Re: A test for replay of regression tests

Thomas Munro <thomas.munro@gmail.com> writes:
> Seen again today on prairiedog.  Erm, scratch that idea, HS feedback
> interferes with test results.  I guess max_standby_streaming_delay
> should be increased to 'forever', like in the attached, since pg_dump
> runs for a very long time on prairiedog:

FWIW, I'd vote for keeping a finite timeout, but making it say
ten minutes.  If the thing gets stuck for some reason, you don't
really want the test waiting forever.  (Some buildfarm animals
have overall-test-time limits, but I think it's not the default,
and the behavior when that gets hit is pretty unfriendly anyway
-- you don't get any report of the run at all.)

            regards, tom lane