You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Dean Gaudet <dg...@arctic.org> on 1997/09/26 06:12:22 UTC

[STATUS] 1.3b1: Thu Sep 25 21:05:41 PDT 1997

Thanks guys for voting today, it's nice to get that backlog cleared. 
Other than win32 issues I think we're ready to roll 1.3b1. 

proposal: everyone start testing now, if there's no progress on win32 by
wed oct 1 we release anyhow.

Committed:

    * PR#1155: RedirectMatch does not deal with urls that need to be escaped
    * RedirectMatch /advertiser/(.*) $1 doesn't work
    * mod_autoindex: deal with &esc; and truncate only when necessary
    * makefile: exit when a subdir fails
    * PR#974: proxy matching bug
    * suexec wouldn't build without LOG_EXEC
    * mod_imap declines non-GET methods
    * PR#1139: mod_include evaluates some exprs with && or || incorrectly
    * send_fb doesn't do aborts right
    * PR#918: use proper User/Group when testing CGI permissions

Open issues:

    * WIN32 problem reports

    * PR#1159: Configure always exits with return code 0

    * Need a version of Garey's OS/2 proxy patch.

    * PR#1049: name-based, multi-port servers don't work
      Dean posted a patch but Martin mentioned it was wrong and proposed a
      better solution.  Waiting for his work.

    * new logging verbosity issues

    * Martin gets a SEGV occasionally
    * Marc is having some auth problems
	Dean thinks these are related and probably his fault.  Hopes to
	work on it shortly.

    * Brian had an undying child problem which Dean thinks is fixed by the
	send_fb change above.


Re: [STATUS] 1.3b1: Thu Sep 25 21:05:41 PDT 1997

Posted by Marc Slemko <ma...@worldgate.com>.
On Thu, 25 Sep 1997, Dean Gaudet wrote:

> Thanks guys for voting today, it's nice to get that backlog cleared. 
> Other than win32 issues I think we're ready to roll 1.3b1. 
> 
> proposal: everyone start testing now, if there's no progress on win32 by
> wed oct 1 we release anyhow.

I am concerned about the core dump / auth oddities and the unresolved
logging and makefile stuff.

The weekend is coming (I think... erm... yes, it is) so I may get some
time to work on some of that on the weekend.


Re: [STATUS] 1.3b1: Thu Sep 25 21:05:41 PDT 1997

Posted by Marc Slemko <ma...@worldgate.com>.
On Sun, 28 Sep 1997, Marc Slemko wrote:

> On Thu, 25 Sep 1997, Dean Gaudet wrote:
> 
> >     * Martin gets a SEGV occasionally
> >     * Marc is having some auth problems
> > 	Dean thinks these are related and probably his fault.  Hopes to
> > 	work on it shortly.
> 
> Also note that I am seeing oddities with proxy access restrictions not
> working; may or may not be related.
> 

Grr.  The problem with "require user marcs" letting user joe in is caused
by only the check_user_id handler being run and not the check_auth.  This
is because mod_example was complied into the server, and it 
returns OK for all requests in its example_auth_checker, so nothing works
right.

While you can argue that it is just for testing and shouldn't be enabled,
I think that having a module that demonstrates how to do things but causes
such bogus behavior isn't good.  Sigh.

Note that this does _not_ fix my problems with proxy access restrictions
not working.


Re: [STATUS] 1.3b1: Thu Sep 25 21:05:41 PDT 1997

Posted by Marc Slemko <ma...@worldgate.com>.
On Thu, 25 Sep 1997, Dean Gaudet wrote:

>     * Martin gets a SEGV occasionally
>     * Marc is having some auth problems
> 	Dean thinks these are related and probably his fault.  Hopes to
> 	work on it shortly.

Also note that I am seeing oddities with proxy access restrictions not
working; may or may not be related.