From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Peter J(dot) Holzer" <hjp-pgsql(at)hjp(dot)at> |
Cc: | pgsql-general(at)lists(dot)postgresql(dot)org |
Subject: | Re: Permissions on postgresql.conf, psql and patroni |
Date: | 2019-09-09 14:03:57 |
Message-ID: | 28297.1568037837@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Peter J. Holzer" <hjp-pgsql(at)hjp(dot)at> writes:
> Yesterday I "apt upgrade"d patroni (to version 1.6.0-1.pgdg18.04+1 from
> http://apt.postgresql.org/pub/repos/apt bionic-pgdg/main).
> Today I noticed that I couldn't invoke psql as an unprivileged user
> anymore:
> % psql
> Error: Invalid data directory for cluster 11 main
FYI, there's no such error string in the core Postgres sources.
I imagine that it's coming from some wrapper script.
> 1) Is there a reason to restrict read access to postgresql.conf to the
> user postgres?
Probably not, though it only matters if you keep that outside the
data directory.
> 2) Why does psql need to read postgresql.conf, and more specifically,
> why does it care about the location of the data directory?
psql itself won't/doesn't. I suspect you're dealing with a wrapper
that is trying to figure out which version of psql to invoke.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2019-09-09 14:17:23 | Re: database "cdf_100_1313" does not exist |
Previous Message | nikhil raj | 2019-09-09 13:43:43 | Re: database "cdf_100_1313" does not exist |