Re: 2.0.36 .zip files WAS, RE: 2.0.36 release

[William A. Rowe, Jr.]

At 01:40 PM 5/6/2002, you wrote:
>Hi,
>
>Is there any special voodoo to be performed to create the apache release 
>zip files?

The nastiest.  Need to build all, then export make files (so that all 
dependent files
exist when creating dependencies) ... and pack up .mak, .dep, and .rc (awk 
version
resource stubs) along with cr/lf'ed files.

Really requires VC5 or 6.  I started some of that jazz in my win_compiling.html
file, and had started a file on how-to-roll-the-win32 zip files.  I'll try 
to wrap that up.

ITMT, I'll put together this .zip in a bit and try rolling a 2.0.36 binary 
for release.
No later than tommorow morning.

>Could we document this somewhere?

I'll try to recall where I left things and pick them back up where I left off.

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c

[Brian Havard]

On 6 May 2002 21:00:50 -, [EMAIL PROTECTED] wrote:

>aaron   02/05/06 14:00:50
>
>  Modified:modules/mappers mod_rewrite.c
>  Log:
>  Protect us from this dirty hack that gets SysV Sem working again
>  on mod_rewrite. This, along with the original usage of a unix-only
>  function in mod_rewrite, is a temporary stopgap measure designed only
>  to workaround limitations in APR's handling of permission attributes.
>  It shall be removed as soon as that interface is improved.
>  
>  Revision  ChangesPath
>  1.112 +4 -0  httpd-2.0/modules/mappers/mod_rewrite.c
>  
>  Index: mod_rewrite.c
>  ===
>  RCS file: /home/cvs/httpd-2.0/modules/mappers/mod_rewrite.c,v
>  retrieving revision 1.111
>  retrieving revision 1.112
>  diff -u -r1.111 -r1.112
>  --- mod_rewrite.c6 May 2002 18:24:03 -   1.111
>  +++ mod_rewrite.c6 May 2002 21:00:50 -   1.112
>  @@ -964,6 +964,7 @@
>   return HTTP_INTERNAL_SERVER_ERROR;
>   }
>   
>  +#if !defined(OS2) && !defined(WIN32) && !defined(BEOS)  && !defined(NETWARE)
>   rv = unixd_set_global_mutex_perms(rewrite_log_lock);

Shouldn't we #define UNIX or some other feature macro somewhere so we don't
have to use such a long winded expressions that must all be adjusted every
time a new non-unix platform is added?

-- 
 __
 |  Brian Havard |  "He is not the messiah!   |
 |  [EMAIL PROTECTED]  |  He's a very naughty boy!" - Life of Brian |
 --

call for votes: preventing mmap of file buckets

[Brian Pane]

For performance reasons, it's sometimes preferable to read a file
instead of mmap'ing it.  Read scales better on some multiprocessor
systems, for example, and read appears to require fewer NFS ops than
mmap if the file is on an NFS-mounted filesystem.

Here's my proposal for changing the httpd and apr-util to support
run-time selection of mmap:

apr-util changes:
  - Add "apr_bucket_file_allow_mmap(int)" to allow apps to enable or
disable mmap on a per-file basis.  (Default is to allow mmap,for
backward compatibility.)

httpd changes:
  - Add a new config directive "EnableMMAP (on|off)", default=on.
  - In the core handler, after creating a file bucket, call
apr_bucket_file_allow_mmap(false) if mmap has been disabled for this 
dir.

--Brian

Re: Fixing BrowserMatch: multi-string search algorithms?

[Scott Lamb]

Scott Lamb wrote:
> I suggest creating a new regular expression which is each of the ones in 
> the list separated by '|'. It will use regular expression alternatives 
> to match multiple patterns with a single state machine. And add almost 
> no code.

Oops, I forgot to mention that I think this is possible only where you 
are doing the same thing with each match. I.e., these three in the 
standard config could be combined in this way:

 BrowserMatch "RealPlayer 4\.0" force-response-1.0
 BrowserMatch "Java/1\.0" force-response-1.0
 BrowserMatch "JDK/1\.0" force-response-1.0

but not with anything that does something other than just 
force-response-1.0.

-- 
Scott Lamb

Re: mod_userdir/mod_alias

[Joshua Slive]

Justin Erenkrantz wrote:

> I agree - I think it's reversed.  +1 on your patch in Bugzilla.
> 
> Feel free to commit it, if you wish.  It'd be nice if your patch in
> bugzilla was unified diff format.  diff -u.  =)  -- justin

Woops.  I lost my .cvsrc on one machine.  Anyway, thanks for the review. 
  I just committed.

Joshua.

Re: Fixing BrowserMatch: multi-string search algorithms?

[Scott Lamb]

Brian Pane wrote:
> Given a set of n BrowserMatch directives (as in,
> for example, the default config that we ship with
> httpd-2.0), mod_setenvif will do O(n) regex comparisons
> on every request.
> 
> I want to fix that.

[...]

> Does anyone know of an efficient algorithm for comparing
> an input string against a set of multiple pattern strings?

I suggest creating a new regular expression which is each of the ones in 
the list separated by '|'. It will use regular expression alternatives 
to match multiple patterns with a single state machine. And add almost 
no code.

--
Scott Lamb

Re: mod_userdir/mod_alias

[Justin Erenkrantz]

On Mon, May 06, 2002 at 02:32:52PM -0700, Joshua Slive wrote:
> Regarding this bug:
> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8841
> 
> Why is mod_alias exlicitly set to run after mod_userdir?  This has the
> effect that you can't use any Alias/ScriptAlias for ~ URLs.  You would
> most often want to do this in order to override mod_userdir for some
> special settings.  I can't image any time you would want mod_userdir to
> override mod_alias, which is what happens now.
> 
> It's been like this since Ben Laurie's original hook work, but I can't
> figure out the justification.

I agree - I think it's reversed.  +1 on your patch in Bugzilla.

