[infinispan-issues] [JBoss JIRA] (ISPN-5242) Minor inconsistencies between local-cache and clustered-cache statistics

[Vladimir Blagojevic (JIRA)]

    [ https://issues.jboss.org/browse/ISPN-5242?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13055104#comment-13055104 ] 

Vladimir Blagojevic edited comment on ISPN-5242 at 3/31/15 4:05 PM:
--------------------------------------------------------------------

Martin,

You are right about 1). We could add timeSinceReset to cluster cache stats i.e org.infinispan.stats.Stats.java API. elapsedTime does not make sense in the context of cache stats as it is related to the elapsed time since cache has started. I'll make a PR with this change and will see what others say. 

Regarding 2) these are here by design requirements. See org.infinispan.stats.ClusterCacheStats.java 


was (Author: vblagojevic):
Martin,

You are right about 1). We could add timeSinceReset to cluster cache stats i.e org.infinispan.stats.Stats.java API. elapsedTime does not make sense in the context of cache stats as it is related to the time since cache has started. I'll make a PR with this change and will see what others say. 

Regarding 2) these are here by design requirements. See org.infinispan.stats.ClusterCacheStats.java 

> Minor inconsistencies between local-cache and clustered-cache statistics
> ------------------------------------------------------------------------
>
>                 Key: ISPN-5242
>                 URL: https://issues.jboss.org/browse/ISPN-5242
>             Project: Infinispan
>          Issue Type: Bug
>    Affects Versions: 7.1.0.Final
>            Reporter: Martin Gencur
>            Priority: Minor
>
> There are some inconsistencies among MBeans for local and clustered cache:
> 1) 
> * local-cache stats: timeSinceReset, elapsedTime
> * cluster-cache stats name: timeSinceStart (NO timeSinceReset or elapsedTime stats here)
> 2) cluster cache now has some statistics that are not available for local cache: CacheLoaderLoads, CacheLoaderMisses, etc.  I'm not sure but maybe it would be better to have this consistent.



--
This message was sent by Atlassian JIRA
(v6.3.11#6341)