Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Thu, 2009-01-29 at 18:18 +0100, Matteo Tescione wrote: > I had a similar problem with my drbd datastore, don’t know if it’s the > same but I remember that I switched off the “tickless” kernel > configuration option and it solved my random crashes. Mmmh, interesting, I might give that a try. I've also seen occasional hangs using DRBD 8.0.11 on two Ubuntu (Hardy) systems running kernel 2.6.24-22-server with Areca RAID controllers. The lockups I am seeing seem to be triggered by I/O load, and more often than not, they tear down _both_ nodes in quick succession, which kind of defeats the idea of redundancy :-/. I've never seen that happen on any other systems with Areca controllers (without DRBD). It is indeed quite likely that the DRBD happens to trigger a kernel bug because it just happens to stress disk and network concurrently. I just upgraded one node to DRBD 8.2.6 on a 2.6.27-11-server kernel (Intrepid), and will take care of the other node shortly. If the problem persists, I'll switch to non-tickless kernel to see if that makes a difference. Cheers, Steven