[
https://issues.apache.org/jira/browse/HADOOP-10090?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13820421#comment-13820421
]
Luke Lu commented on HADOOP-10090:
----------------------------------
I was aware of the suboptimal behavior and hoped it'd be OK for metrics, which
don't require strong consistency.
#1 incurs unnecessary overhead (updating jmx cache) for people who don't use
JMX. This is the reason of the current cache logic.
#2 is risky, as we don't know all existing jmx query patterns (especially due
to HDFS-5333). User (admins) actually already has a choice to use a small JMX
cache TTL for refreshness.
How about #3: we only initialize and update the JMX cache when JMX is first
used and stops updating after a period inactivity and reinitialize and update
JMX cache upon activity. Initialize/reinitialize is a "dense" update, while
"update" means the "sparse" update with the current lastRecs mechanisms. I
think #3 is should be a fairly straightforward patch and more flexible than #1
and #2.
> Jobtracker metrics not updated properly after execution of a mapreduce job
> --------------------------------------------------------------------------
>
> Key: HADOOP-10090
> URL: https://issues.apache.org/jira/browse/HADOOP-10090
> Project: Hadoop Common
> Issue Type: Bug
> Components: metrics
> Affects Versions: 1.2.1
> Reporter: Ivan Mitic
> Assignee: Ivan Mitic
> Attachments: OneBoxRepro.png
>
>
> After executing a wordcount mapreduce sample job, jobtracker metrics are not
> updated properly. Often times the response from the jobtracker has higher
> number of job_completed than job_submitted (for example 8 jobs completed and
> 7 jobs submitted).
> Issue reported by Toma Paunovic.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
