[DRBD-user] What causes nodes to become out-of-sync?

Jeffrey Froman tcijf at olympus.net
Thu Jul 24 22:28:09 CEST 2008

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


On Thursday 24 July 2008 06:16:30 am Lars Ellenberg wrote:
> whenever drbd "data-integrity-alg" comparison fails,
> drbd disconnects, reconnects, and does a resync.
>
> if you get that frequently, you basically have a degraded cluster.

Aha, I see.

> if you turn tcp checksum offloading off, thus have the tcp checksum
> cover in-kernel-memory to in-kernel-memory, any transfer errors in
> between (catched by the tcp checksum, which may be too weak to
> catch all of them -- but then you have a different problem anyways)
> cause a tcp resent, which is completely transparent to drbd.

This sounds like a good first step. I've disabled checksum offloading 
on the interface. If online verification continues to find mismatched 
blocks with any regularity, I'll add data-integrity-alg as well.


Thank you,
Jeffrey



More information about the drbd-user mailing list