Hi,
I've configured two node DRBD9 cluster, all looks good. but I've faced with a strange behavior - when I start both nodes, cluster get into split brain state (looks like). And it can being in that state for some time (5,10,30 minutes) then ocasionaly became ok.
right after nodes start I see:
[root@node2 tmp]# drbdmanage list-nodes
+------------------------------------------------------------------------------------------------------------+
| Name | Pool Size | Pool Free | | State |
|------------------------------------------------------------------------------------------------------------|
| node1 | 3068 | 8 | | OFFLINE |
| node2 | 3068 | 8 | | ok |
+------------------------------------------------------------------------------------------------------------+
+------------------------------------------------------------------------------------------------------------+
| Name | Pool Size | Pool Free | | State |
|------------------------------------------------------------------------------------------------------------|
| node1 | 3068 | 8 | | OFFLINE |
| node2 | 3068 | 8 | | ok |
+------------------------------------------------------------------------------------------------------------+
[root@node1 tmp]# drbdmanage list-nodes
+------------------------------------------------------------------------------------------------------------+
| Name | Pool Size | Pool Free | | State |
|------------------------------------------------------------------------------------------------------------|
| node1 | 3068 | 8 | | ok |
| node2 | 3068 | 8 | | OFFLINE |
+------------------------------------------------------------------------------------------------------------+
+------------------------------------------------------------------------------------------------------------+
| Name | Pool Size | Pool Free | | State |
|------------------------------------------------------------------------------------------------------------|
| node1 | 3068 | 8 | | ok |
| node2 | 3068 | 8 | | OFFLINE |
+------------------------------------------------------------------------------------------------------------+
in good state:
[root@node2 tmp]# drbdmanage list-nodes
+------------------------------------------------------------------------------------------------------------+
| Name | Pool Size | Pool Free | | State |
|------------------------------------------------------------------------------------------------------------|
| node1 | 3068 | 8 | | ok |
| node2 | 3068 | 8 | | ok |
+------------------------------------------------------------------------------------------------------------+
+------------------------------------------------------------------------------------------------------------+
| Name | Pool Size | Pool Free | | State |
|------------------------------------------------------------------------------------------------------------|
| node1 | 3068 | 8 | | ok |
| node2 | 3068 | 8 | | ok |
+------------------------------------------------------------------------------------------------------------+
may anybody suggest what to do to make cluster behavior more determined?
I would appriciate for any suggestions.
Thank you.
OS: RHEL7
DRBD 9.0.1-1
drbd-utils 8.9.6
drbdmanage 0.94
_______________________________________________ drbd-user mailing list [email protected] http://lists.linbit.com/mailman/listinfo/drbd-user
