Horton

Re: BUG #17391: While using --with-ssl=openssl and PG_TEST_EXTRA='ssl' options, SSL tests fail on OpenBSD 7.0

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: Daniel Gustafsson <daniel@yesql.se>, Andres Freund <andres@anarazel.de>, byavuz81@gmail.com, PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>, Heikki Linnakangas <heikki.linnakangas@iki.fi>, Michael Paquier <michael@paquier.xyz>
Date: 2022-02-13T01:06:24Z
Lists: pgsql-bugs 
Thomas Munro <thomas.munro@gmail.com> writes:
> On Sun, Feb 13, 2022 at 10:56 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> I went ahead and pushed the patches-under-discussion (the larger
>> one only to HEAD, for now).  I've confirmed that the ssl tests
>> are solid now under OpenBSD 7.0, on bare metal as well as on the
>> VM I was using before.  They are still broken under OpenBSD 6.8,
>> which seems interesting but not so interesting that I want to
>> spend more time on it (seeing that 6.8 is EOL as far as
>> openbsd.org is concerned).  Does anyone want to recheck 6.9?

> Looks good:

Thanks for checking!

So I think the remaining question here is whether, and if so when,
to back-patch faa189c93.  I'm not sure that the benefit is worth
the risk of new problems (though maybe I'm just feeling particularly
pessimistic because of the regressions we've found in this week's
releases).  Leaving aside the behavior of the ssl tests, which
few people would run anyway, it seems like the benefit is just
to replace a not-very-helpful error message with a more-helpful
one.  That's worth something, but how much?

			regards, tom lane

Commits

  1. Move libpq's write_failed mechanism down to pqsecure_raw_write().

  2. Fix thinko in PQisBusy().

  3. Set SNI ClientHello extension to localhost in tests

  4. Improve worst-case performance of text_position_get_match_pos()