[
https://issues.apache.org/jira/browse/AMBARI-12741?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14694148#comment-14694148
]
Hudson commented on AMBARI-12741:
---------------------------------
FAILURE: Integrated in Ambari-trunk-Commit #3269 (See
[https://builds.apache.org/job/Ambari-trunk-Commit/3269/])
AMBARI-12741 - AMS Collector process takes way more Virtual memory than Xmx
setting (abaranchuk:
http://git-wip-us.apache.org/repos/asf?p=ambari.git&a=commit&h=9e962270bff7d61625617266eff9789c352b5be2)
*
ambari-server/src/main/resources/common-services/AMBARI_METRICS/0.1.0/configuration/ams-env.xml
*
ambari-metrics/ambari-metrics-timelineservice/conf/unix/ambari-metrics-collector
> AMS Collector process takes way more Virtual memory than Xmx setting
> --------------------------------------------------------------------
>
> Key: AMBARI-12741
> URL: https://issues.apache.org/jira/browse/AMBARI-12741
> Project: Ambari
> Issue Type: Task
> Components: ambari-metrics
> Affects Versions: 2.1.2
> Reporter: Artem Baranchuk
> Assignee: Artem Baranchuk
> Priority: Critical
> Fix For: 2.1.2
>
>
> Ambari Metrics Collector grabs almost whole free memory for VSZ just after
> start.
> {noformat}
> free -g
> total used free shared buffers cached
> Mem: 29 14 15 0 0 4
> {noformat}
> As you can see free mem is 15GB
> AMS Collector process:
> {noformat}
> ps -u -p 15921 | awk '{print $0}'
> USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
> ams 25954 4.9 7.0 13782216 2167184 ? Sl 14:00 1:26
> /usr/jdk64/jdk1.8.0_40/bin/java
> -Djava.library.path=/usr/lib/ams-hbase/lib/hadoop-native -Xmx512m -cp
> /usr/lib/ambari-metrics-collector/*:/etc/ambari-metrics-collector/conf
> -Djava.net.preferIPv4Stack=true
> -Dams.log.dir=/var/log/ambari-metrics-collector -Dproc_timelineserver
> org.apache.hadoop.yarn.server.applicationhistoryservice.ApplicationHistoryServer
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
