[
https://issues.apache.org/jira/browse/HBASE-2138?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12801287#action_12801287
]
Lars George commented on HBASE-2138:
------------------------------------
The above error you see seems to go deeper. It seems that if inc() is called
without any real numOps (on idle cluster?) there is no check for "0" and the
division must fail.
{code}
public synchronized void inc(final int numOps, final long time) {
currentData.numOperations += numOps;
currentData.time += time;
long timePerOps = time/numOps;
minMax.update(timePerOps);
}
{code}
> unknown metrics type
> --------------------
>
> Key: HBASE-2138
> URL: https://issues.apache.org/jira/browse/HBASE-2138
> Project: Hadoop HBase
> Issue Type: Bug
> Reporter: Jean-Daniel Cryans
> Assignee: Lars George
> Fix For: 0.20.3, 0.21.0
>
>
> Since the recent metric commits I see this on the master and RS at boot:
> {code}
> 2010-01-16 11:24:59,730 INFO org.apache.hadoop.metrics.jvm.JvmMetrics:
> Initializing JVM Metrics with processName=RegionServer,
> sessionId=regionserver/10.10.1.49:60020
> 2010-01-16 11:24:59,732 ERROR org.apache.hadoop.metrics.MetricsUtil: unknown
> metrics type: org.apache.hadoop.hbase.metrics.MetricsRate
> 2010-01-16 11:24:59,732 ERROR org.apache.hadoop.hbase.metrics: unknown
> metrics instance: org.apache.hadoop.metrics.util.MetricsTimeVaryingRate
> 2010-01-16 11:24:59,732 ERROR org.apache.hadoop.hbase.metrics: unknown
> metrics instance: org.apache.hadoop.metrics.util.MetricsTimeVaryingRate
> 2010-01-16 11:24:59,732 ERROR org.apache.hadoop.hbase.metrics: unknown
> metrics instance: org.apache.hadoop.metrics.util.MetricsTimeVaryingRate
> 2010-01-16 11:24:59,732 ERROR org.apache.hadoop.hbase.metrics: unknown
> metrics instance: org.apache.hadoop.metrics.util.MetricsTimeVaryingRate
> {code}
> We need to clean that for 0.20.3
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