Feel free to commit it, if you wish.  It'd be nice if your patch in
bugzilla was unified diff format.  diff -u.  =)  -- justin

mod_userdir/mod_alias

[Joshua Slive]

Regarding this bug:
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=8841

Why is mod_alias exlicitly set to run after mod_userdir?  This has the
effect that you can't use any Alias/ScriptAlias for ~ URLs.  You would
most often want to do this in order to override mod_userdir for some
special settings.  I can't image any time you would want mod_userdir to
override mod_alias, which is what happens now.

It's been like this since Ben Laurie's original hook work, but I can't
figure out the justification.

Joshua.

Re: Win32 Build Failure (mod_rewrite.c)

[Aaron Bannert]

> Yep, Aaron's working on a fix.

I just committed a fix, let me know if you have any problems.

-aaron

Re: Win32 Build Failure (mod_rewrite.c)

[Cliff Woolley]

On Mon, 6 May 2002, Jerry Baker wrote:

> Anyone aware of this?

Yep, Aaron's working on a fix.

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Win32 Build Failure (mod_rewrite.c)

[Jerry Baker]

Anyone aware of this?

-
Compiling...
mod_rewrite.c
D:\Apache\httpd-2.0\modules\mappers\mod_rewrite.c(967) : warning C4013: 
'unixd_set_global_mutex_perms' undefined; assuming extern returning int
Linking...
   Creating library Release/mod_rewrite.lib and object Release/mod_rewrite.exp
mod_rewrite.obj : error LNK2001: unresolved external symbol 
_unixd_set_global_mutex_perms
Release/mod_rewrite.so : fatal error LNK1120: 1 unresolved externals
Error executing link.exe.

Apache.exe - 2 error(s), 1 warning(s)
-

-- 
Jerry Baker

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.cmod_rewrite.h

[Brad Nicholes]

I understand that this is just a temporary fix and I am all for
whatever gets the code compiling again.  I guess I'm just suggesting
that a cleaner  temporary fix (if there is such a thing) would be to
move both unixd_set_proc_mutex_perms() and
unixd_set_global_mutex_perms() into the APR mutex code as
apr_os_unixd_set_proc_mutex_perms() and
apr_os_unixd_set_global_mutex_perms() since these functions perform APR
kinds of stuff.  Then use an APR_HAVE_XXX #define to #ifdef out the call
in the HTTPD code for the platforms that don't support this
functionality yet.  Anyway, whatever works, works for me.

Brad

Brad Nicholes
Senior Software Engineer
Novell, Inc., a leading provider of Net business solutions
http://www.novell.com 

>>> [EMAIL PROTECTED] Monday, May 06, 2002 2:17:54 PM >>>
On Mon, May 06, 2002 at 02:09:59PM -0600, Brad Nicholes wrote:
> That would work but it doesn't really seem like the right thing
to
> do.  APR.h already has some #defines for other situations such as
> APR_USE_SYSVSEM_SERIALIZE.  Maybe if that #define doesn't make
since,
> another one should be added to platform versions of APR.h and then
used
> wherever SysV needs it.

I'm really just treating this as a stopgap until we figure out a
long-term solution for the problem of describing access permissions
across platforms. FWIW, the function itself already checks
APR_HAS_SYSVSEM_SERIALIZE internally.

-aaron

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h

[Aaron Bannert]

On Mon, May 06, 2002 at 02:09:59PM -0600, Brad Nicholes wrote:
> That would work but it doesn't really seem like the right thing to
> do.  APR.h already has some #defines for other situations such as
> APR_USE_SYSVSEM_SERIALIZE.  Maybe if that #define doesn't make since,
> another one should be added to platform versions of APR.h and then used
> wherever SysV needs it.

I'm really just treating this as a stopgap until we figure out a
long-term solution for the problem of describing access permissions
across platforms. FWIW, the function itself already checks
APR_HAS_SYSVSEM_SERIALIZE internally.

-aaron

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.cmod_rewrite.h

[Brad Nicholes]

That would work but it doesn't really seem like the right thing to
do.  APR.h already has some #defines for other situations such as
APR_USE_SYSVSEM_SERIALIZE.  Maybe if that #define doesn't make since,
another one should be added to platform versions of APR.h and then used
wherever SysV needs it.

Brad

Brad Nicholes
Senior Software Engineer
Novell, Inc., a leading provider of Net business solutions
http://www.novell.com 

>>> [EMAIL PROTECTED] Monday, May 06, 2002 2:04:32 PM >>>
On Mon, May 06, 2002 at 01:49:36PM -0600, Brad Nicholes wrote:
>   Why are we calling a Unix specific function in the middle of a
module?
>  Since unixd_set_global_mutex_perms() doesn't exist on NetWare,
> mod_rewrite won't compile.  Should this be handled in APR somewhere?

For now, would this take care of it? I know this is a stupid thing
to have, but until we have a better permissions setup in APR this
seems like the only way to get it working. The same thing needs to
happen anywhere we use an inherited proc or global mutex in a module
(mod_ssl is the next target).

-aaron



Index: modules/mappers/mod_rewrite.c
===
RCS file: /home/cvs/httpd-2.0/modules/mappers/mod_rewrite.c,v
retrieving revision 1.111
diff -u -u -r1.111 mod_rewrite.c
--- modules/mappers/mod_rewrite.c   6 May 2002 18:24:03
-   1.111
+++ modules/mappers/mod_rewrite.c   6 May 2002 20:02:08 -
@@ -964,6 +964,7 @@
 return HTTP_INTERNAL_SERVER_ERROR;
 }
 
+#if !defined(OS2) && !defined(WIN32) && !defined(BEOS)  &&
!defined(NETWARE)
 rv = unixd_set_global_mutex_perms(rewrite_log_lock);
 if (rv != APR_SUCCESS) {
 ap_log_error(APLOG_MARK, APLOG_CRIT, rv, s,
@@ -971,6 +972,7 @@
  "rewrite_log_lock; check User and Group
directives");
 return HTTP_INTERNAL_SERVER_ERROR;
 }
+#endif
 
 rv = rewritelock_create(s, p);
 if (rv != APR_SUCCESS) {
@@ -3327,6 +3329,7 @@
 return rc;
 }
 
+#if !defined(OS2) && !defined(WIN32) && !defined(BEOS)  &&
!defined(NETWARE)
 rc = unixd_set_global_mutex_perms(rewrite_mapr_lock_acquire);
 if (rc != APR_SUCCESS) {
 ap_log_error(APLOG_MARK, APLOG_CRIT, rc, s,
@@ -3334,6 +3337,7 @@
  "on RewriteLock; check User and Group
directives");
 return rc;
 }
+#endif
 
 return APR_SUCCESS;
 }

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h

[Aaron Bannert]

On Mon, May 06, 2002 at 01:49:36PM -0600, Brad Nicholes wrote:
>   Why are we calling a Unix specific function in the middle of a module?
>  Since unixd_set_global_mutex_perms() doesn't exist on NetWare,
> mod_rewrite won't compile.  Should this be handled in APR somewhere?

