Hi,
the logs indicate that Appia detects some failure, separates the
group and rejoins it again.
The failure could be caused by some process that makes the machine
very heavy and the "alive" messages are not exchanged. But the issue
is that when Appia decides to rejoin the two members, Sequoia ignores
that and I could not understand why. Maybe Emmanuel (or someone from
sequoia) could help us with that?
cheers,
nuno
On Feb 22, 2007, at 12:11 , Schnabl, Sebastian wrote:
Does this happen during the dump?
Yes, it occurs during dump. 2 controllers (Sequoia 3.0.beta3).
- start controller A
- start controller B
- initialize backend of controller B
- make a dump of backend of controller B
--> connection is lost between both controllers
Can you turn on the DEBUG mode on hedera channel and gms and post it
(also) >to the Appia mailing list?
See attached logs.
Cheers Sebastian
<controllerA_full_cluster.log>
<ControllerB_full_cluster.log>
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
--
Nuno Carvalho
University of Lisbon, Portugal
http://dialnp.di.fc.ul.pt
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
