ptests are stuck

[Zoltan Haindrich <ki...@rxd.hu>]

Hello,

It's stuck again; but the current state is very interesting:

* current build is testing HIVE-18237
   https://issues.apache.org/jira/browse/HIVE-18237
* it's currently running for >5 hours
* hiveqa have posted ptest results for HIVE-18237 around 4 hours ago!
* from the jenkins build console it's currently executing batch #134 
(134-TestSparkCliDriver)
https://builds.apache.org/job/PreCommit-HIVE-Build/8139/consoleFull
* but at the corresponding ptest site:
http://104.198.109.242/logs/PreCommit-HIVE-Build-8139/succeeded/
   not that batch 134 have been finished; but all of them! (299?)
* I've searched for batch "177-TestMiniSparkOnYarnCliD" in the console 
output
https://builds.apache.org/job/PreCommit-HIVE-Build/8139/consoleFull
   and it only appears in the "generation" phase.
* taking a closer look at:
http://104.198.109.242/logs/PreCommit-HIVE-Build-8139/
   * according to the directory creation dates:
     all these directories are seem to be newer than the creation date 
of http://104.198.109.242/logs/PreCommit-HIVE-Build-8139/ + 2 hours;
     the dir http://104.198.109.242/logs/PreCommit-HIVE-Build-8139/ is 
lesser than 2 hours later created than 
http://104.198.109.242/logs/PreCommit-HIVE-Build-8138/
   * the patch matches with the one submitted in HIVE-18237
* closer look
   * 122-TestSpark
     the "completed executing" appears in 
http://104.198.109.242/logs/PreCommit-HIVE-Build-8139/execution.txt; but 
not in the actual build output console
   * 134-TestSpark
     the completed executing appears in both logs; with exactly the same 
timestamp

 From the above: I suspect that somehow there are 2 "builds" executing 
the same set of tests is happening somehow....
but anyway...could someone please take a closer look?

cheers,
Zoltan