For now, would this take care of it? I know this is a stupid thing
to have, but until we have a better permissions setup in APR this
seems like the only way to get it working. The same thing needs to
happen anywhere we use an inherited proc or global mutex in a module
(mod_ssl is the next target).

-aaron



Index: modules/mappers/mod_rewrite.c
===
RCS file: /home/cvs/httpd-2.0/modules/mappers/mod_rewrite.c,v
retrieving revision 1.111
diff -u -u -r1.111 mod_rewrite.c
--- modules/mappers/mod_rewrite.c   6 May 2002 18:24:03 -   1.111
+++ modules/mappers/mod_rewrite.c   6 May 2002 20:02:08 -
@@ -964,6 +964,7 @@
 return HTTP_INTERNAL_SERVER_ERROR;
 }
 
+#if !defined(OS2) && !defined(WIN32) && !defined(BEOS)  && !defined(NETWARE)
 rv = unixd_set_global_mutex_perms(rewrite_log_lock);
 if (rv != APR_SUCCESS) {
 ap_log_error(APLOG_MARK, APLOG_CRIT, rv, s,
@@ -971,6 +972,7 @@
  "rewrite_log_lock; check User and Group directives");
 return HTTP_INTERNAL_SERVER_ERROR;
 }
+#endif
 
 rv = rewritelock_create(s, p);
 if (rv != APR_SUCCESS) {
@@ -3327,6 +3329,7 @@
 return rc;
 }
 
+#if !defined(OS2) && !defined(WIN32) && !defined(BEOS)  && !defined(NETWARE)
 rc = unixd_set_global_mutex_perms(rewrite_mapr_lock_acquire);
 if (rc != APR_SUCCESS) {
 ap_log_error(APLOG_MARK, APLOG_CRIT, rc, s,
@@ -3334,6 +3337,7 @@
  "on RewriteLock; check User and Group directives");
 return rc;
 }
+#endif
 
 return APR_SUCCESS;
 }

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h

[Aaron Bannert]

On Mon, May 06, 2002 at 01:49:36PM -0600, Brad Nicholes wrote:
>   Why are we calling a Unix specific function in the middle of a module?
>  Since unixd_set_global_mutex_perms() doesn't exist on NetWare,
> mod_rewrite won't compile.  Should this be handled in APR somewhere?

Hmm...good point. What #ifdef protection should I use?

This *should* be handled by APR, but I think that will take a lot more
work and for now SysV Sem is broken whenever we use setuid(). We need to
abstract out some concept of permissions that works across all platforms
and can be used for things like this. In this case, it would be useful
to perscribe to the apr_global_mutex.h interface that we want a lock
that is accessible to the httpd child processes. I have no idea how we
can do this across the different types of authn/authz frameworks on our
various platforms.

-aaron

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.cmod_rewrite.h

[Brad Nicholes]

  Why are we calling a Unix specific function in the middle of a module?
 Since unixd_set_global_mutex_perms() doesn't exist on NetWare,
mod_rewrite won't compile.  Should this be handled in APR somewhere?

Brad



Brad Nicholes
Senior Software Engineer
Novell, Inc., a leading provider of Net business solutions
http://www.novell.com 

>>> [EMAIL PROTECTED] Monday, May 06, 2002 12:24:04 PM >>>
aaron   02/05/06 11:24:04

  Modified:modules/mappers mod_rewrite.c mod_rewrite.h
  Log:
  Fix mod_rewrite hang when APR uses SysV Semaphores. Before we were
  silently failing when locking/unlocking the mutex, since httpd
  child processes didn't have permissions to access the root-created
  semaphore.
  
  PR: 8143
  
  Revision  ChangesPath
  1.111 +26 -6 httpd-2.0/modules/mappers/mod_rewrite.c
  
  Index: mod_rewrite.c
  ===
  RCS file: /home/cvs/httpd-2.0/modules/mappers/mod_rewrite.c,v
  retrieving revision 1.110
  retrieving revision 1.111
  diff -u -r1.110 -r1.111
  --- mod_rewrite.c 6 May 2002 18:04:53 -   1.110
  +++ mod_rewrite.c 6 May 2002 18:24:03 -   1.111
  @@ -964,7 +964,19 @@
   return HTTP_INTERNAL_SERVER_ERROR;
   }
   
  -rewritelock_create(s, p);
  +rv = unixd_set_global_mutex_perms(rewrite_log_lock);
  +if (rv != APR_SUCCESS) {
  +ap_log_error(APLOG_MARK, APLOG_CRIT, rv, s,
  + "mod_rewrite: Could not set permissions on "
  + "rewrite_log_lock; check User and Group
directives");
  +return HTTP_INTERNAL_SERVER_ERROR;
  +}
  +
  +rv = rewritelock_create(s, p);
  +if (rv != APR_SUCCESS) {
  +return HTTP_INTERNAL_SERVER_ERROR;
  +}
  +
   apr_pool_cleanup_register(p, (void *)s, rewritelock_remove,
apr_pool_cleanup_null);
   
   /* step through the servers and
  @@ -3296,26 +3308,34 @@
   
   #define REWRITELOCK_MODE ( APR_UREAD | APR_UWRITE | APR_GREAD |
APR_WREAD )
   
  -static void rewritelock_create(server_rec *s, apr_pool_t *p)
  +static apr_status_t rewritelock_create(server_rec *s, apr_pool_t
*p)
   {
   apr_status_t rc;
   
