-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/46384/
-----------------------------------------------------------

Review request for Ambari and Dmitro Lisnichenko.


Bugs: AMBARI-15972
    https://issues.apache.org/jira/browse/AMBARI-15972


Repository: ambari


Description
-------

While express upgrade is in progress and if disk space becomes full on any
node, the EU operation's failure happens on that particular node. When task
failure happens, appropriate message should be shown corresponding to failed
task indicating that failure happened since disk space was not available.

But for some task failures, message shown is not very informative. In the
recent Eu run, test failure happened while restarting node manager. But in
stdout.txt and stderr.txt no information is displayed which indicates that
failure is due to disk space.  
But from yarn-nodemanager logs, its evident that failure happened since disk
space was not available on node.  
"java.lang.UnsatisfiedLinkError: Could not load library. Reasons: [no
leveldbjni64-1.6.1.2.4.2.0-178 in java.library.path, no
leveldbjni-1.6.1.2.4.2.0-178 in java.library.path, no leveldbjni in
java.library.path, No space left on device]"

Attaching the logs.


Diffs
-----

  
ambari-server/src/main/resources/common-services/YARN/2.1.0.2.0/package/scripts/nodemanager_upgrade.py
 65709ac 
  
ambari-server/src/main/resources/common-services/YARN/2.1.0.2.0/package/scripts/service.py
 0703c1f 
  ambari-server/src/test/python/stacks/2.0.6/YARN/test_nodemanager.py 665ebfb 

Diff: https://reviews.apache.org/r/46384/diff/


Testing
-------

mvn clean test


Thanks,

Andrew Onischuk

Reply via email to