You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Martin Kraemer <Ma...@mch.sni.de> on 1998/06/01 22:09:45 UTC
Re: 1.3b8 ==> 1.3.0
On Fri, May 29, 1998 at 08:13:17PM -0700, Dean Gaudet wrote:
> Hey by background gc do you mean an external program that goes around
> doing it, rather than something that happens to get spawned as a
> side-effect of a request? yeah would be cool.
No, sorry. We talked about my patch (based on a discussion between you
and dirk last fall) where the GC is spawned off by a double-fork()
and setsid()/setpgrp() to avoid blocking the current request to finish.
> I'm +1 on the fixes going in as well.
Hmm... Under these circumstances, does that mean +0?
Too late anyway, no?
Martin
--
| S I E M E N S | <Ma...@mch.sni.de> | Siemens Nixdorf
| ------------- | Voice: +49-89-636-46021 | Informationssysteme AG
| N I X D O R F | FAX: +49-89-636-44994 | 81730 Munich, Germany
~~~~~~~~~~~~~~~~My opinions only, of course; pgp key available on request
Re: 1.3b8 ==> 1.3.0
Posted by Brian Behlendorf <br...@hyperreal.org>.
At 10:53 PM 6/1/98 +0200, Martin Kraemer wrote:
>And I have more proxy fixes to come.
Since you're paying attention to the proxy code these days, could you take
a look at PR2277, a patch to fix a bug in aborted xfers being cached, as
well as complete caching of those aborted xfers? The author is thinking
about adding a few things to it, along the lines of what Squid does.
Brian
--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
pure chewing satisfaction brian@apache.org
brian@hyperreal.org
Re: 1.3b8 ==> 1.3.0
Posted by Martin Kraemer <Ma...@mch.sni.de>.
On Mon, Jun 01, 1998 at 01:52:44PM -0700, Dean Gaudet wrote:
>
> No I'm still +1 -- I had one comment on your patch (the missing
> ap_cleanup_for_exec). The above comment was because I thought you guys
> were talking about some other new concept.
Yes, I incorporated that (replaced by comment with the ap_cleanup_for_exec()
call) without sending out a second [PATCH]. But as nobody seems to have
tested it, and only you and Randy studied the code, I agree with Jim
that it's best to add this to the 1.3.1-to-be.
And I have more proxy fixes to come. That will mostly be code cleanup,
improved ftp authentication (by Basic authentication method instead of
clean text URL password). I'm currently testing and will post some
questions to the list soon.
I'm also thinking about a last-modified header for ftp-proxy served files.
That requires an intermediate LIST request and an "ls -l" unparser, which
I'll have to write (it's in wget, but wget is GPL'ed). There's also the
problem with time zones... Are ftp directory listimgs local or GMTime?
Martin
--
| S I E M E N S | <Ma...@mch.sni.de> | Siemens Nixdorf
| ------------- | Voice: +49-89-636-46021 | Informationssysteme AG
| N I X D O R F | FAX: +49-89-636-44994 | 81730 Munich, Germany
~~~~~~~~~~~~~~~~My opinions only, of course; pgp key available on request
Re: 1.3b8 ==> 1.3.0
Posted by Dean Gaudet <dg...@arctic.org>.
On Mon, 1 Jun 1998, Martin Kraemer wrote:
> On Fri, May 29, 1998 at 08:13:17PM -0700, Dean Gaudet wrote:
> > Hey by background gc do you mean an external program that goes around
> > doing it, rather than something that happens to get spawned as a
> > side-effect of a request? yeah would be cool.
>
> No, sorry. We talked about my patch (based on a discussion between you
> and dirk last fall) where the GC is spawned off by a double-fork()
> and setsid()/setpgrp() to avoid blocking the current request to finish.
>
> > I'm +1 on the fixes going in as well.
>
> Hmm... Under these circumstances, does that mean +0?
No I'm still +1 -- I had one comment on your patch (the missing
ap_cleanup_for_exec). The above comment was because I thought you guys
were talking about some other new concept.
Dean