Re: Time's a wastin'

[Randy Terbush <ra...@zyzzyva.com>]

> > What else needs to be done:
> > 
> > 2) There are about 25 unresolved issues I've seen over the last
> > three months, since 1.1.  I'll be forwarding them along, with
> > comment.  They are mostly small bugfixes, usually witch patches
> > attached, which somehow we've skipped over, I think.  They
> > were all posted to new-httpd but just got lost in the noise.
> 
> Dat's why we need the bugs database!
> 
> Cheers,
> 
> Ben.

FWIW, I've been squirreling away patches that look useful. I have the
following things on my list that I will try to address in the next few
days. Some of these include application of patches we have received.
Let me know if some of these have been fixed that I have missed.

* Change limit directives to use TAKE1|2
	(looking now, I thought Alexei commited such a beast?)

* Make SymLinksIfOwnerMatch imply FollowSymLinks

* Compiler warnings on platforms without setrlimit()

* Docs for Limit directives

Patches to apply:

* Continuation header support - Paul Sutton (fixed?)

* Chroot support. We received what looks like a good patch.

* SIGBUS in mod_dir - (patch but looks like added work)

* Here is one that at the time seemed important, but can't remember
  what the problem was. Anyone?
Slawa V. Olhovchenkov wrote:

*** alloc.c:bug	Tue Jun 18 00:43:45 MSK DST 1996
--- alloc.c	Wed Jul 17 17:38:35 MSK DST 1996
***************
*** 865,870 ****
--- 865,871 ----
      /* Child process */
      
      if (pipe_out) {
+       kill_cleanup(p, (void *)STDOUT_FILENO, fd_cleanup);
        close (out_fds[0]);
        dup2 (out_fds[1], STDOUT_FILENO);
        close (out_fds[1]);
***************
*** 871,876 ****
--- 872,878 ----
      }
  
      if (pipe_in) {
+       kill_cleanup(p, (void *)STDIN_FILENO, fd_cleanup);
        close (in_fds[1]);
        dup2 (in_fds[0], STDIN_FILENO);
        close (in_fds[0]);