Re: PostgreSQL12 and older versions of OpenSSL

Поиск
Список
Период
Сортировка
От Michael Paquier
Тема Re: PostgreSQL12 and older versions of OpenSSL
Дата
Msg-id 20190925065514.GJ1815@paquier.xyz
обсуждение исходный текст
Ответ на Re: PostgreSQL12 and older versions of OpenSSL  (Tom Lane <tgl@sss.pgh.pa.us>)
Список pgsql-hackers
On Tue, Sep 24, 2019 at 11:25:30AM -0400, Tom Lane wrote:
> Alvaro Herrera <alvherre@2ndquadrant.com> writes:
>> ... I wonder if we should really continue to support
>> OpenSSL 0.9.8.
>
> Fair question, but post-rc1 is no time to be moving that goalpost
> for the v12 branch.

Yeah.  I worked in the past with SUSE-based appliances, and I recall
that those folks have been maintaining their own patched version of
OpenSSL 0.9.8 with a bunch of custom patches, some of them coming from
newer versions of upstream to take care of security issues with 0.9.8.
So even if they call their version 0.9.8j, I think that they include
much more security-related fixes than their version string suggests.
I don't know at which extent though.

>> Anyway I suppose it's not impossible that third parties are still
>> maintaining their 1.0.0 branch,
>
> Another data point on that is that Red Hat is still supporting
> 1.0.1e in RHEL6.  I don't think we should assume that just because
> OpenSSL upstream has dropped support for a branch, it no longer
> exists in the wild.
>
> Having said that, if it makes our lives noticeably easier to
> drop support for 0.9.8 in HEAD, I won't stand in the way.

Agreed.  There is an argument for dropping support for OpenSSL 0.9.8
in 13~, but I don't agree of doing that in 12.  Let's just fix the
issue.
--
Michael

Вложения

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Kyotaro Horiguchi
Дата:
Сообщение: Re: Remove page-read callback from XLogReaderState.
Следующее
От: Kyotaro Horiguchi
Дата:
Сообщение: Re: PATCH: standby crashed when replay block which truncated instandby but failed to truncate in master node