Horton

Re: Removal of support for OpenSSL 0.9.8 and 1.0.0

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Michael Paquier <michael@paquier.xyz>
Cc: Daniel Gustafsson <daniel@yesql.se>, Postgres hackers <pgsql-hackers@lists.postgresql.org>, Peter Eisentraut <peter.eisentraut@2ndquadrant.com>
Date: 2020-01-02T14:22:47Z
Lists: pgsql-hackers 
Michael Paquier <michael@paquier.xyz> writes:
> Sorry for letting this thread down for a couple of weeks, but I was
> hesitating to apply the last patch of the series as the cleanup of the
> code related to OpenSSL 0.9.8 and 1.0.0 is not that much.  An extra
> argument in favor of the removal is that this can allow more shaving
> of past Python versions, as proposed by Peter here:
> https://www.postgresql.org/message-id/98b69261-298c-13d2-f34d-836fd9c29b21@2ndquadrant.com

> So, let's do it.

FWIW, I'm not sure I see why there's a connection between moving up
the minimum Python version and minimum OpenSSL version.  Nobody is
installing bleeding-edge Postgres on RHEL5, not even me ;-), so I
don't especially buy Peter's line of reasoning.

I'm perfectly okay with doing both things in HEAD, I just don't
see that doing one is an argument for or against doing the other.

			regards, tom lane

Commits

  1. Remove support for OpenSSL 0.9.8 and 1.0.0

  2. Fix handling of OpenSSL's SSL_clear_options

  3. Remove configure check for OpenSSL's SSL_get_current_compression()