On Mon, Feb 18, 2002 at 09:17:19PM -0800, Ian Holsman wrote:
it doesn't do what it is supposed to do ALL the time.
for example.. take mod-status.
inside the handler it decides what type of content-type
the program will return.
bad module, fix module.
That's bogus. Consider
This works well with perfork and worker under Linux. I have a couple
of comments though:
1) There are some not infrequent cases I have run into where apache
needs to be killed (for unrelated reasons) and the shared memory
segment does not get cleaned up. When this happens, you
Bill Stoddard wrote:
Design and implementation of Apache 2.0 is nearing completion. Module
authors are encouraged to review the Apache 2.0 API and share any
concerns with the Apache development team at [EMAIL PROTECTED]
This is your best opportunity to ensure that your issues are
FWIW, my JRE_1 tag does *not* have this change. rbb and I agreed
that it wasn't necessary to have the signal-less worker MPM in any
beta as it doesn't affect modules. It also gives time to stablize
problems like these.
yep... roll on...
unclear what problem was actually being
On Tue, 12 Feb 2002, Ryan Bloom wrote:
This has historically been a Covalent hosted and driven project, as
a
service to the Apache community. Covalent has decided to open that
project up to the community to improve. To that end, sometime next
week, I will be setting up a machine
This is normal. Apache does two initializations, the first bootstraps
the server, and the second actually initializes it.
Ryan
I've noticed that the module init sometimes gets called twice. Is
this
something I'm doing wrong or is this normal?
--
Brian Akins
Systems Engineer III
CNN
This has historically been a Covalent hosted and driven project,
as
a
service to the Apache community. Covalent has decided to open
that
project up to the community to improve. To that end, sometime
next
week, I will be setting up a machine as a web server and CVS
server
to
be
Mads Toftum wrote:
On Tue, Feb 12, 2002 at 08:44:16PM +0100, Thomas Eibner wrote:
Are there any projects regarding modules.apache.org that will
require
development effort? I'd be interested in helping if that is the
case.
IMHO a cleanup of the current entries is much more needed than
On Wed, Feb 13, 2002 at 09:08:38AM -0800, Ian Holsman wrote:
even easier
a nightly cron job which does a HEAD on the URL link.
if it 404's x in a row it gets marked as old
if it 404's for a week/month it gets deleted
what are you planning to write this in JSP/perl/PHP ?
or a custom
FWIW, my JRE_1 tag does *not* have this change. rbb and I
agreed
that it wasn't necessary to have the signal-less worker MPM in
any
beta as it doesn't affect modules. It also gives time to
stablize
problems like these.
yep... roll on...
unclear what problem was
asked about
this project every few weeks for the last year or so. I wanted to let
everybody know that we are making progress and this project will be
available to hack on in about a week or two.
Please be patient.
Ryan
--
Ryan Bloom
, 2002 at 02:30:28PM -0500, Rodent of Unusual Size
wrote:
Ryan Bloom wrote:
That machine will be hosted by Covalent, and accounts on that
machine
will be given out as it makes sense. For the time being, I will
be
admin'ing the machine in my spare time, so things may take a while
Doesn't the race condition still exist? For example, I can delete the
object after the if but before the hash_set. You need a mutex, don't
you?
Ryan
-apr_hash_set(sconf-cacheht, obj-key, strlen(obj-key), NULL);
+/*
+ * RACE .. some one might have just deleted this
/apache_modules/mod_apachecon.c
The reason we did this, was because we removed the 'extra' directory
from underneath modules, and people were very much against putting it
back.
Ryan
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St
Ryan Bloom [EMAIL PROTECTED] writes:
--with-module takes two arguments with a ':' separating them.
The first argument is the module type, so that the configure script
knows where to copy the file. The second is the file to copy. So,
the
following will add mod_apachecon.c to my
Ryan Bloom [EMAIL PROTECTED] writes:
--with-module takes two arguments with a ':' separating them.
The first argument is the module type, so that the configure
script
knows where to copy the file. The second is the file to copy.
So,
the
following will add mod_apachecon.c
Justin Erenkrantz wrote:
On Sat, Feb 09, 2002 at 08:11:45AM -0800, Brian Behlendorf wrote:
Heh, anyone want to check this out?
