Hi everybody,
I'm running drbd 0.6.1-pre7 with 3 drbd devices syncing over one crossover
ethernet interface and I have experienced one rather strange problem.
The whole cluster works just fine, SyncAll, SyncQuick, everything.
Just _sometimes_ one device (drbd1) loses it's connection. As far as I
understand this should be no problem and drbd should reconnect as
specified by "connect-int" in the config file.
When starting heartbeat, I sometimes get:
Nov 27 11:00:36 server-194 heartbeat: info: Running /etc/ha.d/resource.d/datadisk drbd1 start
Nov 27 11:00:36 server-194 kernel: drbd1: Connection lost.(pc=0,uc=0)
Now the state of drbd1 on the first node is primary/unknown and the state on
the second node is "Standalone/unknown".
My drbd.conf values are like this:
net {
sync-rate=5000
#skip-sync
tl-size=256
timeout=60
connect-int=10
ping-int=10
}
and thats the same for all 3 drbd devices.
Anything I can tune in the config file to make the reconnect work?
Cheers,
Kai Groshert
I'm running drbd 0.6.1-pre7 with 3 drbd devices syncing over one crossover
ethernet interface and I have experienced one rather strange problem.
The whole cluster works just fine, SyncAll, SyncQuick, everything.
Just _sometimes_ one device (drbd1) loses it's connection. As far as I
understand this should be no problem and drbd should reconnect as
specified by "connect-int" in the config file.
When starting heartbeat, I sometimes get:
Nov 27 11:00:36 server-194 heartbeat: info: Running /etc/ha.d/resource.d/datadisk drbd1 start
Nov 27 11:00:36 server-194 kernel: drbd1: Connection lost.(pc=0,uc=0)
Now the state of drbd1 on the first node is primary/unknown and the state on
the second node is "Standalone/unknown".
My drbd.conf values are like this:
net {
sync-rate=5000
#skip-sync
tl-size=256
timeout=60
connect-int=10
ping-int=10
}
and thats the same for all 3 drbd devices.
Anything I can tune in the config file to make the reconnect work?
Cheers,
Kai Groshert