[
https://issues.apache.org/jira/browse/HDFS-9187?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14940608#comment-14940608
]
Colin Patrick McCabe commented on HDFS-9187:
--------------------------------------------
This is a different approach that just removes Tracer from FileSystem, given
the difficulty of ensuring it's initialized. (The other Tracers like DFSClient
don't have this problem since the constructor ALWAYS does it... unlike FS where
there is a no-argument constructor.) I also added a unit test which is a
regression test for this bug.
> Check if tracer is null before using it
> ---------------------------------------
>
> Key: HDFS-9187
> URL: https://issues.apache.org/jira/browse/HDFS-9187
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: tracing
> Affects Versions: 2.8.0
> Reporter: stack
> Assignee: Colin Patrick McCabe
> Attachments: HDFS-9187.001.patch, HDFS-9187.002.patch
>
>
> Saw this where an hbase that has not been updated to htrace-4.0.1 was trying
> to start:
> {code}
> Oct 1, 5:12:11.861 AM FATAL org.apache.hadoop.hbase.master.HMaster
> Failed to become active master
> java.lang.NullPointerException
> at org.apache.hadoop.fs.Globber.glob(Globber.java:145)
> at org.apache.hadoop.fs.FileSystem.globStatus(FileSystem.java:1634)
> at org.apache.hadoop.hbase.util.FSUtils.getTableDirs(FSUtils.java:1372)
> at
> org.apache.hadoop.hbase.util.FSTableDescriptors.getAll(FSTableDescriptors.java:206)
> at
> org.apache.hadoop.hbase.master.HMaster.finishActiveMasterInitialization(HMaster.java:619)
> at org.apache.hadoop.hbase.master.HMaster.access$500(HMaster.java:169)
> at org.apache.hadoop.hbase.master.HMaster$1.run(HMaster.java:1481)
> at java.lang.Thread.run(Thread.java:745)
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