Yeah, I see it too. Add:
ForceLanguagePriority Prefer Fallback
to daedalus's httpd.conf. I added it via .htaccess to docs-2.0
in
So we shouldn't just patch httpd.conf. We need to figure out what
is
going wrong.
I believe nothing is wrong.
I believe it is working as intended. Fix the config file, and
move on. I will attempt to commit a change that makes PREFER
and FALLBACK the default for ForceLangPriority in
Modified:.CHANGES
server/mpm/worker Makefile.in mpm.h worker.c
Added: server/mpm/worker pod.c pod.h
Log:
Remove all signal handling from the worker MPM's child processes.
Instead,
we use the pipe of death for all communication between parent and
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
-Original Message-
From: Bill Stoddard [mailto:[EMAIL PROTECTED]]
Sent: Saturday, February 09, 2002 7:12 AM
To: [EMAIL PROTECTED]
Subject: Window MPM debugging
FYI
Can we fix the service restart problem before we go making other
changes
to the MPM? I
see over the last few days there have been dozens of changes to the
code,
none of which
are directed at fixing the service restart problem. All this monkeying
around with the
code does is obscure
I disagree completely. The Windows MPM is fragile, because it is
almost
impossible to read. Bill has been cleaning it up so that multiple
people can easily modify the code and actually understand what is
happening.
Knowing how much time Bill has spent reading through that code, and
Why is this defined in the MPM? Shouldn't it exist outside of the
MPM? I'd like to let the MPM leave ap_scoreboard_fname as NULL if it
doesn't encounter a ScoreboardFile directive, so I can opt for
anonymous
shared memory as was discussed on here a week or two ago, but I'll
have to
change
On Wed, Feb 06, 2002 at 09:44:22AM -0500, Rodent of Unusual Size
wrote:
Let's see, then. Here are some test cases:
[...]
Does that seem about right?
Yes, with the addition that any Port directive is used to create the
canonical name in preference to the port from the Listen
ServerName MyServer.Com
Listen 1
Listen 2
Canonical name should be: MyServer.Com:port-used-by-the-request
I agree with all of them up through this last one. It's not that I
disagree with this, just that I'd be perfectly happy if the Canonical
name used 1 or 2 regardless
On Wed, Feb 06, 2002 at 09:44:22AM -0500, Rodent of Unusual Size
wrote:
ServerName MyServer.Com
Listen 1
Listen 2
Canonical name should be: MyServer.Com:port-used-by-the-request
It's this case that makes me think we should just rename ServerName to
something like
ServerName MyServer.Com
Listen 1
Listen 2
Canonical name should be:
MyServer.Com:port-used-by-the-request
I agree with all of them up through this last one. It's not that I
disagree with this, just that I'd be perfectly happy if the
Canonical
name used 1 or
Why do your web servers rely on DNS at all? If your config file doesn't
specify HostNameLookup on, and you always use dotted quad notation, the
server shouldn't ever use the DNS server. If it does, I would consider
that a showstopper.
Ryan
--
Ryan
*c)
+static int perchild_pre_connection(conn_rec *c, void *csd)
{
ap_add_input_filter(PERCHILD_BUFFER, NULL, NULL, c);
return OK;
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
-Original Message-
From: Rodent of Unusual Size [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, February 05, 2002 1:14 PM
To: [EMAIL PROTECTED]
Subject: Re: UseCanonicalName considered harmful
Ryan Bloom wrote:
Shouldn't we fix the canonicalisation then? If you have
configured
-Original Message-
From: William A. Rowe, Jr. [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, February 05, 2002 1:18 PM
To: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: Re: Any vetos on releasing 1.3.24...
