufferOffset += xdr_getpos (&xdr);
- len = msg->bufferOffset;
+ /* Update length word to include payload*/
+ len = msg->bufferOffset = xdr_getpos (&xdr);
if (xdr_setpos (&xdr, 0) == 0)
goto xdr_error;
--
1.6.2.5
14 years, 7 months
[libvirt] Tunnelled migration
by Chris Lalancette
At long last, now that the stream stuff is committed to the tree,
I can repost the tunnelled migration patches. Note that these
patches will conflict in some places with danpb's previously
posted "PEER2PEER" patches, but it should be easy to
resolve the differences.
14 years, 7 months
[libvirt] Fix apibuild.py warnings
by Matthias Bolte
apibuild.py warns about misformatted function comment for
virStreamEventAddCallback, virStreamEventUpdateCallback and
virStreamEventRemoveCallback.
The attached patch adds the missing ':' and contains the resulting
changes in docs/libvirt-api.xml and docs/libvirt-refs.xml
Matthias
14 years, 7 months
[libvirt] [PATCH 0/7] Add support for qcow2 backing files to sVirt
by Mark McLoughlin
Hey,
https://bugzilla.redhat.com/497131 shows that sVirt cannot
currently handle qcow2 files with a backing store. Because we do
not re-label the backing store, qemu cannot read it.
The fix is fairly straightforward - when we go to re-label
a file, check its header and if it has a backing store, re-label
that too.
Most of this series involves moving the header probing code
from the storage driver into libvirt_util so that the security
driver can use it.
Cheers,
Mark.
14 years, 7 months
[libvirt] Fix memleaks in libvirtd's message processing
by Matthias Bolte
I tracked down a memleak in libvirtd's message processing. The leak
was introduced in commit 47cab734995fa9521b1df05d37e9978eedd8d3a2
"Split out code for handling incoming method call messages". This
commit changed the way qemud_client_message objects were reused.
Before this commit:
- qemudWorker() removes a message from dispatch queue and passes it to
remoteDispatchClientRequest()
- remoteDispatchClientRequest() handles the message and reuses the
same message for the response. If an error occurs the same message is
used to report it (rpc_error label). If another error occurs while
trying to report the first error remoteDispatchClientRequest() would
return -1 and the message will be freed in qemudWorker()
After this commit:
- qemudWorker() removes a message from dispatch queue and passes it to
remoteDispatchClientRequest()
- remoteDispatchClientRequest() calls remoteDispatchClientCall() if
the message it is an remote call or otherwise respond with a new error
message by calling remoteSerializeReplyError() and the original
message leaks
- remoteDispatchClientCall() handles the message and reuses the same
message for the response. If an error occurs it calls
remoteSerializeReplyError() and the original message leaks. If a fatal
error occurs remoteDispatchClientCall() returns -1 and the original
message will be freed in qemudWorker()
To fix this leak the original message has to be freed if
remoteSerializeReplyError() succeeds. If remoteSerializeReplyError()
fails the original message will be freed in qemudWorker().
In addition I came across another leak in remoteSerializeError(). If
an error occurs while serializing the initial error the message leaks.
To fix this leak the message has to be freed in case of an XDR error.
Matthias
14 years, 7 months
[libvirt] virtio network problems with kvm guests with 2.6.26 kernel
by Jim Paris
Hi,
After upgrading libvirt on one system, my kvm guests running a 2.6.26
kernel with virtio networking could no longer communicate with the
network.
It seems that the problem is caused by newer libvirt versions
unconditionally enabling GSO support by setting IFF_VNET_HDR [1].
However, support for this feature is apparently broken in 2.6.26.
Ubuntu seems to have also discovered this problem [2] and they
currently work around it by removing GSO support from kvm [3].
Currently, I'm running with a patched libvirt that never sets
IFF_VNET_HDR, and it's working fine once again.
