[
https://issues.apache.org/jira/browse/DERBY-3364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12568934#action_12568934
]
V.Narayanan commented on DERBY-3364:
------------------------------------
Now the stopSlave on a slave when the connection is active will fail saying that
the connection with the master is still active. I was earlier using this
particular
situation as the reproducible
1) Start Slave
2) start master
3) execute a few transactions on the master (a few inserts)
4) Issue stopSlave on the slave side
5) failover on the master
This hanged for me. But after stopSlave on a active connection being disabled
I am running short of ideas to reproduce this.
With the earlier reproducible doing a setSoTimeout on the socket connection in
the
Transmitter worked for me.
I please request for help in generating a reproducible in the present trunk. I
can
submit a fix with the setSoTimeout if that is acceptable since logically the
read
on the InputStream obtained from the socket should time out.
> Replication failover implementation must be modified to fail at the master
> after slave has been stopped
> -------------------------------------------------------------------------------------------------------
>
> Key: DERBY-3364
> URL: https://issues.apache.org/jira/browse/DERBY-3364
> Project: Derby
> Issue Type: Bug
> Components: Replication
> Reporter: V.Narayanan
> Assignee: V.Narayanan
>
> Jorgen says...
> I tried to run the failover command on the master, which seems to work fine
> as long as the master and slave are still connected. If the slave has been
> stopped for some reason, however, failover hangs on
> MasterController#startFailover here:
> ack = transmitter.readMessage();
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
