I am using heartbeat together with ddrbd v8 and dopd. Now I ran into problems while testing recovery scenarios. lets say there is a cluster of two nodes with a niúmber of DRBD-devices. one node is outdated due to networ failure in the ddrbd-link, the other node is completely lost (fire, theft, whatever). there is no time to get a replacement-machine before restarting everything, stale data is acceptable. How do I access the data on the DRBD-devices? I get
m:res cs st ds p mounted fstype 0:r0 WFConnection Secondary/Unknown Consistent/DUnknown C 1:r1 WFConnection Secondary/Unknown Consistent/DUnknown C 2:r2 WFConnection Secondary/Unknown Consistent/DUnknown C 3:r3 WFConnection Secondary/Unknown Consistent/DUnknown C 4:r4 WFConnection Secondary/Unknown Consistent/DUnknown C 5:r5 WFConnection Secondary/Unknown Consistent/DUnknown C 6:r6 WFConnection Secondary/Unknown Consistent/DUnknown C 7:r7 WFConnection Secondary/Unknown Consistent/DUnknown C 8:r8 WFConnection Secondary/Unknown Consistent/DUnknown C 9:r9 WFConnection Secondary/Unknown Consistent/DUnknown C 10:r10 WFConnection Secondary/Unknown Consistent/DUnknown C 11:r11 WFConnection Secondary/Unknown Consistent/DUnknown C 12:r12 WFConnection Secondary/Unknown Consistent/DUnknown C 13:r13 WFConnection Secondary/Unknown Consistent/DUnknown C 14:r14 WFConnection Secondary/Unknown Consistent/DUnknown C 15:r15 WFConnection Secondary/Unknown Consistent/DUnknown C 16:r16 WFConnection Secondary/Unknown Consistent/DUnknown C 17:r17 WFConnection Secondary/Unknown Consistent/DUnknown C c3100kc2:~ # drbdadm primary all State change failed: (-2) Refusing to be Primary without at least one UpToDate disk Command 'drbdsetup /dev/drbd0 primary' terminated with exit code 11 drbdsetup exited with code 11 c3100kc2:~ # drbdadm -- --overwrite-data-of-peer primary all State change failed: (-2) Refusing to be Primary without at least one UpToDate disk Command 'drbdsetup /dev/drbd0 primary --overwrite-data-of-peer' terminated with exit code 11 drbdsetup exited with code 11 If there is connection to the other node it works fine, but we need to take care of scenarios where there is none. Robert Köppl _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