While this is not a libvirt problem per se, it certainly violated the
principle of least surprise given that libvirt was the only thing that
changed! But I don't know if there's much that libvirt can do to
avoid this problem, as it's really something that needs to get fixed
in kvm, or the guest needs to be upgraded, or switched away from
virtio.
At the very least, maybe this info helps someone else who runs across
the same problem.
-jim
[1] http://www.redhat.com/archives/libvir-list/2009-January/msg00642.html
[2] https://bugs.launchpad.net/kvm/+bug/331128
[3] http://patches.ubuntu.com/k/kvm/extracted/virtio-net_disable_gso.patch
14 years, 7 months
[libvirt] [PATCH] Minor comment changes.
by Laine Stump
Fix some minor grammer (and one other) nits in comments that end up in
generated API reference documentation. No functional/binary differences.
---
src/libvirt.c | 24 ++++++++++++------------
1 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/src/libvirt.c b/src/libvirt.c
index ca8e003..ce3d6ba 100644
--- a/src/libvirt.c
+++ b/src/libvirt.c
@@ -592,9 +592,9 @@ virRegisterNetworkDriver(virNetworkDriverPtr driver)
/**
* virRegisterInterfaceDriver:
- * @driver: pointer to a interface driver block
+ * @driver: pointer to an interface driver block
*
- * Register a interface virtualization driver
+ * Register an interface virtualization driver
*
* Returns the driver priority or -1 in case of error.
*/
@@ -5365,7 +5365,7 @@ error:
/**
* virNetworkGetXMLDesc:
* @network: a network object
- * @flags: and OR'ed set of extraction flags, not used yet
+ * @flags: an OR'ed set of extraction flags, not used yet
*
* Provide an XML description of the network. The description may be reused
* later to relaunch the network with virNetworkCreateXML().
@@ -5545,7 +5545,7 @@ error:
/**
* virInterfaceGetConnect:
- * @iface: pointer to a interface
+ * @iface: pointer to an interface
*
* Provides the connection pointer associated with an interface. The
* reference counter on the connection is not increased by this
@@ -5819,7 +5819,7 @@ error:
/**
* virInterfaceGetName:
- * @iface: a interface object
+ * @iface: an interface object
*
* Get the public name for that interface
*
@@ -5842,9 +5842,9 @@ virInterfaceGetName(virInterfacePtr iface)
/**
* virInterfaceGetMACString:
- * @iface: a interface object
+ * @iface: an interface object
*
- * Get the MAC for a interface as string. For more information about
+ * Get the MAC for an interface as string. For more information about
* MAC see RFC4122.
*
* Returns a pointer to the MAC address (in null-terminated ASCII
@@ -5867,11 +5867,11 @@ virInterfaceGetMACString(virInterfacePtr iface)
/**
* virInterfaceGetXMLDesc:
- * @iface: a interface object
- * @flags: and OR'ed set of extraction flags, not used yet
+ * @iface: an interface object
+ * @flags: an OR'ed set of extraction flags, not used yet
*
* Provide an XML description of the interface. The description may be reused
- * later to recreate the interface with virInterfaceCreateXML().
+ * later to redefine the interface with virInterfaceDefineXML().
*
* Returns a 0 terminated UTF-8 encoded XML instance, or NULL in case of error.
* the caller must free() the returned value.
@@ -6100,7 +6100,7 @@ error:
* This method is typically useful for applications where multiple
* threads are using a connection, and it is required that the
* connection remain open until all threads have finished using
- * it. ie, each new thread using a interface would increment
+ * it. ie, each new thread using an interface would increment
* the reference count.
*
* Returns 0 in case of success, -1 in case of failure.
@@ -6121,7 +6121,7 @@ virInterfaceRef(virInterfacePtr iface)
/**
* virInterfaceFree:
- * @iface: a interface object
+ * @iface: an interface object
*
* Free the interface object. The interface itself is unaltered.
* The data structure is freed and should not be used thereafter.
--
1.6.2.5
14 years, 7 months
