[DRBD-user] "syncer" crash when doing full resync

Philipp Reisner philipp.reisner at linbit.com
Fri Jul 2 17:11:42 CEST 2004

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


[...OOPS deleted...]
> ohwell :(

After finding time to analyze this problem it is clear that the 
peer sent an invalid response-cookie, which caused this node to crash.

We understand what happened there:

drbdd
 -> receive_block_ack
     -> drbd_end_req
         -> struct Drbd_Conf* mdev = drbd_conf + MINOR(req->bh->b_rdev);
                                                    ^^^^^^^^^^
The crash happens while dereferencing req->bh

The register holding req (%esi) has 0xffffffff .

This req comes from the peer, it got it with the request to write a block.

Either your peer is corrupting its memory (more probabely), or 
the node, on which the OOPS happened is corrupting its memory (less
probabel)

Please go and get new hardware.

[In case memtest86 does not find anything, it could be that the 
 IO operation to disk does the memory corruption...]

-philipp
-- 
: Dipl-Ing Philipp Reisner                      Tel +43-1-8178292-50 :
: LINBIT Information Technologies GmbH          Fax +43-1-8178292-82 :
: Schönbrunnerstr 244, 1120 Vienna, Austria    http://www.linbit.com :



More information about the drbd-user mailing list