← Back to team overview

launchpad-dev team mailing list archive

Re: test_min_time_to_next_builder failing

 

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Muharem Hrnjadovic wrote:
> On 03/09/2010 08:48 PM, Aaron Bentley wrote:
> I'm getting test failures in stable.  Both my devel and test machine are
> affected.
> 
> I've done make clean and make schema, updated my download-cache and my
> sourcecode, and still I get test failures.  Can anyone help?  See below:
>> Hello Aaron,
> 
>> I am sorry for the trouble you're having with this test. The only reason
>> I can think of is that the lines that directly precede both failing
>> tests
> 
>>         build, bq = find_job(self, 'postgres')
>>         set_remaining_time_for_running_job(bq, 30)
> 
>> for some reason did not take effect. I am at a loss to explain why this
>> would be the case though.

It appears that you are entering the ELSE clause of your delay_query
SELECT.  This patch fixes test_min_time_to_next_builder.

=== modified file 'lib/lp/buildmaster/model/buildqueue.py'
- --- lib/lp/buildmaster/model/buildqueue.py	2010-03-08 12:35:15 +0000
+++ lib/lp/buildmaster/model/buildqueue.py	2010-03-10 18:14:05 +0000
@@ -242,7 +242,7 @@
                 -- Please note that this is entirely innocuous i.e. if our
                 -- guess is off nothing bad will happen but our
estimate will
                 -- not be as good as it could be.
- -                120
+                30
               END)
             FROM
                 BuildQueue, Job, Builder
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkuX4lAACgkQ0F+nu1YWqI1OEwCeNR5FnffY9NcHcd7R8gAnzZFA
xOcAnRwoCCBH06QbZEQO2cGH0bqQgdwe
=vCJf
-----END PGP SIGNATURE-----



Follow ups

References