From: Jim Jagielski [EMAIL PROTECTED]
Sent: Tuesday, February 05, 2002 3:09 PM
From: Bill Stoddard [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, February 05, 2002 1:44 PM
To: [EMAIL PROTECTED]
Subject: Re: [PATCH] Remove the insert_transport_filters hook
This approach has an aesthetic problem that annoys me. If
mod_yadda
wants
to insert its
own replacement of
From: Ryan Bloom [mailto:[EMAIL PROTECTED]]
From: Rodent of Unusual Size [mailto:[EMAIL PROTECTED]]
Ryan Bloom wrote:
Shouldn't we fix the canonicalisation then? If you have
configured your server so that it can't be reached through
the canonical name, then you have
I had 1.3.22 working on one of my XP boxes a few months ago. It is a
real pity that I now have multiple XP machines, and just one lonely
Linux VM.
:-(
Ryan
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED
Ahhh, I hadn't tried that yet.
Ryan
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
-Original Message-
From: Bill Stoddard [mailto:[EMAIL PROTECTED]]
Sent: Monday
Modified:.CHANGES STATUS
server/mpm/perchild perchild.c
server/mpm/prefork prefork.c
server/mpm/worker worker.c
Log:
Not being able to bind to a socket is a fatal error. This makes all
MPMs treat it as such. We now print a
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
Since it sounds like Win32 is busted (I'd say it'd be fine to
say that .31 is beta but doesn't work on Win32, but wrowe doesn't
like that),
IMNSHO, .31 should be released as a beta, with the understanding that it
doesn't work on Win32. That is a PITA, but it wouldn't be the first
time that a
I finally had time to review this patch, and I have some comments.
Log:
This patch restores most of Ryan's patch (11/12/2001) to remove the
client_socket from the conn_rec. Diffs from Ryan's patch include:
- rename the create_connection hook to install_transport_filters
- move
From: Justin Erenkrantz [mailto:[EMAIL PROTECTED]]
Ian was hesitant to bump to 2.0.32 because he was under the
impression that it was not permitted to bump so close to a previous
tag. He was the RM, so it was his call.
This argument has been had befor (ad naseum), but...
This is
From: Jim Jagielski [mailto:[EMAIL PROTECTED]]
Ryan Bloom wrote:
Not long after the current tag/roll procedure was developed, we had
this
same situation, and Roy himself agreed that rolling more than once a
week discouraged people from testing the tarballs.
Not sure what this Roy
According to Ryan Bloom:
Mod_ldap was in 2.0, but the group decided to remove it. The docs
should be removed as well. Instead of losing the code and docs, a
new httpd sub-project was created, and the docs should be moved
there, the code has been there for a while already.
The sub
Bill Stoddard wrote:
On Fri, Feb 01, 2002 at 02:45:56PM -0700, Brad Nicholes wrote:
httpd-2.0.31 does not build on NetWare because of a screwed up
#ifdef
APR_HAS_SHARED_MEMORY in scoreboard.c/ap_reopen_scoreboard().
The
fix
for this has already been checked in but I'm not
Ryan Bloom wrote:
Once the tarball is rolled, that's it, move on to the next
version.
+1 (for Apache 2)
I disagree, strongly. :-) In this case, the tarball was rolled,
but it
was rolled incorrectly (my fault for not updating the how_to_release
site). The code was fine
Justin Erenkrantz wrote:
On Fri, Feb 01, 2002 at 04:58:00PM -0500, Greg Ames wrote:
daedalus is unhappy when I run ./config.nice. I intentionally
didn't
run
./buildconf first so I could test the included configure scripts.
Greg
config.status: creating support/envvars-std
mv:
All you have to do to roll a release is:
ssh cvs.apache.org
cvs co httpd-2.0
and apr and apr-util I would guess
Nope. The script checks out the source that it will package. The only
reason to checkout the httpd-2.0 repository is to get the
httpd_roll_release script, so that should
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
-Original Message-
From: [EMAIL PROTECTED] [mailto:trawick@rdu163-40-
092.nc.rr.com] On Behalf Of Jeff Trawick
Sent: Monday
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
-Original Message-
From: [EMAIL PROTECTED] [mailto:trawick@rdu163-40-
092.nc.rr.com] On Behalf Of Jeff Trawick
Sent
The create_connection hook has a fatal design flaw. create_conn is
run
before
ap_update_vhost_given_ip(), which means that it is impossible to
install
input and
output
filters based on vhost info.