   /* only operate if a lockfile is used */
   if (lockname == NULL || *(lockname) == '\0') {
  -return;
  +return APR_EINVAL;
   }
   
   /* create the lockfile */
   rc = apr_global_mutex_create(&rewrite_mapr_lock_acquire,
lockname,
APR_LOCK_DEFAULT, p);
   if (rc != APR_SUCCESS) {
  -ap_log_error(APLOG_MARK, APLOG_ERR, rc, s,
  +ap_log_error(APLOG_MARK, APLOG_CRIT, rc, s,
"mod_rewrite: Parent could not create
RewriteLock "
"file %s", lockname);
  -exit(1);
  +return rc;
   }
   
  -return;
  +rc = unixd_set_global_mutex_perms(rewrite_mapr_lock_acquire);
  +if (rc != APR_SUCCESS) {
  +ap_log_error(APLOG_MARK, APLOG_CRIT, rc, s,
  + "mod_rewrite: Parent could not set permissions
"
  + "on RewriteLock; check User and Group
directives");
  +return rc;
  +}
  +
  +return APR_SUCCESS;
   }
   
   static apr_status_t rewritelock_remove(void *data)
  
  
  
  1.32  +1 -1  httpd-2.0/modules/mappers/mod_rewrite.h
  
  Index: mod_rewrite.h
  ===
  RCS file: /home/cvs/httpd-2.0/modules/mappers/mod_rewrite.h,v
  retrieving revision 1.31
  retrieving revision 1.32
  diff -u -r1.31 -r1.32
  --- mod_rewrite.h 6 May 2002 03:10:24 -   1.31
  +++ mod_rewrite.h 6 May 2002 18:24:03 -   1.32
  @@ -451,7 +451,7 @@
   static char *current_logtime(request_rec *r);
   
   /* rewriting lockfile support */
  -static void rewritelock_create(server_rec *s, apr_pool_t *p);
  +static apr_status_t rewritelock_create(server_rec *s, apr_pool_t
*p);
   static apr_status_t rewritelock_remove(void *data);
   
   /* program map support */
  
  
  

2.0.36 .zip files WAS, RE: 2.0.36 release

[Sander Striker]

Hi,

Is there any special voodoo to be performed to create the apache release zip files?

Could we document this somewhere?

Sander

Re: cvs commit: httpd-2.0/os/unix unixd.c unixd.h

[Aaron Bannert]

On Mon, May 06, 2002 at 06:19:53PM -, [EMAIL PROTECTED] wrote:
> aaron   02/05/06 11:19:53
> 
>   Modified:os/unix  unixd.c unixd.h
>   Log:
>   Add unixd_set_global_mutex_perms so we can set permissions on things like
>   SysV Semaphores in the core and modules.

I checked that this compiles whether APR_PROCESS_LOCK_IS_GLOBAL is defined
or not, but only really tested when it is not defined. If you're on one
of those other systems, please holler if it's broken.

-aaron

Re: 2.0.36 release

[Pier Fumagalli]

"Cliff Woolley" <[EMAIL PROTECTED]> wrote:

> On Mon, 6 May 2002, Pier Fumagalli wrote:
> 
>> No segfaults on worker un nagoya whatsoever...
> 
> Great!  Thanks for testing it out.

Was about time that we updated HTTPd on there... And Nagoya is the perfect
place to test out multi-threaded stuff (6 procs on Solaris 8)

>> The only thing "nagging" me is:
>> [Mon May 06 00:01:15 2002] [warn] child process 18743 still did not exit,
>> sending a SIGTERM
>> Not always processes die nicely...
> 
> Yeah, that's a known nastiness with 2.0.36's worker.  At least it shuts
> down, though.  We'll clean it up for 2.0.37.

Was also in 2.0.35... It's just a PITA because you have to wait 30 secs
before doing a startup if one of the children didn't die...

--
I think that it's extremely foolish to name a server after the current U.S.
President. B.W. Fitzpatrick

Re: proxy_cache handles "304 Not Modified" incorrectly

[Joshua Slive]

On Mon, 6 May 2002, Cliff Woolley wrote:

> On Mon, 6 May 2002, Joshua Slive wrote:
>
> > See:
> > http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7847
> > At most, we could do a browsermatch env-variable for this, but we
> > can't just ignore the header.
>
> I don't see why not.  The RFC says it's invalid IIRC, so why can't we
> just ignore it?  "Be lenient in what you accept."

Hmmm One quote is


   If a cache uses a received 304 response to update a cache entry, the
   cache MUST update the entry to reflect any new field values given in
   the response.


So one possibility would be to simply invalidate the cache and make an
unconditional request.  This would, of course, kill performance.

Joshua.

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c

[Cliff Woolley]

On 6 May 2002, Jeff Trawick wrote:

> IBM's native compiler for AIX failed the compile
> HP's native compiler for HP-UX issued a warning

As if I couldn't have guessed.  ;)  Thanks again.

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c

[Jeff Trawick]

Cliff Woolley <[EMAIL PROTECTED]> writes:

> On 6 May 2002 [EMAIL PROTECTED] wrote:
> 
> > trawick 02/05/06 07:02:50
> >
> >   Modified:modules/mappers mod_rewrite.c
> >   Log:
> >   fix the type of the pointer returned by the hash lookup
> 
> Thanks for the fix.  Just out of curiosity, which compiler choked on
> casting a (void *) as a function pointer implicitly?

IBM's native compiler for AIX failed the compile
HP's native compiler for HP-UX issued a warning

ISTR that IBM's native compiler for OS/390 and the native compiler for
Tru64 will both fail the compile with such an error

(this is with the compiler settings normally used for Apache)

-- 
Jeff Trawick | [EMAIL PROTECTED]
Born in Roswell... married an alien...

Re: proxy_cache handles "304 Not Modified" incorrectly

[Cliff Woolley]

On Mon, 6 May 2002, Joshua Slive wrote:

> See:
> http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7847
> At most, we could do a browsermatch env-variable for this, but we
> can't just ignore the header.

I don't see why not.  The RFC says it's invalid IIRC, so why can't we
just ignore it?  "Be lenient in what you accept."

