Re: libpq: system-wide root.crt
От | Magnus Hagander |
---|---|
Тема | Re: libpq: system-wide root.crt |
Дата | |
Msg-id | AANLkTikt9v2ntHY0mkU1R5G-TKzoat+P18r6_K5X+L4r@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: libpq: system-wide root.crt (Peter Eisentraut <peter_e@gmx.net>) |
Ответы |
Re: libpq: system-wide root.crt
|
Список | pgsql-bugs |
On Fri, Sep 24, 2010 at 00:41, Peter Eisentraut <peter_e@gmx.net> wrote: > On tor, 2010-09-23 at 08:36 +0200, Magnus Hagander wrote: >> I wonder if we want to have a default value for this rather than >> disabling it when it's not specified by configure. But is there any >> kind of reasonable default that's not going to be >> platform/distribution specific? > > I would like for us to find a way to do it without configure options. > Compiling in stuff is always weird. I think the compile-time default really is designed for distributions - in this case debian, right? I would assume debian would prefer a combination of a compile-time default and the below, rather than just the default? > I'm thinking, libpq could really use a global (and, for that matter, a > per-user) configuration file, where you could set defaults for some of > the things that you currently have to use environment variables for, > e.g., sslmode. =A0And then you can configure the system-wide root.crt > location there. We already have this - pg_service.conf - no? > Alternatively, if you think that that is overkill, then using an > environment variable to configure this feature would be consistent with > the existing mechanisms. So in this case, a "sslfallbackroot=3D/etc/somewhere" parameter, that you could then stick in etc/pg_service.conf, or in any of the other places you can specify it? The one thing I think would be really needed for that to work is to support wildcard database names in pg_service.conf? --=20 =A0Magnus Hagander =A0Me: http://www.hagander.net/ =A0Work: http://www.redpill-linpro.com/
В списке pgsql-bugs по дате отправления: