Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
On Jun 3, 2009, at 6:52 AM, Heiko wrote: > > > On Wed, Jun 3, 2009 at 3:43 PM, Heiko <rupertt at gmail.com> wrote: > > > On Wed, Jun 3, 2009 at 3:34 PM, Tony Lownds <tony at pagedna.com> wrote: > > On Jun 2, 2009, at 10:59 PM, Heiko wrote: > > Hello, > > i am investigating why our server pairs reboot themselves from time > to time. > This is very annoing because these machines are in production and i > always > have to fix mysql replications or drbd splitbrains after these > reboots. > > We have 3 pairs that use a drbd/xen/heartbeat setup and 2 of these > pairs crash, > sometimes every 2 week sometimes only twice a year. > > I first thought it could be heartbeat, but I stopped the service on > 1 pair and we also had a crash. > Are there other people who had these kind of crashes? > > Yes, sounds familiar... switching to protocol "A" is a workaround > that helped for me. > > http://article.gmane.org/gmane.comp.linux.drbd/17684 > Hello Tony, > > you are giving me hope, but I dont understand that post you linked. > I now have set to Protocoll to A. > Can I just run drbdadm adjust "resource" without getting any trouble/ > data los? I think you'll have to try it and see. I personally brought down xen, then restarted drbd on both sides, then brought back xen during a downtime window. That worked well for me. You can see the current protocol in /proc/drbd near the end of the first line for each resource. # cat /proc/drbd | grep -w A 0: cs:Connected st:Secondary/Primary ds:UpToDate/UpToDate A r--- 1: cs:Connected st:Secondary/Primary ds:UpToDate/UpToDate A r--- 2: cs:Connected st:Secondary/Primary ds:UpToDate/UpToDate A r--- 3: cs:Connected st:Secondary/Primary ds:UpToDate/UpToDate A r--- -Tony -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20090603/9b7b971a/attachment.htm>