Re: Convert encrypted SSL test keys to PKCS#8 format

From: Jacob Champion <jchampion(at)timescale(dot)com>
To: Peter Eisentraut <peter(at)eisentraut(dot)org>
Cc: pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Convert encrypted SSL test keys to PKCS#8 format
Date: 2023-08-22 19:02:02
Message-ID: CAAWbhmh8GMZvhGy097cHiy584KUb9ee7Gs0x+gD8+w0hryJW+A@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, Aug 22, 2023 at 1:07 AM Peter Eisentraut <peter(at)eisentraut(dot)org> wrote:
> I have attached two patches, one to update the generation rules, and one
> where I have converted the existing test files. (I didn't generate them
> from scratch, so for example
> src/test/modules/ssl_passphrase_callback/server.crt that corresponds to
> one of the keys does not need to be updated.)

Looks good from here. I don't have a FIPS setup right now, but the new
files pass tests on OpenSSL 1.0.2u, 1.1.1v, 3.0.2-0ubuntu1.10, and
LibreSSL 3.8. Tests continue to pass after a full clean and rebuild of
the sslfiles.

> It's also interesting that if you generate all private keys from scratch
> using the existing rules on a new OpenSSL version (3+), they will be
> generated in PKCS#8 format by default. In those OpenSSL versions, the
> openssl-rsa command has a -traditional option to get the old format, but
> of course old OpenSSL versions don't have that. As OpenSSL 3 gets more
> widespread, we might need to rethink these rules anyway to make sure we
> get consistent behavior.

Yeah. Looks like OpenSSL 3 also adds new v3 extensions to the
certificates... For now they look benign, but I assume someone's going
to run into weirdness at some point.

Thanks!
--Jacob

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2023-08-22 19:14:31 Re: PG 16 draft release notes ready
Previous Message Andrew Dunstan 2023-08-22 18:46:46 Re: Make all Perl warnings fatal