[DRBD-user] FAQ: Reconnecting after a temporary primary node failure

Mauricio Tavares raubvogel at gmail.com
Tue Mar 8 13:33:58 CET 2011

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


On Tue, Mar 8, 2011 at 7:13 AM, Felix Frank <ff at mpexnet.de> wrote:
>> root at db1:~#
>> root at db1:~# cat /proc/drbd
>> version: 8.3.7 (api:88/proto:86-91)
>> GIT-hash: ea9e28dbff98e331a62bcbcc63a6135808fe2917 build by root at db1,
>> 2011-03-07 15:01:39
>>  0: cs:StandAlone ro:Primary/Unknown ds:UpToDate/DUnknown   r----
>>     ns:0 nr:0 dw:240977 dr:37746 al:36 bm:20 lo:0 pe:0 ua:0 ap:0 ep:1
>> wo:b oos:18588
>
> For some reason, it won't connect. Check the logs during your attempted
> connect to find out why that is.
>
      Which logs should I check besides the kernel ones I included in
the first post (for both db1 and db2)? The db1 one has not changed,
but since I've done drbdadm connect r0 in db1, db2's kernel log
(included below on its entirety from the moment I told db1 to
reconnect) is complaining about a split-brain

Mar  8 06:33:15 db2 kernel: [77457.601739] block drbd0: Handshake
successful: Agreed network protocol version 91
Mar  8 06:33:15 db2 kernel: [77457.601757] block drbd0: conn(
WFConnection -> WFReportParams )
Mar  8 06:33:15 db2 kernel: [77457.601791] block drbd0: Starting
asender thread (from drbd0_receiver [15294])
Mar  8 06:33:15 db2 kernel: [77457.604332] block drbd0: data-integrity-alg: md5
Mar  8 06:33:15 db2 kernel: [77457.604493] block drbd0: drbd_sync_handshake:
Mar  8 06:33:15 db2 kernel: [77457.604502] block drbd0: self
952185A0C76CE6A6:201C5D8F032E15FA:BE4CF69DB972BF67:5445D15D92EAA8D9
bits:0 flags:0
Mar  8 06:33:15 db2 kernel: [77457.604512] block drbd0: peer
AD47F2C0FCB0CE1B:201C5D8F032E15FB:BE4CF69DB972BF66:5445D15D92EAA8D9
bits:4647 flags:0
Mar  8 06:33:15 db2 kernel: [77457.604521] block drbd0:
uuid_compare()=100 by rule 90
Mar  8 06:33:15 db2 kernel: [77457.604526] block drbd0: Split-Brain
detected, dropping connection!
Mar  8 06:33:15 db2 kernel: [77457.679298] block drbd0: helper
command: /sbin/drbdadm split-brain minor-0
Mar  8 06:33:15 db2 kernel: [77457.679590] block drbd0: meta
connection shut down by peer.
Mar  8 06:33:15 db2 kernel: [77457.748585] block drbd0: conn(
WFReportParams -> NetworkFailure )
Mar  8 06:33:15 db2 kernel: [77457.748598] block drbd0: asender terminated
Mar  8 06:33:15 db2 kernel: [77457.748600] block drbd0: Terminating
asender thread
Mar  8 06:33:15 db2 kernel: [77457.751299] block drbd0: helper
command: /sbin/drbdadm split-brain minor-0 exit code 0 (0x0)
Mar  8 06:33:15 db2 kernel: [77457.751305] block drbd0: conn(
NetworkFailure -> Disconnecting )
Mar  8 06:33:15 db2 kernel: [77457.751311] block drbd0: error
receiving ReportState, l: 4!
Mar  8 06:33:15 db2 kernel: [77457.822425] block drbd0: Connection closed
Mar  8 06:33:15 db2 kernel: [77457.822434] block drbd0: conn(
Disconnecting -> StandAlone )
Mar  8 06:33:15 db2 kernel: [77457.822523] block drbd0: receiver terminated
Mar  8 06:33:15 db2 kernel: [77457.822525] block drbd0: Terminating
receiver thread

 even though it never became a primary node,

root at db2:~# drbdadm role r0
Secondary/Unknown
root at db2:~#

> HTH,
> Felix
>



More information about the drbd-user mailing list