Am 21.02.22 um 10:21 schrieb Strahil Nikolov:
Are you building your VMs from template ?

No everything by hand.

Any chance there is another system (outside oVirt ?) with the same MAC ?

you can run arping and check if you get response from more than 1 system.

I have check that to, but no the mac is unique.

The problem with " ifup: failed to bring up enp1s0" I could also solve and I think it has nothing to do with the problem. I used a custom script in /etc/network/if-up.d/ for mounting my fstab correctly.


On Mon, Feb 21, 2022 at 11:15, jb
<jonbae77@gmail.com> wrote:

Thank you Nikolov,

I have descript the problem a bit wrong. In the UI is see the interface, with virsh dumpxml I get:

  <interface type='bridge'>
      <mac address='00:1a:4a:16:01:83'/>
      <source bridge='ovirtmgmt'/>
      <target dev='vnet9'/>
      <model type='virtio'/>
      <driver name='vhost' queues='4'/>
      <filterref filter='vdsm-no-mac-spoofing'/>
      <link state='up'/>
      <mtu size='1500'/>
      <alias name='ua-c8a50041-2d13-456d-acb6-b57fdaea434b'/>
      <address type='pci' domain='0x0000' bus='0x01' slot='0x00' function='0x0'/>
  </interface>

lspci on the VM show also the interface and I can also bring the interface up again with: ifup enp1s0. So it only loose the connection, not the interface.

When I restart the VM I get this log message:

/var/log/syslog:Feb 21 10:11:29 tv-planer sh[391]: ifup: failed to bring up enp1s0
/var/log/syslog:Feb 21 10:11:29 tv-planer systemd[1]: ifup@enp1s0.service: Main process exited, code=exited, status=1/FAILURE
/var/log/syslog:Feb 21 10:11:29 tv-planer systemd[1]: ifup@enp1s0.service: Failed with result 'exit-code'.

But the VM got his IP and is reachable.



Am 20.02.22 um 14:06 schrieb Strahil Nikolov:
Do you see all nic in the UI ? What type are they ? 

Set this alias on the Hypervisors:
alias virsh='virsh -c qemu:///system?authfile=/etc/ovirt-hosted-engine/virsh_auth.conf'
 
and then use 'virsh dumpxml name-of-vm' to identify how many nics the vm has .

If gou got the correct settings in ovirt, use 'lspci -vvvvv' .

Best Regards,
Strahil Nikolov

On Sun, Feb 20, 2022 at 11:32, Jonathan Baecker
Hello everybody,

I have here a strange behavior: We have a 3 node self hosted cluster
with around 20 VMs running on it. Since longer I had the problem with
one VM that after some days it lose the network interface. But because
this VM was only for testing I was to lazy to dive more in, to figure
out what is happen.

Now I have a second VM, with the same problem and this VM is more
important. Both VMs running debian 10 and use cifs mounts, so maybe that
is related?

Have some one of you seeing this behavior? And can give me a hint, how I
can fix this?

At the moment I can't provide a log file, because I didn't know the
exact time, when this was happen. And I also don't know, if the problem
comes from ovirt or from the operating system inside the VMs.

Have a nice day!

Jonathan

_______________________________________________
Users mailing list -- users@ovirt.org
To unsubscribe send an email to users-leave@ovirt.org