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: 2.0.36 release

[Cliff Woolley]

On Mon, 6 May 2002, Pier Fumagalli wrote:

> No segfaults on worker un nagoya whatsoever...

Great!  Thanks for testing it out.

> The only thing "nagging" me is:
> [Mon May 06 00:01:15 2002] [warn] child process 18743 still did not exit,
> sending a SIGTERM
> Not always processes die nicely...


Yeah, that's a known nastiness with 2.0.36's worker.  At least it shuts
down, though.  We'll clean it up for 2.0.37.

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: [Patch] Concept: have MPM identify itself in Server header

[Austin Gonyou]

Sorry for chiming in after the weekend on this. Just to clarify my
statement earlier though, I didn't mean that this info should be
presented to the client, but rather made available, say in the
server-signature string or some such thing. I've not seen it in there.
It would be nice to be able to identify the type of MPM used, mainly
like in signature. 

Not sure how *safe* it is in general though.

On Fri, 2002-05-03 at 17:24, Cliff Woolley wrote:
> On Fri, 3 May 2002, Rasmus Lerdorf wrote:
> 
> > Ok, but where should this information go then?  Apache has definitely
> > benefitted by having this information available.  Some sort of
> > X-SERVER-INFO: header then?
> 
> What I meant was I don't think the MPM should be announced to the
> client.
> What possible benefit could there be to doing that?
> 
> --Cliff
> 
> --
>Cliff Woolley
>[EMAIL PROTECTED]
>Charlottesville, VA
> 
-- 
Austin Gonyou
Systems Architect, CCNA
Coremetrics, Inc.
Phone: 512-698-7250
email: [EMAIL PROTECTED]

"It is the part of a good shepherd to shear his flock, not to skin it."
Latin Proverb



signature.asc
Description: This is a digitally signed message part

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h(fwd)

[Cliff Woolley]

On Mon, 6 May 2002, Tahiry Ramanamampanoharana wrote:

> One thing I don't understand: why make ap_register_rewrite_mapfunc()
> static and then export it as an optional function? Why isn'it directly a
> public function provided by mod_rewrite?

Good question.  The answer is that if you didn't use an optional function
and some other module (or even the core) tried to extend mod_rewrite using
ap_register_rewrite_mapfunc() and mod_rewrite was not loaded, there would
be a link failure (possibly even at runtime if the other module is a DSO).
Using an optional function allows the other module to skip over the
extension gracefully.

Practical example:  consider mod_cgi and mod_include.  mod_include exports
the optional function ap_register_include_handler(), which allows other
modules to extend mod_include with their own SSI tag types.  mod_cgi does
this for the  tag.  If ap_register_include_handler() were
not an optional function, mod_cgi could not be loaded without mod_include
also being loaded, which is clearly not what you'd want, since mod_cgi is
quite functional without the SSI extension it happens to provide.

Make sense?

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: proxy_cache handles "304 Not Modified" incorrectly

[Joshua Slive]

On Mon, 6 May 2002, Zvi Har'El wrote:
>
> I have noticed today that when Microsoft-IIS/5.0 sends a "304 Not Modified"
> response to IMS GET request, it (incorrectly IMHO) sends with it a
> "Content-Length: 0" header.

See:
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=7847
At most, we could do a browsermatch env-variable for this, but we
can't just ignore the header.

Joshua.

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c

[Cliff Woolley]

On 6 May 2002 [EMAIL PROTECTED] wrote:

> trawick 02/05/06 07:02:50
>
>   Modified:modules/mappers mod_rewrite.c
>   Log:
>   fix the type of the pointer returned by the hash lookup

Thanks for the fix.  Just out of curiosity, which compiler choked on
casting a (void *) as a function pointer implicitly?  I didn't know there
was a problem with that.

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

proxy_cache handles "304 Not Modified" incorrectly

[Zvi Har'El]

Dear Apache developers,

I have noticed today that when Microsoft-IIS/5.0 sends a "304 Not Modified"
response to IMS GET request, it (incorrectly IMHO) sends with it a
"Content-Length: 0" header. This is (again IMHO) a bug in IIS, but because of
that, IMS GET request with "Pragma: no-cache" headers are not handled
correctly: When the cache revalidation is done, in ap_proxy_cache_update(), the
content length is reset (in line 1481 in apache_1.3.24, line 1508 in the
current snapshot of apache_1.3.25-dev, file proxy_cache.c), before the status
is checked to be HTTP_NOT_MODIFIED (in line 1513 or 1547 resp). This makes the
erroneous Content-Length: 0 header replace the real length in the cache file
headers, and the file is destroyed. Not this doesn't happen without the
"Pragma: no-cache" header. A possible fix would be do ignore the
"Content-Length" in the response in case of "304 Not Modified", but I think
that perhaps no header in such a response should replace the cached headers!

Best,

Zvi.


-- 
Dr. Zvi Har'El mailto:[EMAIL PROTECTED] Department of Mathematics
tel:+972-54-227607   Technion - Israel Institute of Technology
fax:+972-4-8324654 http://www.math.technion.ac.il/~rl/ Haifa 32000, ISRAEL
"If you can't say somethin' nice, don't say nothin' at all." -- Thumper (1942)
   Monday, 25 Iyyar 5762,  6 May 2002,  6:42PM

Re: 2.0.36 release

[Bill Stoddard]

+1

> 
> +1 for GA.  Let's get this bad boy out the door.  I'm sick of dealing with
> duplicate PRs on the same three bugs in 2.0.35.  :)
> 
> --Cliff
> 

Re: 2.0.36 release

[Pier Fumagalli]

"Jeff Trawick" <[EMAIL PROTECTED]> wrote:

> Cliff Woolley <[EMAIL PROTECTED]> writes:
> 
>> +1 for GA.  Let's get this bad boy out the door.
> 
> +1 from me too

No segfaults on worker un nagoya whatsoever... The only thing "nagging" me
is:

[Mon May 06 00:01:15 2002] [warn] child process 18743 still did not exit,
sending a SIGTERM
[Mon May 06 00:01:15 2002] [warn] child process 18744 still did not exit,
sending a SIGTERM
[Mon May 06 00:01:15 2002] [warn] child process 18744 still did not exit,
sending a SIGTERM
[Mon May 06 00:01:16 2002] [warn] child process 18744 still did not exit,
sending a SIGTERM
[Mon May 06 00:01:21 2002] [error] child process 18744 still did not exit,
sending a SIGKILL
[Mon May 06 00:01:37 2002] [notice] SIGHUP received.  Attempting to restart
[Mon May 06 00:01:38 2002] [notice] Apache/2.0.36 (Unix)
mod_webapp/1.2.0-dev configured -- resuming normal operations

Not always processes die nicely...

