Hari Sekhon created AMBARI-10799:
------------------------------------
Summary: HDP 2.2.0.0 => HDP 2.2.4.0 upgrade Oozie did not upgrade,
health check failed with Error: java.io.IOException: Split class
org.apache.oozie.action.hadoop.OozieLauncherInputFormat$EmptySplit not found
Key: AMBARI-10799
URL: https://issues.apache.org/jira/browse/AMBARI-10799
Project: Ambari
Issue Type: Bug
Components: ambari-server
Affects Versions: 2.0.0
Environment: HDP 2.2.0.0 => HDP 2.2.4.2
Reporter: Hari Sekhon
When upgrading HDP stack from 2.2.0.0 => 2.2.4.2 via Ambari, the Oozie health
check failed with Failed/Killed for the job in the Oozie output. On digging in
to the task logs it turns out this is the root cause:
{code}2015-04-28 15:55:48,670 INFO [AsyncDispatcher event handler]
org.apache.hadoop.mapreduce.v2.app.job.impl.TaskAttemptImpl: Diagnostics report
from attempt_1430229797293_0015_m_000000_1: Error: java.io.IOException: Split
class org.apache.oozie.action.hadoop.OozieLauncherInputFormat$EmptySplit not
found
at org.apache.hadoop.mapred.MapTask.getSplitDetails(MapTask.java:363)
at org.apache.hadoop.mapred.MapTask.runOldMapper(MapTask.java:423)
at org.apache.hadoop.mapred.MapTask.run(MapTask.java:343)
at org.apache.hadoop.mapred.YarnChild$2.run(YarnChild.java:163)
at java.security.AccessController.doPrivileged(Native Method)
at javax.security.auth.Subject.doAs(Subject.java:415)
at
org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1628)
at org.apache.hadoop.mapred.YarnChild.main(YarnChild.java:158)
Caused by: java.lang.ClassNotFoundException: Class
org.apache.oozie.action.hadoop.OozieLauncherInputFormat$EmptySplit not found
at
org.apache.hadoop.conf.Configuration.getClassByName(Configuration.java:1982)
at org.apache.hadoop.mapred.MapTask.getSplitDetails(MapTask.java:361)
... 7 more{code}
This seemed to be a shared library problem. I then found that
/usr/hdp/current/oozie-server and /usr/hdp/current/oozie-current were both
still set to the old oozie installation on 2.2.0.0.
Workaround: I shut down Oozie, did hdp-select to the new version for both
client and server, then regenerated and deployed the Oozie war to HDFS and
upgraded the db, before starting Oozing and retrying after which the health
check succeeded and got past this upgrade blocker. Fyi this is a Kerberized
cluster.
Hari Sekhon
http://www.linkedin.com/in/harisekhon
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
