Re: Re: Removing SSL renegotiation (Was: Should we back-patch SSL renegotiation fixes?)

Поиск
Список
Период
Сортировка
От Michael Paquier
Тема Re: Re: Removing SSL renegotiation (Was: Should we back-patch SSL renegotiation fixes?)
Дата
Msg-id CAB7nPqR_6i77AcN=goWg=FMszdacNbPWUhhv4PN9k+qHL8PXGQ@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Re: Removing SSL renegotiation (Was: Should we back-patch SSL renegotiation fixes?)  (Andres Freund <andres@anarazel.de>)
Ответы Re: Re: Removing SSL renegotiation (Was: Should we back-patch SSL renegotiation fixes?)  (Andres Freund <andres@anarazel.de>)
Список pgsql-hackers


On Sat, Jul 11, 2015 at 9:28 PM, Andres Freund <andres@anarazel.de> wrote:
On 2015-07-11 21:09:05 +0900, Michael Paquier wrote:
> Something like the patches attached

Thanks for that!

> could be considered, one is for master
> and REL9_5_STABLE to remove ssl_renegotiation_limit, the second one for
> ~REL9_4_STABLE to change the default to 0.

> diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
> index c669f75..16c0ce5 100644
> --- a/doc/src/sgml/config.sgml
> +++ b/doc/src/sgml/config.sgml
> @@ -1040,7 +1040,7 @@ include_dir 'conf.d'
>          cryptanalysis when large amounts of traffic can be examined, but it
>          also carries a large performance penalty. The sum of sent and received
>          traffic is used to check the limit. If this parameter is set to 0,
> -        renegotiation is disabled. The default is <literal>512MB</>.
> +        renegotiation is disabled. The default is <literal>0</>.

I think we should put in a warning or at least note about the dangers of
enabling it (connection breaks, exposure to several open openssl bugs).

This sounds like a good idea to me. Here is an idea:
+       <warning>
+        <para>
+         Enabling <varname>ssl_renegotiation_limit</> can cause various
+         problems endangering the stability of a <productname>PostgreSQL</>
+         instance like connection breaking suddendly and exposes the
+         server to bugs related to the internal implementation of renegotiation
+         done in the SSL libraries used.
+        </para>
+       </warning>
Attached is v2 for ~9.4.
Regards,
--
Michael
Вложения

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Amit Kapila
Дата:
Сообщение: Re: drop/truncate table sucks for large values of shared buffers
Следующее
От: Michael Paquier
Дата:
Сообщение: Fixes for a couple of resource leaks