Pier

--
I think that it's extremely foolish to name a server after the current U.S.
President. B.W. Fitzpatrick

Re: 2.0.36 release

[Jeff Trawick]

Cliff Woolley <[EMAIL PROTECTED]> writes:

> +1 for GA.  Let's get this bad boy out the door.

+1 from me too

-- 
Jeff Trawick | [EMAIL PROTECTED]
Born in Roswell... married an alien...

Re: Self-contained modules?

[Pier Fumagalli]

"Justin Erenkrantz" <[EMAIL PROTECTED]> wrote:

> Got a bit of an interesting problem I encountered with Darwin and
> building PHP the other day.  I brought this up on the php-dev list
> and I'd like to consider what our options are to fixing this.
> 
> First off, Darwin requires that all modules be self-contained by
> default.  So, if we try to build with PHP with apxs, we will get
> symbol not defined errors for all of the apr and apr-util functions.
> 
> This can be turned off by -flat_namespace -undefined suppress - see
> <[EMAIL PROTECTED]>.  But, for certain reasons
> the PHP developers would prefer to use bundles instead - the
> approach Apple uses outlined in Sander's post above.  So, we have
> a few alternatives and I'd like to post them here to see what we
> think the preferred solution is before replying on php-dev.
> 
> 1) Tell PHP tough - they must use -flat_namespace.  Problem solved.
> 2) Teach PHP about apr-config and apu-config so that they know
> where APR is and can link them when building the module.  I think
> the problem would be solved.
> 3) Have APXS expose apr and apr-util in a consistent manner so
> that it is possible to build self-contained libraries.
> 
> In email exchanges with a PHP developer (Marko?), he really didn't
> like option 1 or 2.  He would like to see #3.  Personally, I say
> #2, but I'd like to see what everyone else thinks.  -- justin

Hmm... AFAIK, building mod_webapp, libtool under darwin has several troubles
in linking against a static and a dynamic library (two at the same time)...
It simply doesn't want to pass the stupid -all_load parameter to the
compiler.

Under WebApp we solved it by not creating AR libraries, but simply linking
all the .lo files all together with APXS (see my latest patches on that),
and of course, the idiot won't build a dso with all symbols correctly
referenced if you don't build a library named libsomething.la...

Go figure, libtool sucks.

BTW, related to this, anyone tried to build mod_proxy under Darwin?

Pier




--
I think that it's extremely foolish to name a server after the current U.S.
President. B.W. Fitzpatrick

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h (fwd)

[Tahiry Ramanamampanoharana]

Cliff:

One thing I don't understand: why make ap_register_rewrite_mapfunc() static 
and then export it as an optional function? Why isn'it directly a public 
function provided by mod_rewrite?

Tahiry

>From: Cliff Woolley <[EMAIL PROTECTED]>
>Reply-To: [EMAIL PROTECTED]
>To: Tahiry Ramanamampanoharana <[EMAIL PROTECTED]>, 
><[EMAIL PROTECTED]>
>Subject: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h 
>(fwd)
>Date: Sun, 5 May 2002 23:14:05 -0400 (EDT)
>MIME-Version: 1.0
>Received: from [63.251.56.142] by hotmail.com (3.2) with ESMTP id 
>MHotMailBE9F403A00094004311E3FFB388EF3CB0; Sun, 05 May 2002 20:15:07 -0700
>Received: (qmail 5010 invoked by uid 500); 6 May 2002 03:14:50 -
>Received: (qmail 4997 invoked from network); 6 May 2002 03:14:50 -
>From dev-return-29426-nomentsoa Sun, 05 May 2002 20:15:54 -0700
>Mailing-List: contact [EMAIL PROTECTED]; run by ezmlm
>Precedence: bulk
>list-help: 
>list-unsubscribe: 
>list-post: 
>Delivered-To: mailing list [EMAIL PROTECTED]
>X-X-Sender: [EMAIL PROTECTED]
>Message-ID: 
><[EMAIL PROTECTED]>
>X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N
>
>
>Tahiry:
>
>I took your patch, played with it a bit, and here's what I came up with.
>Rather than using the optional function to pull IN the third-party
>function to mod_rewrite, it seems better to have mod_rewrite *export* an
>optional function which is a registration function.  Support for multiple
>extenders is more straightforward that way.  This is also how mod_include
>does it, by the way.
>
>Thanks for the idea!
>
>--Cliff
>
>


_
MSN Photos is the easiest way to share and print your photos: 
http://photos.msn.com/support/worldwide.aspx

Re: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h (fwd)

[Tahiry Ramanamampanoharana]

Cliff:

Thanks a lot for taking care of it. I only have internet access at work so I 
didn't have access to the forum this week-end :(. I am actually happy that 
you changed it so I could learn something out of it :)