More concisely, it is now impossible to install filters in place of
CORE_IN and
Because the shutdown/restart code has traditionally been
trouble-prone, I'm posting this for comments before
committing. This patch changes the worker MPM to use the
ap_mpm_pod_* routines for pipe-of-death management in the
same manner as prefork. My main goal was to eliminate the
poll
stoddard02/01/27 04:52:08
Modified:.CHANGES
include http_connection.h httpd.h
modules/http http_core.c
modules/proxy proxy_ftp.c proxy_http.c
server connection.c core.c
server/mpm/beos beos.c
--
Ryan Bloom [EMAIL PROTECTED]
645 Howard St. [EMAIL PROTECTED]
San Francisco, CA
-Original Message-
From: Bill Stoddard [mailto:[EMAIL PROTECTED]]
Sent: Sunday, January 27, 2002 12:55 PM
To: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: Re: cvs commit
hook install_networkio or something like that. If you like this
idea, I
will implement it tonight.
___
Ryan Bloom [EMAIL PROTECTED]
550 Jean St
Oakland CA 94610
---
One Nov. 12, Ryan committed a patch creating the create_conn hook. The
idea was to move
the client_socket out of the conn_rec presumably to make available
only to
the core_in and
core_out filters. However, I just found a backdoor...
In core_create_conn() the socket is saved away thusly:
Please don't let two mis-behaved modules color your judgment on
this.
Both proxy and perchild must be re-written if they are going to be
clean, and once that is done the stupid set_module_config can be
removed. In fact, the server ran for over a day without the
set_module_config, but
The real solution is to pass the core_net_rec structure to the
NET_TIME
filter as user-data for the filter, the same way we do for CORE_IN
and
CORE_OUT.
But you'll have to play some sort of trick to do that since the
NET_TIME filter is added from the create_request hook which does
On Fri, Jan 25, 2002 at 02:01:26PM -0800, Ryan Bloom wrote:
Please don't let two mis-behaved modules color your judgment on
this.
Both proxy and perchild must be re-written if they are going to
be
clean, and once that is done the stupid set_module_config can be
removed
Bill,
This triggered a question...
I have not followed the discussion on this thread closely, but are
you
requiring the
ScoreBoardFile directive on Windows to name the shmem? I hope
not.
At this moment, yes, otherwise it defaults to the parent/private,
child/private
On Thu, Jan 24, 2002 at 11:23:39AM -0500, Bill Stoddard wrote:
The issue is not whether we need a file backing the scoreboard. On
Unix, you make the
scoreboard use a file when the OS does not support shared memory.
I think Aaron pointed out in a recent exchange with a FreeBSD user
that
.
--
___
Ryan Bloom [EMAIL PROTECTED]
550 Jean St
Oakland CA 94610
---
The easiest way to remove the S_L_U_A from the worker MPM is to
remove
all
signals from the MPM.
1!
Basically, we have a single thread that currently
sits in sigwait, which is just waiting for a signal. We then have
every
other thread sitting in accept waiting for something
the possibility of workers_may_exit getting set asynchronously
on thread x, and apparently we wanted thread y to know ASAP. (same disclaimer)
Greg
--
___
Ryan Bloom [EMAIL PROTECTED]
550 Jean St
the ap_run_open_logs call to use plog instead of pconf, and
that should solve this problem.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Sunday 06 January 2002 11:55 pm, Sander van Zoest wrote:
On Mon, 7 Jan 2002, William A. Rowe, Jr. wrote:
From: Ryan Bloom [EMAIL PROTECTED]
I would like to discuss the reasoning behind this change. Why are we trying
to overload the meaning of DocumentRoot this way?
Because
On Monday 07 January 2002 11:20 am, Sander van Zoest wrote:
On Mon, 7 Jan 2002, Ryan Bloom wrote:
On Monday 07 January 2002 10:48 am, Sander van Zoest wrote:
So, you are saying that I can create an apache httpd daemon where it would never
make a
stat(2) in the default DocumentRoot
On Monday 07 January 2002 01:05 pm, Sander Striker wrote:
From: Ryan Bloom [mailto:[EMAIL PROTECTED]]
Sent: 07 January 2002 22:00
To: [EMAIL PROTECTED]; William A. Rowe, Jr.
