Stopped VM raise zabbix notification of the network interface due backup

[czechsys]

Hi,
running backup of the stopped VM regularly and when it meets some zabbix check, it will raise notification because VM interface on PVE is going up/down:

PVE:

INFO: Finished Backup of VM 101 (00:00:10)
INFO: Backup finished at 2022-02-11 02:00:35
INFO: Starting Backup of VM 102 (qemu)
INFO: Backup started at 2022-02-11 02:00:40
INFO: status = stopped
INFO: backup mode: stop
INFO: ionice priority: 7
...
INFO: stopping kvm after backup task
INFO: Finished Backup of VM 102 (00:00:59)
INFO: Backup finished at 2022-02-11 02:01:39

Zabbix:

Problem started at 02:00:44 on 2022.02.11
Problem name: Interface tap102i1: Link down
Host: PVE
Severity: Average
Operational data: Current state: down (2)

Can PVE handle this case better? Monitored network interfaces are generated automatically by zabbix templates.

 

[Stoiko Ivanov]

running backup of the stopped VM regularly and when it meets some zabbix check, it will raise notification because VM interface on PVE is going up/down:

I would consider this an issue of zabbix - you need to teach the agent to ignore interfaces with names starting with `tap` or `fwbr`

one example where this is also problematic is in a cluster - if you migrate a guest the interfaces will wander from source to target node

sadly am not too experienced with zabbix - so have no idea if this can easily be done (for some other monitoring solutions I think you can ignore interfaces based on their name)

I hope this helps!