Re: [HACKERS] TAP tests take a long time - Mailing list pgsql-hackers

From Stephen Frost
Subject Re: [HACKERS] TAP tests take a long time
Date
Msg-id 20170412194123.GL9812@tamriel.snowman.net
Whole thread Raw
In response to Re: [HACKERS] TAP tests take a long time  (Andrew Dunstan <andrew.dunstan@2ndquadrant.com>)
List pgsql-hackers
Andrew,

* Andrew Dunstan (andrew.dunstan@2ndquadrant.com) wrote:
> On 04/12/2017 08:40 AM, Andrew Dunstan wrote:
> > On 04/12/2017 01:27 AM, Craig Ringer wrote:
> >> BTW, I suggest adding --timer to our default PROVE_FLAGS, so we can
> >> collect more data from the buildfarm on what takes a while. Sample
> >> output:
> >
> > I'll add that to the commandline the buildfarm uses in the upcoming release.
>
> See
> <https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=nightjar&dt=2017-04-12%2013%3A54%3A22>
>
> One thing I noticed was this
>
>     [10:02:40] t/001_basic.pl ......... ok      320 ms
>     [10:02:59] t/002_pg_dump.pl ....... ok    19441 ms
>     [10:03:47] t/010_dump_connstr.pl .. ok    47430 ms
>
> Why does the last test set take 47s?

It's doing a bunch of dropdb/createdb stuff, and also creates multiple
clusters.  Part of this is because it's generating random names for
databases to make sure they are handled correctly.  I'd certainly love
to see it take less time but I've not really done very much with it.

Thanks!

Stephen

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: [HACKERS] Inadequate parallel-safety check for SubPlans
Next
From: Peter Eisentraut
Date:
Subject: Re: [pgsql-www] [HACKERS] Small issue in online devel documentationbuild