[jira] [Updated] (HDFS-9655) NN should start JVM pause monitor before loading fsimage
[ https://issues.apache.org/jira/browse/HDFS-9655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yongjun Zhang updated HDFS-9655: Labels: supportability (was: ) > NN should start JVM pause monitor before loading fsimage > > > Key: HDFS-9655 > URL: https://issues.apache.org/jira/browse/HDFS-9655 > Project: Hadoop HDFS > Issue Type: Bug >Reporter: John Zhuge >Assignee: John Zhuge >Priority: Critical > Labels: supportability > Fix For: 2.8.0 > > Attachments: HDFS-9655.001.patch > > > We have seen many cases of NameNode startup either extremely slow or even > hung. Most of them were caused by insufficient heap size with regard to the > metadata size. Those cases were resolved by increasing the heap size. > However it did take support team some time to root cause. JVM pause warning > messages would greatly assist in such diagnosis, but NN starts JVM pause > monitor after fsimage/edits loading. > Propose to start JVM pause monitor before loading fsimage/edits. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HDFS-9655) NN should start JVM pause monitor before loading fsimage
[ https://issues.apache.org/jira/browse/HDFS-9655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lei (Eddy) Xu updated HDFS-9655: Resolution: Fixed Fix Version/s: 2.8.0 Status: Resolved (was: Patch Available) +1. The test failures seems not relevant to me. Thanks a lot for working on this, [~jzhuge]! Committed to trunk, branch-2 and branch-2.8. > NN should start JVM pause monitor before loading fsimage > > > Key: HDFS-9655 > URL: https://issues.apache.org/jira/browse/HDFS-9655 > Project: Hadoop HDFS > Issue Type: Bug >Reporter: John Zhuge >Assignee: John Zhuge >Priority: Critical > Fix For: 2.8.0 > > Attachments: HDFS-9655.001.patch > > > We have seen many cases of NameNode startup either extremely slow or even > hung. Most of them were caused by insufficient heap size with regard to the > metadata size. Those cases were resolved by increasing the heap size. > However it did take support team some time to root cause. JVM pause warning > messages would greatly assist in such diagnosis, but NN starts JVM pause > monitor after fsimage/edits loading. > Propose to start JVM pause monitor before loading fsimage/edits. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HDFS-9655) NN should start JVM pause monitor before loading fsimage
[ https://issues.apache.org/jira/browse/HDFS-9655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] John Zhuge updated HDFS-9655: - Status: Patch Available (was: Open) > NN should start JVM pause monitor before loading fsimage > > > Key: HDFS-9655 > URL: https://issues.apache.org/jira/browse/HDFS-9655 > Project: Hadoop HDFS > Issue Type: Bug >Reporter: John Zhuge >Assignee: John Zhuge >Priority: Critical > Attachments: HDFS-9655.001.patch > > > We have seen many cases of NameNode startup either extremely slow or even > hung. Most of them were caused by insufficient heap size with regard to the > metadata size. Those cases were resolved by increasing the heap size. > However it did take support team some time to root cause. JVM pause warning > messages would greatly assist in such diagnosis, but NN starts JVM pause > monitor after fsimage/edits loading. > Propose to start JVM pause monitor before loading fsimage/edits. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HDFS-9655) NN should start JVM pause monitor before loading fsimage
[ https://issues.apache.org/jira/browse/HDFS-9655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] John Zhuge updated HDFS-9655: - Attachment: HDFS-9655.001.patch Patch 001: * Start JVM pause monitor before loading fsimage during NN startup. * Log an info message when JVM pause monitor starts. > NN should start JVM pause monitor before loading fsimage > > > Key: HDFS-9655 > URL: https://issues.apache.org/jira/browse/HDFS-9655 > Project: Hadoop HDFS > Issue Type: Bug >Reporter: John Zhuge >Assignee: John Zhuge >Priority: Critical > Attachments: HDFS-9655.001.patch > > > We have seen many cases of NameNode startup either extremely slow or even > hung. Most of them were caused by insufficient heap size with regard to the > metadata size. Those cases were resolved by increasing the heap size. > However it did take support team some time to root cause. JVM pause warning > messages would greatly assist in such diagnosis, but NN starts JVM pause > monitor after fsimage/edits loading. > Propose to start JVM pause monitor before loading fsimage/edits. -- This message was sent by Atlassian JIRA (v6.3.4#6332)