Bug 729567

I tested in newest environment:

kernel-2.6.32-178.el6.x86_64
libvirt-0.9.4-2.el6.x86_64
virt-manager-0.9.0-5.el6.x86_64
qemu-kvm-0.12.1.2-2.177.el6.x86_64

new migration test in two situations
1, add hostname in /etc/hosts
2. not add hostname

test 1 : success
test 2 : failed

and error is:
Unable to migrate guest: internal error unable to execute QEMU command
'migrate': An undefined error has ocurred

and like this bug  

https://bugzilla.redhat.com/show_bug.cgi?id=615941#c36
"Bug 615941 - Migrate fail with error 'An undefined error has ocurred'."

Two different bugs. 

-------------------------------------------------------------
this bug environment:
virt-manager-0.9.0-5.el6.x86_64
libvirt-0.9.4-1.el6.x86_64
qemu-kvm-0.12.1.2-2.175.el6.x86_64
kernel-2.6.32-175.el6.x86_64

node A： zhpeng              10.66.6.209
node B： pengzhimoutest      10.66.5.246

not edit /etc/hosts

error:
Unable to migrate guest: out of memory

The bug is seems report wrong error message.
------------------------------------------------------------
and tested in newest environment:
virt-manager-0.9.0-5.el6.x86_64
libvirt-0.9.4-2.el6.x86_64
qemu-kvm-0.12.1.2-2.177.el6.x86_64
kernel-2.6.32-178.el6.x86_64

node A： zhpeng              10.66.6.209
node B： pengzhimoutest      10.66.5.246

not edit /etc/hosts

and error is:
Unable to migrate guest: internal error unable to execute QEMU command
'migrate': An undefined error has ocurred

Hmm, I'm seriously confused now. The "Unable to migrate guest: out of memory" error was fixed upstream but not in RHEL (and definitely not in libvirt-0.9.4-2.el6.x86_64) so it's strange you don't see the same error in newer environment. Are you sure nothing else changed? Especially wrt. DNS resolving on both machines? You should be able to reproduce the bug even with 0.9.4-2 version of libvirt.

Patch sent for review:
http://post-office.corp.redhat.com/archives/rhvirt-patches/2011-August/msg00341.html

Sorry about that.

I did it again.

That's Ctrl+c Ctrl+v's mistake

I can reproduce the bug at libvirt-0.9.4-2.el6.x86_64

The same error


Unable to migrate guest: out of memory

Traceback (most recent call last):
  File "/usr/share/virt-manager/virtManager/asyncjob.py", line 44, in cb_wrapper
    callback(asyncjob, *args, **kwargs)
  File "/usr/share/virt-manager/virtManager/migrate.py", line 557, in _async_migrate
    vm.migrate(dstconn, migrate_uri, rate, live, secure, meter=meter)
  File "/usr/share/virt-manager/virtManager/domain.py", line 1123, in migrate
    self._backend.migrate(destconn.vmm, flags, newname, interface, rate)
  File "/usr/lib64/python2.6/site-packages/libvirt.py", line 741, in migrate
    if ret is None:raise libvirtError('virDomainMigrate() failed', dom=self)
libvirtError: out of memory

Hi,
I have Reproduced the bug with :
virt-manager-0.9.0-5.el6.x86_64
libvirt-0.9.4-1.el6.x86_64
qemu-kvm-0.12.1.2-2.175.el6.x86_64
kernel-2.6.32-175.el6.x86_64

And i have verified on the latest version on:
kernel-2.6.32-189.el6.x86_64
libvirt-0.9.4-4.el6.x86_64
qemu-kvm-0.12.1.2-2.183.el6.x86_64
virt-manager-0.9.0-5.el6.x86_64

The steps are as follows:
there are two nodes:
nodeA hostname: zhpeng ip: 10.66.6.209
nodeB hostname: pengzhimoutest ip: 10.66.5.246

1, prepare a nfs server for two nodes. # cat /etc/exports
/tmp/libvirt *(rw,no_root_squash) # service nfs start
2, create a new nfs pool of two nodes. # pool-list --all
Name State Autostart
-----------------------------------------
boot-scratch active yes
default active yes

