"make check" changes have caused buildfarm deterioration.

From: Andrew Dunstan <andrew(dot)dunstan(at)pgexperts(dot)com>
To: PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>
Subject: "make check" changes have caused buildfarm deterioration.
Date: 2015-07-20 20:51:13
Message-ID: 55AD5F41.10601@pgexperts.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Somewhere along the way some changes to the way we do "make check" have
caused a significant deterioration in the buildfarm's logging. Compare
these two from animal crake, which happens to be my test instance:
<http://www.pgbuildfarm.org/cgi-bin/show_stage_log.pl?nm=crake&dt=2015-07-20%2013%3A09%3A02&stg=check>
and
<http://www.pgbuildfarm.org/cgi-bin/show_stage_log.pl?nm=crake&dt=2015-07-20%2017%3A23%3A08&stg=check>

In the first place, we now have all that extraneous installation logging
at the top, and the stuff we are most likely to be interested in right
at the bottom. But more importantly, we are now missing the initdb log
and the postmaster log, and the function to get a stack trace if
required is almost certainly going to be broken as well.

What's most annoying is that some of the stuff for this hasn't just
moved, but some is now apparently now cleaned up and removed so the
buildfarm script can't get at it at all.

I'm going to look at what can be done to repair the damage. No doubt I
should have been paying more attention, but sometimes I just can't keep
track of everything going on. Fixing this will almost certainly involve
some core changes, though.

Incidentally, this has real consequences: I just went looking to find
the postmaster log of a case that had an error and it was missing -
that's how I noticed this.

cheers

andrew

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Adam Brightwell 2015-07-20 21:10:48 Unnecessary #include in objectaddress.h?
Previous Message Heikki Linnakangas 2015-07-20 20:23:41 Re: All-zero page in GIN index causes assertion failure