[ https://issues.apache.org/jira/browse/AMBARI-11903?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Yusaku Sako updated AMBARI-11903: --------------------------------- Assignee: Andrew Onischuk > metrics-collector in distributed mode fails to stop > --------------------------------------------------- > > Key: AMBARI-11903 > URL: https://issues.apache.org/jira/browse/AMBARI-11903 > Project: Ambari > Issue Type: Bug > Components: ambari-metrics > Environment: sles11sp3 > ambari-server-2.1.0-1064 > 2.3.0.0-2346 > Reporter: Michael Harp > Assignee: Andrew Onischuk > Priority: Critical > Fix For: 2.1.0 > > > Stopping metrics-collector is slow and mostly fails when configured in > distributed mode. I believe the reason is the hbase stop order is the same as > start order which results in zookeeper stopping fist instead of last. > {code} > # ams_service.py > if params.is_hbase_distributed: > hbase_service('zookeeper', action=action) > hbase_service('master', action=action) > hbase_service('regionserver', action=action) > cmd = format("{cmd} --distributed") > {code} -- This message was sent by Atlassian JIRA (v6.3.4#6332)