Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Plug it in, give it a /30 network on a unused subnet. Shutdown drbd on both sides, change the IPs in the config, start drbd back up - It should auto resync anything and become uptodate/uptodate. You probably would also want to configure ocfs2 to use the same interface for its DLM. On 2/17/12 7:32 AM, Lawrence Strydom wrote: > I might be able to add NIC's -the machines are geographically quite > far from me so will have to get some one to check first but say there > is NIC space, will DRBD behave if I change the replicating IP > addresses? Can you reccomend a procedure to follow in this case? > > > > > On 17 February 2012 14:28, David Coulson <david at davidcoulson.net > <mailto:david at davidcoulson.net>> wrote: > > Can you add NICs? There should be no requirement for DRBD > replication to happen over any particular interface, and obviously > having it's own dedicated interfaces without a switch is much > nicer than what you have now. > > DRBD doesn't support multiple networks, as it uses a constant TCP > Connection for updates - You could probably do it with some > routing protocol, but that'd be really ugly. > > > > > On 2/17/12 7:21 AM, Lawrence Strydom wrote: >> Hi Dave, >> >> Thanks for all the help - The default log rotation time for >> ubuntu seems to be 7 days so it seems this will remain a mystery. >> >> A dedicated cross over for replication is not an option because >> the web servers connect to the DB over the same link. Is it >> possible to add the public interface as a second replication >> interface? Do I simply add it in the .res file with the existing >> ip address? >> >> The setup was primary/primary uptodate/uptodate before I copied >> the sites across. It then sat for a couple weeks doing nothing >> untill this week when dev tested on it and this hapenned. >> >> L >> >> >> >> On 17 February 2012 14:01, David Coulson <david at davidcoulson.net >> <mailto:david at davidcoulson.net>> wrote: >> >> >> >> On 2/17/12 6:56 AM, Lawrence Strydom wrote: >>> Couldn't find any PingAck's in the syslog - what else could >>> have caused the split brain? >> >> ./syslog.2:Feb 15 13:40:35 web02 kernel: [ 16.455206] block drbd0: self D3CCDACF6FD7FDB8:4579E80074D400D3:C117CFF0A5777F0F:0000000000000004 bits:14607528 flags:0 >> ./syslog.2:Feb 15 13:40:35 web02 kernel: [ 16.455211] block drbd0: peer 37C841BC2AA49AC4:4579E80074D400D3:C117CFF0A5777F0F:0000000000000004 bits:1407177 flags:0 >> ./syslog.2:Feb 15 13:40:35 web02 kernel: [ 16.455215] block drbd0: uuid_compare()=100 by rule 90 >> ./syslog.2:Feb 15 13:40:35 web02 kernel: [ 16.455217] block drbd0: Split-Brain detected, dropping connection! >> >> Was this ever really primary/primary, uptodate/uptodate? I can't find anything in the logs indicating that it was ever truly dual primary. Your logs seem to rotate quickly, so if you are really only keeping logs going back three days you'll probably never know root cause. >> >> Personally, I'd bypass the switch and go with a cross-connect direct between the boxes - Or two if you have enough NICs. >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20120217/4b5491da/attachment.htm>