[DRBD-user] filesystem corruptions

Poyner, Brandon bpoyner at ccac.edu
Fri Sep 30 14:53:20 CEST 2005

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


> we'll ive spent all night building a new system with same 
> configuration.
> and yes, ive run fsck after upgrading kernel, it said that 
> everything is 
> in perfect condition.
> so for now i have :
> Linux serv1 2.6.13.2 #1 SMP
> DRBD 0.7.13
> 
> and STILL i have those damn errors, just now(bout 10mins ago) primary 
> node crashed with kernel panic, last on screen there was:
> 
> ata3: status0xa0 { Busy }
> SCSI error: <2 0 0 0> return code = 0x800082
> sdc: Current: sense key: Aborted Command
>       Additional sense: SCSI parity error
> end_request: I/O error, dev sdc, sector 325582919
> drbd1: IO error on backing device!
> kernel panic - not syncing: drbd1: IO error on backing device!
> 
> now this is getting realy annoying, any ideas what could i do 
> about this 
> problem?

DRBD is doing what you asked it to do.  The problem isn't with DRBD.

   disk {
     on-io-error   panic;
   }

I would start by checking the underlying device, /dev/sdc, for hardware
problems.  My favorite tool for doing that from a running Linux system
is smartmontools (http://smartmontools.sourceforge.net), but you also
have tools like badblocks and most hard drive manufacturers make an
offline test utility.  If this is a true scsi disk I would make sure the
cable connections are tight and that you have plugged the drive into the
correct socket (LVD to LVD, etc).

Brandon Poyner
Network Engineer III
CCAC - College Office
412-237-3086




More information about the drbd-user mailing list