| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Santiago DIEZ <santiago(dot)diez(at)caoba(dot)fr> |
| Cc: | "David G(dot) Johnston" <david(dot)g(dot)johnston(at)gmail(dot)com>, Andres Freund <andres(at)anarazel(dot)de>, PostgreSQL LIST <pgsql-admin(at)postgresql(dot)org> |
| Subject: | Re: Change Unix Socket directory at build time |
| Date: | 2017-04-14 17:40:41 |
| Message-ID: | 8473.1492191641@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-admin |
Santiago DIEZ <santiago(dot)diez(at)caoba(dot)fr> writes:
> no no, if I set -t to 60 seconds (or even 1000), it's the same problem.
David's right though: any reasonably modern version of pg_ctl (since
about 9.1 I believe) gets everything it needs to know out of the
postmaster.pid file. Compiling in a different default
unix_socket_directory might save some effort when using psql or other
clients, but it should make no difference to pg_ctl.
It might be useful to strace pg_ctl to see what it's doing in the
problem case.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Steven Chang | 2017-04-17 03:59:45 | Bitmap Index Scan when btree index created |
| Previous Message | Santiago DIEZ | 2017-04-14 17:12:23 | Re: Change Unix Socket directory at build time |