[infinispan-issues] [JBoss JIRA] (ISPN-10051) Cluster stats error after node leaves

[Dan Berindei (Jira)]

Dan Berindei created ISPN-10051:
-----------------------------------

             Summary: Cluster stats error after node leaves
                 Key: ISPN-10051
                 URL: https://issues.jboss.org/browse/ISPN-10051
             Project: Infinispan
          Issue Type: Bug
          Components: Core
    Affects Versions: 9.4.9.Final
            Reporter: Dan Berindei
             Fix For: 9.4.10.Final


Up to 9.4.x, {{ClusterCacheStatsImpl}} uses the distributed executor service to collect statistics from all the members of the cache. However, DES will fail with a {{SuspectException}} if one of the cache members is no longer in the cluster view, which is very common (a crashed node is always removed from the cluster view first and from the cache topology later):

{noformat}
23:40:57,029 ERROR [org.infinispan.stats.impl.ClusterCacheStatsImpl] (pool-1-thread-1) Could not execute cluster wide cache stats operation : java.util.concurrent.ExecutionException: 
org.infinispan.remoting.transport.jgroups.SuspectException: ISPN000400: Node rhdg73-server-11-9pl9h was suspected
{noformat}

In 10.0.x the distributed executor was removed and stats use the cluster executor, which only works with the cluster view.



--
This message was sent by Atlassian Jira
(v7.12.1#712002)