[PVE-User] PVE 7.2 unstability

Eneko Lacunza elacunza at binovo.es
Thu May 12 15:15:09 CEST 2022


Hi,

This time VM didn't crash but kernel noticed the time issue:

May 12 09:48:57 monitor-cloud systemd[1]: session-38.scope: Succeeded.
May 12 18:08:57 monitor-cloud kernel: [31097.014795] clocksource: 
timekeeping watchdog on CPU0: Marking clocksource 'tsc' as unstable 
because the skew is too large:
May 12 18:08:57 monitor-cloud kernel: [31097.014803] 
clocksource:                       'kvm-clock' wd_now: a601cec4c1d9 
wd_last: 8aba12430493 mask: ffffffffffffffff
May 12 18:08:57 monitor-cloud kernel: [31097.014806] 
clocksource:                       'tsc' cs_now: 1f116978cedab cs_last: 
19f66ae2f9bbb mask: ffffffffffffffff
May 12 18:08:57 monitor-cloud kernel: [31097.014810] tsc: Marking TSC 
unstable due to clocksource watchdog
May 12 09:49:02 monitor-cloud systemd[1]: Starting Clean php session 
files...

Seems that the issue is more easily triggered live migrating the VMs, 
another VM just hung but no time-issues in syslog (I had to hard reset...)

We have downgraded from pve-qemu-kvm:amd64 6.2.0-5 to 6.2.0-2 (version 
before issues started)

We have downgraded from qemu-server from 7.2-2 to 7.1-4 (version before 
issues started):

Issue continues.

We have seen that when bulk migrating VMs from node1 to node2, VMs in 
node2 ALSO start to have issues.

We'll try setting max workers for bulk actions to 1 next.


El 12/5/22 a las 9:33, Eneko Lacunza via pve-user escribió:
> _______________________________________________
> pve-user mailing list
> pve-user at lists.proxmox.com
> https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-user

Eneko Lacunza
Zuzendari teknikoa | Director técnico
Binovo IT Human Project

Tel. +34 943 569 206 |https://www.binovo.es
Astigarragako Bidea, 2 - 2º izda. Oficina 10-11, 20180 Oiartzun

https://www.youtube.com/user/CANALBINOVO
https://www.linkedin.com/company/37269706/



More information about the pve-user mailing list