[ 
http://issues.apache.org/jira/browse/DERBY-937?page=comments#action_12417814 ] 

Andreas Korneliussen commented on DERBY-937:
--------------------------------------------

This test is failing pretty consistently on Windows. 

I looked at the properties for the test, and found this:

#checkpoint influences the plans in some cases becuase rows count 
#getupdated on checkpoint, following is necessart to check some plans
derby.storage.checkpointInterval=100000

Based on this, I have the following question:
Does really the checkpointinterval influence this test ? The indexes are 
created after the tables are populated, so the statistics should be updated or 
..?

Would it make any difference to the test if we add a sleeping period after the 
population of the tables ? I think this could make a difference, if the 
failures are caused by the optimizer choosing plans based on outdated 
statistics.

> Instability in wisconsin test
> -----------------------------
>
>          Key: DERBY-937
>          URL: http://issues.apache.org/jira/browse/DERBY-937
>      Project: Derby
>         Type: Test

>   Components: Test, Regression Test Failure
>     Versions: 10.1.2.4
>     Reporter: Rick Hillegas
>  Attachments: 10.1_05_05_06_failure.zip
>
> The lang/wisconsin.java test prints out query plans. These plans differ 
> depending on your platform. When run under cygwin on xp, the plans vary from 
> the canonized results. I am seeing this instability in my own test 
> environment and in the tinderbox tests against cygwin.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira

Reply via email to