db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mike Matrigali <mikem_...@sbcglobal.net>
Subject Re: [jira] Created: (DERBY-2563) Intermittent failure in 'lang/autoincrement.sql'
Date Thu, 19 Apr 2007 16:37:59 GMT

Knut Anders Hatlen wrote:
> Mike Matrigali <mikem_app@sbcglobal.net> writes:
>>Ole Solberg wrote:
>>>The full output from the test should now be available  as
>>>-- Ole
>>Thanks, that had the info I needed.  It looks like the tinderbox report
>>cut off the diff - probably be cause of a long line.  The real diff is:
>>*** Start: autoincrement jdk1.6.0_01 derbyall:derbylang 2007-04-18
>>11:11:12 ***
>>                                                        |IX
>>Test Failed.
>>*** End:   autoincrement jdk1.6.0_01 derbyall:derbylang 2007-04-18
>>11:11:21 ***
> Stupid question, but what's the difference between the two diffs? They
> both consist of one line which contains TABLE|T1[lots-of-spaces]|IX. I'm
> sure I'm missing something.

Not a stupid question, I was being stupid.  I didn't realize my browser 
was cutting off the end of the diff.  I am used to reading the diffs in
my favorite editor and just stared at the screen without realizing I 
needed to scroll.

Though it is nice to be able to see the complete state of a test 
failure, often what you need is not just in the diff.  With the old
tests this is pretty well isolated from other tests.  I am a really big
fan of the public posting of the nightly and tinderbox results.  It 
seems like the more info we can automatically post for everyone about
test failures, the more likely more people in the community can jump
in and address the failures.  I really appreciate the help in the last
few weeks to get the tinderbox diffs down to 0.  I think that was the
first time in at least 3 weeks.

I know I am
having problems identifying bugs from the junit suite run as the
stack traces are being cut off and it looks like derby.log are not
kept around for all the tests.

View raw message