Thanks

Tahiry


>From: Cliff Woolley <[EMAIL PROTECTED]>
>Reply-To: [EMAIL PROTECTED]
>To: Tahiry Ramanamampanoharana <[EMAIL PROTECTED]>, 
><[EMAIL PROTECTED]>
>Subject: cvs commit: httpd-2.0/modules/mappers mod_rewrite.c mod_rewrite.h 
>(fwd)
>Date: Sun, 5 May 2002 23:14:05 -0400 (EDT)
>MIME-Version: 1.0
>Received: from [63.251.56.142] by hotmail.com (3.2) with ESMTP id 
>MHotMailBE9F403A00094004311E3FFB388EF3CB0; Sun, 05 May 2002 20:15:07 -0700
>Received: (qmail 5010 invoked by uid 500); 6 May 2002 03:14:50 -
>Received: (qmail 4997 invoked from network); 6 May 2002 03:14:50 -
>From dev-return-29426-nomentsoa Sun, 05 May 2002 20:15:54 -0700
>Mailing-List: contact [EMAIL PROTECTED]; run by ezmlm
>Precedence: bulk
>list-help: 
>list-unsubscribe: 
>list-post: 
>Delivered-To: mailing list [EMAIL PROTECTED]
>X-X-Sender: [EMAIL PROTECTED]
>Message-ID: 
><[EMAIL PROTECTED]>
>X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N
>
>
>Tahiry:
>
>I took your patch, played with it a bit, and here's what I came up with.
>Rather than using the optional function to pull IN the third-party
>function to mod_rewrite, it seems better to have mod_rewrite *export* an
>optional function which is a registration function.  Support for multiple
>extenders is more straightforward that way.  This is also how mod_include
>does it, by the way.
>
>Thanks for the idea!
>
>--Cliff
>
>


_
Join the world’s largest e-mail service with MSN Hotmail. 
http://www.hotmail.com

Re: Stripping Content-Length headers

[Graham Leggett]

Joshua Slive wrote:

> As far as I can tell, the C-L is still trashed.  Doug just removed a
> "flush".  See
> /* In order for ap_set_keepalive to work properly, we can NOT
>  * have any length information stored in the output headers.
>  */
> apr_table_unset(r->headers_out,"Transfer-Encoding");
> apr_table_unset(r->headers_out,"Content-Length");
> 
> in ap_proxy_http_process_response in proxy_http.c.

In the case of the "transfer-encoding", surely this header should be
removed by the dechunk filter? (or whatever bit of the core was doing
dechunking, as the dechunk filter was absorbed into another filter).

In other words, proxy shouldn't be fiddling with headers, rather the
filters should be fiddling with headers instead.

This looks wrong to me.

Regards,
Graham
-- 
-
[EMAIL PROTECTED]"There's a moon
over Bourbon Street
tonight..."


smime.p7s
Description: S/MIME Cryptographic Signature

Re: Stripping Content-Length headers

[Graham Leggett]

Justin Erenkrantz wrote:

> Here's my question: why is C-L filter even getting involved if there
> is a C-L header?  Why shouldn't it just get out of the way if there
> is a C-L already present?  Why do the duplication?  Should we assume
> that any module that changes the content is smart enough to unset it?
> Perhaps we can't do that - a module might change the content but not
> unset C-L?  -- justin

I would argue that a module that changes the content-length, but does
not remove the CL header is broken.

Regards,
Graham
-- 
-
[EMAIL PROTECTED]"There's a moon
over Bourbon Street
tonight..."


smime.p7s
Description: S/MIME Cryptographic Signature

Re: cvs commit: httpd-2.0/include ap_mmn.h

[Cliff Woolley]

On 6 May 2002 [EMAIL PROTECTED] wrote:

> jerenkrantz02/05/06 01:21:10
>
>   Modified:include  ap_mmn.h
>   Log:
>   Removing a field in a core structure (r->boundary) merits a MMN bump,
>   unfortunately.  They got 2 GAs out of the old MMN.
>
>   Reviewed by:Cliff Woolley

If people end up objecting to this (I'm sure they will :-P), we *could*
just leave a dummy field in the middle of the request_rec.  It seems
awfully silly to me to start leaving cruft in the structures this early in
the 2.0 cycle, though.  I suggest that we leave this bump in there for
now, and if other changes come about that also could take advantage of an
MMN bump, we won't feel too bad about it.  If we get to release time and
this was the only change, we can consider sticking in a dummy var as a
*temporary* measure at that time.

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: cvs commit: httpd-2.0/modules/http http_protocol.c

[Cliff Woolley]

On Mon, 6 May 2002, Justin Erenkrantz wrote:

> You know, I *thought* that macro existed.  =)  Should have checked
> first.  Thanks for catching this!  -- justin

No prob.  Functionally equivalent in every way, just ... prettier.  :)
There are probably many places in the code that still use
APR_BRIGADE_CONCAT in a backwards way (_PREPEND is a very recent
addition)...  I'll do a grep later and weed out any remaining cases.

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: cvs commit: httpd-2.0/modules/http http_protocol.c

[Justin Erenkrantz]

On Mon, May 06, 2002 at 08:14:53AM -, [EMAIL PROTECTED] wrote:
>   +/* Prepend any earlier saved brigades. */
>   +APR_BRIGADE_PREPEND(bb, ctx->bb);

You know, I *thought* that macro existed.  =)  Should have checked
first.  Thanks for catching this!  -- justin

Re: cvs commit: httpd-2.0/modules/http http_protocol.c

[Cliff Woolley]

On Mon, 6 May 2002, Cliff Woolley wrote:

> >   +ctx->boundary = apr_psprintf(r->pool, "%qx%lx",
> >   + r->request_time, (long) getpid());
> Is %qx portable?  Shouldn't that be APR_TIME_T_FMT?

Oh duh, this is apr_psprintf(), not printf().  Nevermind.  ;)

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: cvs commit: httpd-2.0/modules/http http_protocol.c

[Cliff Woolley]

On 6 May 2002 [EMAIL PROTECTED] wrote:

>   - Remove r->boundary since it is possible to have this self-contained in
> boundary's ctx.  (May require MMN bump?)

That would definitely require an MMN bump.

>   +ctx->boundary = apr_psprintf(r->pool, "%qx%lx",
>   + r->request_time, (long) getpid());

Is %qx portable?  Shouldn't that be APR_TIME_T_FMT?

>   -if (ctx->bb) {
>   +if (!APR_BRIGADE_EMPTY(ctx->bb)) {
>APR_BRIGADE_CONCAT(ctx->bb, bb);
>bb = ctx->bb;
>   -ctx->bb = NULL; /* ### strictly necessary? call brigade_destroy? */
>}
>   +apr_brigade_destroy(ctx->bb);

That apr_brigade_destroy() call is unnecessary.  ctx->bb is allocated from
a pool, and you know it's empty because you just emptied it.  There's
nothing to destroy.  Ie, the answer to the ### comment was "no, and no."
:)

PS: If you're looking for a good way to test this stuff, run a big PDF
through it and view with Acrobat Reader.  If Acrobat locks up, the
byterange filter is broken.  ;)

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: cvs commit: httpd-2.0/modules/http http_protocol.c

[Justin Erenkrantz]

On Mon, May 06, 2002 at 07:43:40AM -, [EMAIL PROTECTED] wrote:
>   - Remove r->boundary since it is possible to have this self-contained in
> boundary's ctx.  (May require MMN bump?)

