100% kworker process

[encore]

Hi,

we are facing issues on all nodes with a 100% kworker process:
http://prntscr.com/kbnzth

Kernel:
4.15.18-1-pve #1

Already testes with Kernel 4.15.10 and with Kernel option "drm_kms_helper.poll=0", thes process starts occuring some hours after a node reboot again.

Any ideas what can cause that?

 

[fireon]

In the past when we hat such problem it always pointed to an hardware problem. Have you done some changes? Was the failure always since pve installation? HW is ok? HW has enough power?

 

[Jean-Pierre]

Hi See this thread,

https://forum.proxmox.com/threads/lxc-restart-creates-kworker-cpu-100.40934/#post-219092

 

[fireon]

Do you have ZFS? I do. Then we are running in do this issue. https://pve.proxmox.com/pipermail/pve-devel/2018-August/033288.html The fix will be available soon in testingrepo.