Horton

Re: [PATCH] Log details for client certificate failures

Jacob Champion <jchampion@timescale.com>

View thread View on postgr.es
From: Jacob Champion <jchampion@timescale.com>
To: Andres Freund <andres@anarazel.de>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Peter Eisentraut <peter.eisentraut@enterprisedb.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2022-07-19T22:08:38Z
Lists: pgsql-hackers

Attachments

On Tue, Jul 19, 2022 at 10:09 AM Andres Freund <andres@anarazel.de> wrote:
> On 2022-07-19 12:39:43 -0400, Tom Lane wrote:
> > Having said that, I struggle to see why we are panicking about badly
> > encoded log data from this source while blithely ignoring the problems
> > posed by non-ASCII role names, database names, and tablespace names.
>
> I think we should fix these as well. I'm not as concerned about post-auth
> encoding issues (i.e. tablespace name) as about pre-auth data (role name,
> database name) - obviously being allowed to log in already is a pretty good
> filter...

v2 adds escaping to pg_clean_ascii(). My original attempt used
StringInfo allocation, but that didn't play well with guc_malloc(), so
I switched to a two-pass API where the caller allocates. Let me know
if I'm missing something obvious; this way is more verbose than I'd
like...

Thanks,
--Jacob

Commits

  1. Fix tiny memory leaks

  2. Don't reflect unescaped cert data to the logs

  3. pg_clean_ascii(): escape bytes rather than lose them

  4. Log details for client certificate failures