I believe removing a field from request_rec requires a MMN bump.  Am I
right?  -- justin

Re: dbm rewritemaps

[Cliff Woolley]

On Mon, 6 May 2002, Eli Marmor wrote:

> I thought that apr_dbm is a replacement of SDBM (this is written in too
> many places and documents...).

It did used to be that way.  Not anymore.  :)

Thanks,
Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: dbm rewritemaps

[Eli Marmor]

Cliff Woolley wrote:

> > I am also not sure regarding the alternative DBs of SDBM: Either we
> > should make apr_dbm an abstraction layer that will support them too
> > (may cause licensing problems), or we should ignore them.
> 
> It already is such a layer, if I understand what you're saying.  apr_dbm
> lets you choose which flavor you want: sdbm, gdbm, or db.  Of the ones it
> detects on your system, it picks one as a default in case you just don't
> care.  I have all three, and it picks SDBM as the default on my system.

Oops...
Sorry...
I thought that apr_dbm is a replacement of SDBM (this is written in too
many places and documents...).
Next time I should look at the source before writing...

-- 
Eli Marmor
[EMAIL PROTECTED]
CTO, Founder
Netmask (El-Mar) Internet Technologies Ltd.
__
Tel.:   +972-9-766-1020  8 Yad-Harutzim St.
Fax.:   +972-9-766-1314  P.O.B. 7004
Mobile: +972-50-23-7338  Kfar-Saba 44641, Israel

Re: dbm rewritemaps

[Cliff Woolley]

On Mon, 6 May 2002, Eli Marmor wrote:

> Do you mean INSTEAD of the "txt:"?

Um, no.  I'm just fixing the "dbm:" flavor.  "int:", "prg:", "rnd:", and
"txt:" would remain.  :)

> (by the way: use the APR-UTIL functions for it, rather than SDBM).

That's what I'm talking about.

> I am also not sure regarding the alternative DBs of SDBM: Either we
> should make apr_dbm an abstraction layer that will support them too
> (may cause licensing problems), or we should ignore them.

It already is such a layer, if I understand what you're saying.  apr_dbm
lets you choose which flavor you want: sdbm, gdbm, or db.  Of the ones it
detects on your system, it picks one as a default in case you just don't
care.  I have all three, and it picks SDBM as the default on my system.

My only concern is that someone will have an existing configuration that
uses RewriteMap ... dbm:... and will have a dbm database that's some other
flavor than the one APR-util picks by default (say, SDBM), and it will
stop working until they change their config to be more specific.

--Cliff

--
   Cliff Woolley
   [EMAIL PROTECTED]
   Charlottesville, VA

Re: dbm rewritemaps

[Eli Marmor]

Cliff Woolley wrote:

> I'm converting mod_rewrite to use apr_dbm_* instead of its homegrown
> flavor.

Do you mean INSTEAD of the "txt:"?
I'm afraid that it will break many files, compatibilities, and
existing installations.
In addition, TXT map files are easier for tests, debugging, learning,
etc.
Can't it be done IN ADDITION to "txt:" and not INSTEAD?

>  The only problem I see with it is that it will require a
> configuration change for the user to specify what kind of dbm they want.
> Right now, the RewriteMap is set up as follows:
> 
> RewriteMap mymap dbm:/path/to/dbm
> 
> I propose to change that as follows: dbm will mean "default", but the user
> can specify "db:", "gdbm:", or "sdbm:" instead of "dbm:" to get a specific
> flavor.

We can't adopt SDBM on one hand (apr_dbm), but default to DBM on the
other hand.
IF there is a default, it should be SDBM (by the way: use the APR-UTIL
functions for it, rather than SDBM).
I don't think that there should be any default. Alternatively, the
default should be "txt:".
I am also not sure regarding the alternative DBs of SDBM: Either we
should make apr_dbm an abstraction layer that will support them too
(may cause licensing problems), or we should ignore them.

But in any case, we can't adopt a format (SDBM) as a standard, and
later - ignore what we already built.

P.S. your initiative is great, and my notes are only regarding the
compatibility with the existing files on one hand, and with the
APR-UTIL library on the other hand.
-- 
Eli Marmor
[EMAIL PROTECTED]
CTO, Founder
Netmask (El-Mar) Internet Technologies Ltd.
__
Tel.:   +972-9-766-1020  8 Yad-Harutzim St.
Fax.:   +972-9-766-1314  P.O.B. 7004
Mobile: +972-50-23-7338  Kfar-Saba 44641, Israel

Re: dbm rewritemaps

[Justin Erenkrantz]

On Mon, May 06, 2002 at 03:10:32AM -0400, Cliff Woolley wrote:
> On Mon, 6 May 2002, Justin Erenkrantz wrote:
> 
> > Hint: look at apr_dbm_open_ex.  =)  -- justin
> 
> I know that, I've been looking at it for like an hour now.  :)  That's
> why I said "dbm" would get translated into "default" by mod_rewrite.  I
> almost explicitly said:
> 
> apr_dbm_open_ex(..., "default", ...)
> 
> in my last email.  Ugh.  ;]  You didn't answer my question though.  Is it
> acceptable for
> 
> RewriteMap mydbm dbm:/path/to/dbm
> 
> to mean apr_dbm_open_ex(..., "default", ...) ?

What I was thinking was:

RewriteMap mydbm default:/path/to/dbm

since apr_dbm_open_ex could handle that.  (The dbm->default mapping
makes a lot of sense to maintain backwards compatibility.)

But (as you just pointed out to me on IRC), RewriteMap can also take
txt, prg, etc.  So, having "default" may be ambiguous.  Your call if
you want to allow "default" as a type - I think it may be more hassle
to prevent default than it would be to just allow it.  -- justin