Subject: Re: [PATCH] Allow DocumentRoot within Location blocks
On Monday 07 January 2002 11:55 am, William
, but that doesn't mean
we should add another option.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
.;
+}
return NULL;
}
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
haven't
wrapped my head around the ring macros yet.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
to
be sure that they are identical.
Other than that, it looks good.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Wednesday 26 December 2001 08:10 am, Greg Ames wrote:
Ryan Bloom wrote:
Since
we already know that we need this general ability for some of the auth
directives (see John Sterling's posts about some of the problems with AuthType
and AuthName), I would much rather have
you just rename your platform
to unixd, and abstract out the differences.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Sunday 23 December 2001 09:47 am, Jeff Trawick wrote:
Ryan Bloom [EMAIL PROTECTED] writes:
On Sunday 23 December 2001 06:05 am, David Reid wrote:
While trying to build the worker MPM on beos, I found that as we didn't
need/want the _np clutter these small patches were neccesary. I
to become non-static.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Sunday 23 December 2001 09:38 am, Ryan Bloom wrote:
On Sunday 23 December 2001 06:27 am, Jeff Trawick wrote:
Greg Ames [EMAIL PROTECTED] writes:
...are more painful to deal with than you might think, if the user is
allowed to code them in any order.
another thing
On Monday 24 December 2001 10:01 am, Justin Erenkrantz wrote:
On Mon, Dec 24, 2001 at 09:00:09AM -0800, Ryan Bloom wrote:
On Monday 24 December 2001 03:58 am, David Reid wrote:
This is defined, but how about we rename it to something without the unixd_
portion so that it's simply a more
to kill child processes regardless of how
they should die.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
for discussion.
I personally hate the np functions, because they are completely non-portable,
as their name implies. We should either find some way to make them portable,
or remove them IMNSHO.
Ryan
__
Ryan Bloom
of the auth
directives (see John Sterling's posts about some of the problems with AuthType
and AuthName), I would much rather have a generic function that will
manipulate the tree itself.
Ryan
__
Ryan Bloom
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
to mention that if you really want to fix this problem, we should move to
an XML based config language.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Tuesday 18 December 2001 03:35 am, jean-frederic clere wrote:
Ryan Bloom wrote:
Hi.
The big thing about ServerName in 2.0 is that it has taken over for the Port
directive from 1.3. So, the following config means that the server
listens on 127.0.0.1, port 8080, but reports itself
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
locally.
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
not want to detach on Unix, because if you do, the
CGI script will end up forking three times.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
, so it will uncover more bugs than mod_negotation
does alone.
Bill
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
a handler name, not both.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Thursday 13 December 2001 01:44 pm, Greg Ames wrote:
Ryan Bloom wrote:
I still think the handler_fn function is overkill. The performance of
Apache 1.3 wasn't bad, because we did sane string compares, making sure
that the lengths were equal before doing a full strcmp.
You're right
specified?
...
/VirtualHost
Thanks in advance,
Liza
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Wednesday 12 December 2001 02:25 pm, Harrie Hazewinkel wrote:
--On Wednesday, December 12, 2001 2:21 PM -0800 Ryan Bloom
[EMAIL PROTECTED] wrote:
On Wednesday 12 December 2001 02:14 pm, Harrie Hazewinkel wrote:
HI,
Some time ago the post_config phase return types was changed from
times, and everybody seems to be in favor
or if.
Ryan
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
to believe that almost every point has been
covered. -- justin
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
under
the 2.0 filter semantics. And, I'll argue that we must allow
modules to call for a flush. -- justin
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
On Wednesday 05 December 2001 01:22 pm, Ian Holsman wrote:
isn't this VERY similiar to what SGI's state machine thingy was
going to do ?
It most likely is, this is not a uncommon design.
Ryan
__
Ryan Bloom
there for a while already.
Ryan
Did there is a plan to add mod_ldap to Apache 2.0 ?
For now I only see documentation.
Regards
--
__
Ryan Bloom [EMAIL PROTECTED]
Covalent Technologies [EMAIL PROTECTED]
--
301 - 400 of 684 matches
Mail list logo
