Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
/ 2006-02-16 11:04:57 +0100 \ Andrés Cañada: > El Jueves, 16 de Febrero de 2006 10:55, Andrés Cañada escribió: > > Hi! > > Please, I need some help with this. Yesterday again the master director > > failover resources to backup director. Great (althougth it is crashing too > > many times) > > But now, when I want to come back to the old configuration I have some > > problems. > > I had to reboot primary director because it was frozen. > > So when heartbeat started again and with failback off, the backup director > > (now primary) shows in /proc/drbd: > > version: 0.7.15 (api:77/proto:74) > > SVN Revision: 2020 build by root at prot15, 2006-01-24 16:03:36 > > 0: cs:StandAlone st:Primary/Unknown ld:Consistent > > ns:0 nr:0 dw:4821556 dr:80335 al:188 bm:700 lo:0 pe:0 ua:0 ap:0 > > 1: cs:Unconfigured > > And the old primary director, now secondary, shows: > > version: 0.7.15 (api:77/proto:74) > > SVN Revision: 2020 build by root at prot14, 2006-01-24 16:00:25 > > 0: cs:WFConnection st:Secondary/Unknown ld:Consistent > > ns:0 nr:0 dw:656 dr:55713 al:3 bm:260 lo:0 pe:0 ua:0 ap:0 > > 1: cs:Unconfigured > > > > I'd like to have them in Primary/Secondary and Secondary/Primary status but > > I can't go back to that situation. > > Do I have to type #drbdadm syncer all (what I think it should synchronize > > the disks) to get them synchronized?? I can't with this command. > Feb 16 10:59:21 localhost kernel: drbd0: Connection established. > Feb 16 10:59:21 localhost kernel: drbd0: I am(P): 1:00000002:00000001:00000014:00000005:10 > Feb 16 10:59:21 localhost kernel: drbd0: Peer(S): 1:00000003:00000001:00000014:00000003:00 I think here is one log line missing about "current primary shall become sync target. aborting to prevent data corruption"... I can not tell how you managed to get into this situation. but. you have to decide which one has the bad data, which has the good data. !! keep the box with the GOOD data in drbd primary mode !! go on the box with the BAD data, then "drbdadm invalidate" it. after that, you should be able to "drbdadm connect" them (probably has to be issued on both nodes). you will see a full sync. there are ways to avoid a full sync, but they are more involved. I'd go for the full sync. hope that helps. -- : Lars Ellenberg Tel +43-1-8178292-0 : : LINBIT Information Technologies GmbH Fax +43-1-8178292-82 : : Schoenbrunner Str. 244, A-1120 Vienna/Europe http://www.linbit.com : __ please use the "List-Reply" function of your email client.