RE: POC: GUC option for skipping shared buffers in core dumps

From: Craig Ringer <craig(at)2ndquadrant(dot)com>
> Currently my options are "dump all shmem including shared_buffers" or
> "dump no shmem". But I usually want "dump all shmem except
> shared_buffers". It's tolerable to just dump s_b on a test system with
> a small s_b, but if enabling coredumps to track down some
> hard-to-repro crash on a production system I really don't want 20GB
> coredumps...

We have a simple implementation that allows to exclude shared memory. That's been working for years.

[postgresql.conf]
core_directory = 'location of core dumps'
core_contents = '{none | minimum | full}'
# none = doesn't dump core, minimum excludes shared memory, and full dumps all

I can provide it. But it simply changes the current directory and detaches shared memory when postgres receives signals that dump core.

I made this GUC because Windows also had to be dealt with.

From: Andres Freund <andres(at)anarazel(dot)de>
> > Hah. This argument boils down to saying our packagers suck :-)
>
> Hm? I'd say it's a sign of respect to not have each of them do the same
> work. Especially when they can't address it to the same degree core PG
> can. So maybe I'm saying we shouldn't be lazy ;)

Maybe we should add options to pg_ctl just like -c which is available now, so that OS packagers can easily use in their start scripts. Or, can they just use pg_ctl's -o to specify new GUC parameters?

Regards
Takayuki Tsunakawa