Re: buildfarm notifications

From: Andrew Dunstan <andrew(at)dunslane(dot)net>
To: Josh Berkus <josh(at)agliodbs(dot)com>
Cc: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: buildfarm notifications
Date: 2005-06-19 20:08:22
Message-ID: 42B5D0B6.1050404@dunslane.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-patches

Josh Berkus wrote:

>>I am expecting that for the most part people will want the lists of
>>state changes, rather than the lists of all builds or failures. Will
>>Spikesource tests track state changes?
>>
>>
>
>They'd like to. CVS makes this kind of challenging. They'd be happy to
>have suggestions ...
>
>

In the buildfarm system, the state of each machine/branch is either OK
or the stage in the buildfarm process that failed - we stop at the first
such failure. Since all the data reported is stored in our db, finding
the previous state is fairly simple - we just run this:

select coalesce(
(select distinct on (snapshot) stage
from build_status
where sysname = ? and branch = ? and snapshot < ?
order by snapshot desc
limit 1),
'NEW') as prev_status

plugging in the values from the current build.

>
>
>>BTW, these list are being set up only for announcements, so I would have
>>to grant permission before any results started flowing.
>>
>>
>
>Yep, that's why I'm mentioning it.
>
>
>

send me details off-list.

cheers

andrew

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andreas Pflug 2005-06-19 20:53:02 Re: [PATCHES] default database creation with initdb
Previous Message Josh Berkus 2005-06-19 19:42:56 Re: buildfarm notifications

Browse pgsql-patches by date

  From Date Subject
Next Message Andrew Dunstan 2005-06-19 21:09:36 Re: buildfarm notifications
Previous Message Josh Berkus 2005-06-19 19:42:56 Re: buildfarm notifications