Kernel Error: pve-2.6.32-18

[nandex]

Hi, I have a RAID-5 with multipath and I have this kernel error. Always in prompt.

end_request: I/O error, dev sdc, sector 0
end_request: I/O error, dev sdd, sector 0
end_request: I/O error, dev sdg, sector 0
end_request: I/O error, dev sdh, sector 0

Somebody knows how resolve it?

Thanks,

Fernando Sierra.

 

[tom]

what is a raid-5 with multipath?

 

[nandex]

I have a proxmox cluster with 4 nodes. And the data storage is in another chasis. There are 4 disk in RAID-5. I use multipath for recognize this LVM.
Since starts proxmox
I read at bugzilla that its an errors with kernel like SATA disks https://bugzilla.redhat.com/show_bug.cgi?id=211948. Its a big problem because I need to put a logrotate and connect to shell via SSH. If I enter directly in the node I always see
end_request: I/O error, dev sdc, sector 0
end_request: I/O error, dev sdd, sector 0
end_request: I/O error, dev sdg, sector 0
end_request: I/O error, dev sdh, sector 0
end_request: I/O error, dev sdc, sector 0
end_request: I/O error, dev sdd, sector 0
end_request: I/O error, dev sdg, sector 0
end_request: I/O error, dev sdh, sector 0
....

like a waterfall

Thanks for your time!!

 

[tom]

you use iSCSI with multipath?

see http://pve.proxmox.com/wiki/ISCSI_Multipath

 

[nandex]

The problem is that is not scsi, is Fiber Channel with multipath. And I have to do an LVM group, and format the RAID-5 like LVM and only could create raw images

Regards!!,

Fernando Sierra

 

[nandex]

ok,

I think that I fixed this problem with your manual. I put the blacklist and blacklist_exceptions in multipath.conf like:

blacklist{
wwid *
}
blacklist_exceptions{
wwid 3690b11c0000f1fdc000004255198478e
wwid 3690b11c0000f1fcf0000041a3f051d95
}



The errors dissapear. But now i have another message that sais that INFO: task rgmanager: 3275 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.

I'll try to fix this .

Thanks for your help!!