"libvirtError: unsupported configuration: CPU specification not supported by hypervisor" on ci slave

[Aleksandra Fedorova]

We get libvirt error
   libvirtError: unsupported configuration: CPU specification not supported by hypervisor

while running test on ci-slave17:

https://ci.fuel-infra.org/job/master.fuel-library.pkgs.ubuntu.neutron_vlan_ha/1474/console

Please investigate.

The slave is taken offline for now, please contact CI team to take it back once the issue is fixed

[Pawel Brzozowski (pbrzozowski)]

Looks like file system problem:

root@ci-slave17:~# fsck.ext4 -f -n /dev/mapper/vg00-root
e2fsck 1.42.9 (4-Feb-2014)
Warning! /dev/mapper/vg00-root is mounted.
Warning: skipping journal recovery because doing a read-only filesystem check.
Pass 1: Checking inodes, blocks, and sizes
Inodes that were part of a corrupted orphan linked list found. Fix? no

Inode 2100576 was part of the orphaned inode list. IGNORED.
Inode 2100577 was part of the orphaned inode list. IGNORED.
Inode 2102638 was part of the orphaned inode list. IGNORED.
Inode 2102641 was part of the orphaned inode list. IGNORED.
Inode 15335659 was part of the orphaned inode list. IGNORED.
Deleted inode 19005473 has zero dtime. Fix? no

Inode 20709390 was part of the orphaned inode list. IGNORED.
Inode 20709421 was part of the orphaned inode list. IGNORED.
Inode 20709482 was part of the orphaned inode list. IGNORED.
Inode 20709646 was part of the orphaned inode list. IGNORED.
Inode 20709653 was part of the orphaned inode list. IGNORED.
Inode 20709731 was part of the orphaned inode list. IGNORED.
Inode 20713223 was part of the orphaned inode list. IGNORED.
Inode 20713440 was part of the orphaned inode list. IGNORED.
Inode 38274983 was part of the orphaned inode list. IGNORED.
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
Block bitmap differences: -(8438201--8438211) -(8439874--8439877) -(8459442--8459446) -(15776674--15776677) -(15776713--15776726) -(61375488--61375495) -(128330080--128330106) -153137780 -(153157022--153157036) -153165313 -(153351776--153351802) -(154042688--154042725) -(154043872--154043888) -(154047360--154047438) -(154116352--154116589) -(154133504--154134915)
Fix? no

Free blocks count wrong (210711436, counted=207770216).
Fix? no

Inode bitmap differences: -(2100576--2100577) -2102638 -2102641 -15335659 -19005473 -20709390 -20709421 -20709482 -20709646 -20709653 -20709731 -20713223 -20713440 -38274983
Fix? no

Free inodes count wrong (56846892, counted=56746628).
Fix? no

/dev/mapper/vg00-root: ********** WARNING: Filesystem still has errors **********

/dev/mapper/vg00-root: 636372/57483264 files (0.1% non-contiguous), 19213428/229924864 blocks

[Pawel Brzozowski (pbrzozowski)]

Some control files were broken for some reason, that probably caused problems with libvirt.

[Pawel Brzozowski (pbrzozowski)]

It was not the reason. Still looking for root cause.

[Pavel Boldin (pboldin)]

Is this an output from inside the VM?

[Pawel Brzozowski (pbrzozowski)]

No, all this output comes from host machine.

This is the core problem:

root@ci-slave17:~# /usr/sbin/libvirtd -v
2015-10-09 12:12:38.907+0000: 1533: info : libvirt version: 1.2.2
2015-10-09 12:12:38.907+0000: 1533: error : virNodeParseNode:479 : problem reading /sys/devices/system/node/node0: No such file or directory
2015-10-09 12:12:38.907+0000: 1533: warning : virQEMUCapsInit:908 : Failed to get host CPU

Of course the mentioned file exists.

[Pawel Brzozowski (pbrzozowski)]

Host is redeployed - we were not table to find the root cause. It comed to my mind after redeployment, to compare strace output of broken machine with some new machine - but it was too late.