   1. touch /root/nfs.xml
   2. cat /root/nfs.xml
      <pool type="netfs">
      <name>pool-mig</name>
      <source>
      <host name="10.66.6.209"/>
      <dir path="/tmp/libvirt"/>
      </source>
      <target>
      <path>/var/lib/libvirt/images/new</path>
      </target>
      </pool>

   1. pool-define /root/nfs.xml
      Pool pool-mig defined from /root/nfs.xml
   2. pool-start pool-mig
      Pool pool-mig started
   3. pool-autostart pool-mig
      Pool pool-mig marked as autostarted
   4. pool-list --all
      Name State Autostart
      -----------------------------------------
      boot-scratch active yes
      default active yes
      pool-mig active yes

3, create a new guest use pool-mig or define a new guest from pool-mig on nodeA. # touch /root/mig1.xml # cat mig1.xml
<domain type='kvm'>
<name>mig1</name>
<uuid>21b9be2b-fd3f-bea0-f36f-a0dc7a7cde12</uuid>
<memory>262144</memory>
<currentMemory>262144</currentMemory>
<vcpu>1</vcpu>
<os>
<type arch='x86_64' machine='rhel6.2.0'>hvm</type>
<boot dev='hd'/>
</os>
<features>
<acpi/>
<apic/>
<pae/>
</features>
<clock offset='utc'/>
<on_poweroff>destroy</on_poweroff>
<on_reboot>restart</on_reboot>
<on_crash>restart</on_crash>
<devices>
<emulator>/usr/libexec/qemu-kvm</emulator>
<disk type='file' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<source file='/var/lib/libvirt/images/new/aaa.img'/>
<target dev='vda' bus='virtio'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/>
</disk>
<interface type='network'>
<mac address='52:54:00:c9:02:bc'/>
<source network='default'/>
<model type='virtio'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
</interface>
<input type='tablet' bus='usb'/>
<input type='mouse' bus='ps2'/>
<graphics type='vnc' port='-1' autoport='yes'/>
<video>
<model type='cirrus' vram='9216' heads='1'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/>
</video>
<memballoon model='virtio'>
<address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/>
</memballoon>
</devices>
</domain>

   1. define /root/mig1.xml
      Domain mig1 defined from /root/mig1.xml
   2. list --all
      Id Name State
      ----------------------------------
      - mig1 shut off
   3. start mig1
      Domain mig1 started

4,Ensure two nodes have no FQDN.
on nodeA: # hostname
zhpeng
on nodeB: # hostname
pengzhimoutest

5,Ensure /etc/hosts have no domain name resolve on two nodes 
# cat /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 loclhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6

6, migrate from nodeA to nodeB
[root@zhpeng libvirt]# virsh migrate --live mig1 qemu+ssh://root.5.246/system
root.5.246's password:
error: internal error getaddrinfo failed for 'zhpeng': Name or service not known

7, add domain name resolve to /etc/hosts of two nodes
   # cat /etc/hosts
    10.66.6.209    zhpeng
    10.66.5.246    pengzhimoutest

8, repeat step6
 [root@zhpeng libvirt]# virsh migrate --live mig1 qemu+ssh://root.5.246/system
  there is no error now.

9, check the guest migrated.
   On nodeB:
   # virsh list --all
    Id Name                 State
   ----------------------------------
     7 mig1                running

10, migrate guest back to nodeA and check on nodeA
 [root@pengzhimoutest libvirt]# virsh migrate --live mig1 qemu+ssh://root.6.209/system
 [root@zhpeng libvirt]# virsh list --all
    Id Name                 State
   ----------------------------------
     11 mig1                running


So the error information of step 6 is correct now. It's expected results. The test is pass.

Will re-check this bug once bug 731243 fixed, then move it to VERIFIED.

I recheck the migration with:
kernel-2.6.32-192.el6.x86_64
libvirt-0.9.4-6.el6.x86_64
qemu-kvm-0.12.1.2-2.184.el6.x86_64
virt-manager-0.9.0-5.el6.x86_64

The test is pass.

According to Comment 10 and Comment 8, move it to VERIFIED.

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

http://rhn.redhat.com/errata/RHBA-2011-1513.html