Description of problem: Monitoring JDG 6.5.1 using JBoss ON 3.3.3 (with the latest version of JDG plugin for JBoss ON (Data Grid Management Plug-in Pack Update-03 for JBoss ON 3.3 [1]) fails and WARN messages like the following are logged in the agent.log file: ************************************************************* 2015-10-28 10:28:31,197 WARN [ResourceContainer.invoker.daemon-9] (org.infinispan.server.rhq.IspnCacheComponent)- Getting metric [hitRatio] at [ Address{path: subsystem=infinispan,cache-container=local,local-cache=default}] failed: JBAS014792: Unknown attribute hit-ratio, rolled-back=true ************************************************************* Here is complete list of metrics whose collection fails with the same WARN message as above: hitRatio Unknown attribute hit-ratio, rolled-back=true readWriteRatio Unknown attribute read-write-ratio, rolled-back=true replicationCount Unknown attribute replication-count, rolled-back=true timeSinceReset Unknown attribute time-since-reset, rolled-back=true commits Unknown attribute commits, rolled-back=true rollbacks Unknown attribute rollbacks, rolled-back=true averageReplicationTime Unknown attribute average-replication-time, rolled-back=true successRatio Unknown attribute success-ratio, rolled-back=true removeHits Unknown attribute remove-hits, rolled-back=true evictions Unknown attribute evictions, rolled-back=true removeMisses Unknown attribute remove-misses, rolled-back=true prepares Unknown attribute prepares, rolled-back=true averageReadTime Unknown attribute average-read-time, rolled-back=true elapsedTime Unknown attribute elapsed-time, rolled-back=true hits Unknown attribute hits, rolled-back=true replicationFailures Unknown attribute replication-failures, rolled-back=true misses Unknown attribute misses, rolled-back=true stores Unknown attribute stores, rolled-back=true numberOfEntries Unknown attribute number-of-entries, rolled-back=true averageWriteTime Unknown attribute average-write-time, rolled-back=true averageRemoveTime Unknown attribute average-remove-time, rolled-back=true If you enable [CasheStore]Number of cache store stores for Cache resource in JBoss ON - the following is logged: **************************************************************** 2015-10-28 10:28:31,525 WARN [ResourceContainer.invoker.daemon-9] (org.infinispan.server.rhq.IspnCacheComponent)- Getting metric [cacheLoaderStores] at [ Address{path: subsystem=infinispan,cache-container=local,local-cache=default}] failed: JBAS014792: Unknown attribute cache-loader-stores, rolled-back=true **************************************************************** Above message previously was logged in https://bugzilla.redhat.com/show_bug.cgi?id=1168255. According to this Bugzilla above is fixed in JDG 6.4.1. However, JDG 6.5.1 still has the same problem. Version-Release number of selected component (if applicable): JDG 6.5.1 JBoss ON 3.3.3 Data Grid Management Plug-in Pack Update-03 for JBoss ON 3.3 [1] How reproducible: Always Steps to Reproduce: 1. JDG started invoking $JDG-HOME/bin/standalone.sh script 2. Navigate to JBoss ON UI -> JDG resource -> infinispan -> Cache Containers -> local -> Caches -> default -> Monitoring -> Metrics 3. Confirm that agent.log file contains WARN messages like above. 4. Navigate to any other tab for this resource (default - Cache) and the same set of WARN messages will be logged. 5. Navigate to Monitoring -> Schedules and enable [CacheStore]Number of cache store stores metrics; 6. Navigate to Monitoring -> Metrics; 7. Confirm that agent.log file contains the same set of WARN messages plus message for cahceLoaderStores. Actual results: WARN messages are logged in the agent.log file and metrics in JBoss ON UI show NaN values. Expected results: Agent.log file does not contain WARN messages and metrics are properly collected and shown in JBoss ON UI. Additional info: [1] https://access.redhat.com/jbossnetwork/restricted/softwareDownload.html?softwareId=39403
The issue here is that we really only test with shared-state caches (replicated/distributed) in client/server mode and that's why this bug went unnoticed.
Sorry Tristan but I have to correct you. This bug was reported a long time ago: https://bugzilla.redhat.com/show_bug.cgi?id=1181252 So BZ1276000 is actually a duplicate of 1181252. It was decided not to fix this in JDG 6.4 and was re-scheduled to JDG 6.6. See comments in BZ1181252.
This is fixed in *** This bug has been marked as a duplicate of bug 1276035 ***