Magnus Hagander <magnus@hagander.net> writes:
> * On Unix, the core file is dropped in the database directory, we
> don't have a separate directory for crashdumps. If we want to be
> consistent, we should do that here too. I do think that storing them
> in a directory like "crashdumps" is better, but I just wanted to raise
> the comment.
Just a note on that - it's by no means universal that Unix systems will
put the core files in $PGDATA. OS X likes to put them in /cores, which
I think is a convention shared with some other BSDish systems. On Linux
I believe it's possible to configure where the core goes via environment
settings.
> * However, when storing it in crashdumps, I think the code would need
> to create that directory if it does not exist, doesn't it?
If it didn't do so, then manual creation/removal of the directory could
be used as an on/off switch for the feature. Which would have a number
of advantages, not least that you don't need to have the crash dumper
dependent on GUC working. I haven't looked at the patch but this
discussion makes it sound like the dumper is dependent on an
uncomfortably large amount of backend code being functional. You need
to minimize the number of assumptions of that sort.
regards, tom lane