Re: Tracing down buildfarm "postmaster does not shut down" failures

On 2/8/16 2:45 PM, Tom Lane wrote:
> Alvaro Herrera <alvherre(at)2ndquadrant(dot)com> writes:
>> Tom Lane wrote:
>>> What I'd like to do to investigate this is put in a temporary HEAD-only
>>> patch that makes ShutdownXLOG() and its subroutines much chattier about
>>> how far they've gotten and what time it is, and also makes pg_ctl print
>>> out the current time if it gives up waiting.
>
>> Seems like a reasonable place to start. I assume you'll be installing
>> some debug-elog calls, enabled by default, and then once the problem is
>> fixed, we'll change the default to disabled but keep the actual calls?
>
> I had in mind to just "git revert" the patch when we're done with it.
> There might be some parts we want to keep (for instance, I'm thinking of
> logging "postmaster is done" after we've unlinked the pidfile, which might
> be useful permanently). But I plan to err on the side of noisiness for
> the moment, rather than make something I think has long-term value.

It's already difficult enough for DBAs to debug some performance issues,
so getting rid of logging is a step backwards. I realize it's unlikely
that you could run DEBUG2 in a prod environment, but we still shouldn't
be reducing visibility.
--
Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX
Experts in Analytics, Data Architecture and PostgreSQL
Data in Trouble? Get it in Treble! http://BlueTreble.com