Task(s) blocked due to CPU's workqueue stalled by hpsa driver on proxmox 3.4-1

[kaltsi]

Hi,

We have same situation with new kernel: 2.6.32-37-pve

https://access.redhat.com/solutions/1179703

After one day lot of processes hang up. Can login through ssh, but nothing I can do.
All logs are clear.

The system has:
RAID bus controller: Hewlett-Packard Company Smart Array Gen8 Controllers (rev 01)
product: Smart Array Gen8 Controllers P420i version: 3.22

root@sp-05:~# pveversion
pve-manager/3.4-1/3f2d890e

The previous kernel works. kernel: 2.6.32-34-pve

Best regards,
Istvan

 

[davlaw]

Just curious, nic/bond config on storage interfaces? I have pretty much the same hardware and issue and have been free of it now for 2 weeks, I disabled LACP on the bonded nics to my SAN. Nothing in my logs yet, just checked. But I don't consider this fixed yet as it has at times plagued me 3-4 weeks out after a reboot.

 

[adamb]

I am running gen 8's with this same card and haven't run into this one yet.

 

[davlaw]

So here I am, 3 weeks out and crash.

 

[adamb]

So here I am, 3 weeks out and crash.

Bummer. I will say in my setup I am not using bonds. Just direct connect SAS to my storage. So far we have 3 of our 25+ clusters running this kernel and no issues as of yet "knock on wood".