Tom Lane <tgl@sss.pgh.pa.us> wrote:
> "Kevin Grittner" <Kevin.Grittner@wicourts.gov> writes:
> Well, using a different user per instance is a good idea because
> then the safety analysis I gave holds rigorously for each instance.
> It doesn't get you out of the problem by itself, because the problem
> as described can happen with just one instance.
Oh, right -- it does let PostgreSQL automatically deal with the file
left by a different instance, but could still fail on it's own file.
>> It must buy something in our environment, because our attempts to
>> use the sample script with minimal modification were problematic.
>> Unfortunately I forget the details, but our problems vanished when
>> we switched to pg_ctl. (Well, except for that one unfortunate
>> episode mentioned above.)
>
> Hmm. As stated, I would expect pg_ctl to make it worse. It would
> be interesting to have a closer look at your before-and-after
> scripts.
I don't remember whether the problem had anything to do with the lock
files; it could have been that pg_ctl was doing something else which
worked better for us than the direct invocation of postmaster, at
lease with the options we were using. I'll experiment and see what
happens in a test environment. I don't think we saved our failed
attempts from years back. Also, I was quite green with Linux at the
time, and it was my first initscript tinkering -- it's not outside the
range of possibility that I did something dumb with the direct
attempt. That said, knowing so little about what I was doing with it
had me starting from the point of changing as little as I could manage
from the sample to try to get it going.
-Kevin