Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Dear Sirs,
I've installed drbd 8.4 and I use it with Linux 3.0.3. Recently I faced a
problem:
I create configured drbd resource 'prox' with two volumes. (prox/0 and
prox/1)
It starts and works well.
But each time I forget to run 'drbdadm secondary prox' before 'drbdadm down
prox'
I can't connect nodes until I do full resync with 'drbdadm invalidate prox'.
In my opinion, it is becaouse of a bug in tools. I found that when I run:
drbdadm connect --discard-my-data prox
on victim node, the action is applied only to the first volume! See dmesg:
[28298.548846] d-con prox: conn( StandAlone -> Unconnected )
[28298.548857] d-con prox: Starting receiver thread (from drbd_w_prox
[29820])
[28298.548894] d-con prox: receiver (re)started
[28298.548901] d-con prox: conn( Unconnected -> WFConnection )
[28299.048785] d-con prox: Handshake successful: Agreed network protocol
version 100
[28299.048806] d-con prox: conn( WFConnection -> WFReportParams )
[28299.048841] d-con prox: Starting asender thread (from drbd_r_prox
[30149])
[28299.057334] block drbd0: drbd_sync_handshake:
[28299.057340] block drbd0: self
7F4E079DAEAA9BC4:DA80358BFA522EC4:0001000000000004:0000000000000004 bits:0
flags:0
[28299.057344] block drbd0: peer
B3A9337BA1B046A6:DA80358BFA522EC4:0001000000000004:0000000000000004 bits:0
flags:0
[28299.057348] block drbd0: uuid_compare()=100 by rule 90
[28299.057352] block drbd0: helper command: /sbin/drbdadm
initial-split-brain minor-0
[28299.058183] block drbd0: helper command: /sbin/drbdadm
initial-split-brain minor-0 exit code 0 (0x0)
[28299.058192] block drbd0: Split-Brain detected, manually solved. Sync from
peer node
[28299.058195] block drbd0: peer( Unknown -> Secondary ) conn(
WFReportParams -> WFBitMapT ) pdsk( DUnknown -> UpToDate )
[28299.097332] block drbd1: drbd_sync_handshake:
[28299.097337] block drbd1: self
063044FA0D406C66:6FFD6C9E8733A59C:0001000000000004:0000000000000004 bits:0
flags:0
[28299.097342] block drbd1: peer
FB4496E6DCEFC142:6FFD6C9E8733A59C:0001000000000004:0000000000000004 bits:0
flags:0
[28299.097346] block drbd1: uuid_compare()=100 by rule 90
[28299.097350] block drbd1: helper command: /sbin/drbdadm
initial-split-brain minor-1
[28299.098171] block drbd1: helper command: /sbin/drbdadm
initial-split-brain minor-1 exit code 0 (0x0)
[28299.098179] block drbd1: Split-Brain detected but unresolved, dropping
connection!
[28299.098391] block drbd1: helper command: /sbin/drbdadm split-brain
minor-1
[28299.099117] block drbd1: helper command: /sbin/drbdadm split-brain
minor-1 exit code 0 (0x0)
[28299.099125] block drbd0: peer( Secondary -> Unknown ) conn( WFBitMapT ->
Disconnecting ) pdsk( UpToDate -> DUnknown )
[28299.099129] block drbd1: conn( WFReportParams -> Disconnecting )
[28299.099131] d-con prox: error receiving ReportState, e: -5 l: 0!
[28299.099136] d-con prox: asender terminated
[28299.099138] d-con prox: Terminating asender thread
[28299.108961] block drbd0: bitmap WRITE of 0 pages took 0 jiffies
[28299.109094] block drbd0: 0 KB (0 bits) marked out-of-sync by on disk
bit-map.
[28299.109163] d-con prox: Connection closed
[28299.109169] d-con prox: conn( Disconnecting -> StandAlone )
[28299.109171] d-con prox: receiver terminated
[28299.109172] d-con prox: Terminating receiver thread
The as your can see difference between volumes is:
drbd0: Split-Brain detected, manually solved. Sync from peer node
drbd1: Split-Brain detected but unresolved, dropping connection!
Could you advice me please what to do next?
Yours faithfully,
Albert Akchurin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.linbit.com/pipermail/drbd-user/attachments/20110830/4ceea85c/attachment.htm>