[
https://issues.jboss.org/browse/JGRP-2265?page=com.atlassian.jira.plugin....
]
lokesh raheja updated JGRP-2265:
--------------------------------
Description:
Hi Bela
Jgroup 3.6.13
We have a cluster of 105 nodes.I see after the cluster split it merge the cluster every
time.
But sometimes it gets stuck and divided into two or three clusters .1 big cluster and 2
small clusters.
We have to restart the node in the small cluster to join the big cluster.
In the case when it doesn’t join the cluster, I noticed that the node which is not in big
cluster show the coordinator(using MERGE3.dumpviews) is the one which is already in a big
cluster.The strange part is if I shut down the coordinator of a small cluster, but on all
the nodes of the small cluster, it shows that same coordinator which is shut down.
Is some caching issue?Could you please help on the same
was:
Hi Bela
Jgroup 3.6.13
MERGE3 is not merging the cluster:
We have a cluster of 105 nodes.I see after the cluster split it merge the cluster every
time.
But sometimes it gets stuck and divided into two or three clusters .1 big cluster and 2
small clusters.
We have to restart the node in the small cluster to join the big cluster.
In the case when it doesn’t join the cluster, I noticed that the node which is not in big
cluster show the coordinator(using MERGE3.dumpviews) is the one which is already in a big
cluster.The strange part is if I shut down the coordinator of a small cluster, but on all
the nodes of the small cluster, it shows that same coordinator which is shut down.
Is some caching issue?Could you please help on the same
Jgroup MERGE3 is not merging big cluster
----------------------------------------
Key: JGRP-2265
URL:
https://issues.jboss.org/browse/JGRP-2265
Project: JGroups
Issue Type: Feature Request
Reporter: lokesh raheja
Assignee: Bela Ban
Hi Bela
Jgroup 3.6.13
We have a cluster of 105 nodes.I see after the cluster split it merge the cluster every
time.
But sometimes it gets stuck and divided into two or three clusters .1 big cluster and 2
small clusters.
We have to restart the node in the small cluster to join the big cluster.
In the case when it doesn’t join the cluster, I noticed that the node which is not in big
cluster show the coordinator(using MERGE3.dumpviews) is the one which is already in a big
cluster.The strange part is if I shut down the coordinator of a small cluster, but on all
the nodes of the small cluster, it shows that same coordinator which is shut down.
Is some caching issue?Could you please help on the same
--
This message was sent by Atlassian JIRA
(v7.5.0#75005)