On Mon, Sep 17, 2001 at 11:38:25PM -0700, Justin Erenkrantz wrote:
>...
> Since there is a veto against the use of SIGWINCH by default, the
> current code in CVS is broken. I would like to resolve this
> before we tag and roll again. -- justin
By definition, it must be resolved before the nex
On Mon, Sep 17, 2001 at 03:52:21PM -0700, Jon Travis wrote:
> I've got a bit of code that needs to run after a connection to a client
> has been closed. Right now I can (kind of) spoof this by setting the
> keepalive for the client to 0, and registering a cleanup on the
> request_req pool. Unfo
Greg Stein <[EMAIL PROTECTED]> writes:
> 2) move the ap_lingering_close inside ap_process_connection, then call it
>from with ap_process_connection. This *almost* works. All MPMs have a
>call to ap_process_connection followed by a call to ap_lingering_close.
>The only MPM that does ot
Hi,
I'm writing a module for apache 2. What I need is to create a socket from
the server, then connect it to somewhere and keep the connection open
while the server is alive. Also, of course, I would like to be able to
receive data from the socket and then to trigger handling of a request by
the
On Monday 17 September 2001 09:33 pm, William A. Rowe, Jr. wrote:
> From: "Jon Travis" <[EMAIL PROTECTED]>
> Sent: Monday, September 17, 2001 6:32 PM
>
> > I tried setting keepalive == 0 in the handler, and doing my ju-ju in
> > the log_transaction phase. The client was still hanging around.
>
>
Cliff Woolley <[EMAIL PROTECTED]> writes:
> httpd-win.conf only differs from httpd-std.conf in minor ways as well.
> The biggest differences again are in default directory names. IMO a
> comment or two would fix that problem right up.
A default define representing the platform could be useful.
On Tuesday 18 September 2001 01:58 am, Greg Stein wrote:
As I said a LONG time ago. I'm not veto'ing this change. That doesn't
mean I can't gripe about it. I am sick and tired of going back and forth
over issues that were decided years ago.
Ryan
> On Mon, Sep 17, 2001 at 11:38:25PM -0700, Ju
On Tuesday 18 September 2001 02:10 am, Greg Stein wrote:
> On Mon, Sep 17, 2001 at 03:52:21PM -0700, Jon Travis wrote:
> > I've got a bit of code that needs to run after a connection to a client
> > has been closed. Right now I can (kind of) spoof this by setting the
> > keepalive for the client
Hi,
How do i configure and build apache with 2 or more modules of my own?(is it
possible?)
I know for one module you can do it with the following command :
./configure --enable-so --with-module=one --enable-mods-shared=mod_one.so
Thanks
Hany
On Tuesday 18 September 2001 06:22 am, zethix or something wrote:
This is possible, but not easy. You will need to write your own MPM,
so that you can detect when there is data on that socket. Of course,
you could also just modify one of the existing MPMs, but this kind
of logic is unlikely to
On Tuesday 18 September 2001 07:02 am, Farag, Hany M (Hany) wrote:
--with-module only accepts one module, and you can't add more. I keep meaning
to go in and fix that, but I haven't had time recently.
I would suggest just creating a config.m4 file for your own modules, and copying
the files int
On Tue, 18 Sep 2001, Ryan Bloom wrote:
> > You've confused the issue with your subject line (everybody is bugging out
> > because they're relating it to logging). It should not have anything to do
> > with "log". We have a pre-connection hook, so call yours post-connection.
> > That is when you w
Why not let the MPM register the lingerclose with APR_HOOK_MIDDLE in the
post_connection hook? That way, if Jon's (or any other author's) intent is
to work before the lingering close, then it can be APR_HOOK_FIRST. Otherwise
register it APR_HOOK_LAST.
Problem solved?
- Original Message ---
On Tuesday 18 September 2001 08:19 am, Cliff Woolley wrote:
> On Tue, 18 Sep 2001, Ryan Bloom wrote:
> > > You've confused the issue with your subject line (everybody is bugging
> > > out because they're relating it to logging). It should not have
> > > anything to do with "log". We have a pre-con
On Tue, 18 Sep 2001, Ryan Bloom wrote:
> > Actually, I was just about to ask about child_exit for unrelated reasons.
> > Somebody asked me what happened to child_exit... I was just about to tell
> > them that all they had to do was register a cleanup on pchild when I
> > realized that the pchild
Yes, FD_SETSIZE is defined in sys/types.h on UNIX flavored
systems. If you set it to a high enough value
(i.e. #DEFINE FD_SETSIZE 65535 ) before sys/types.h gets included,
it will override the value set in sys/types.h. We (myself and a
couple co-workers) have submitted a patch to APR ([EMAIL PR
doesn't mod-pop3/mod_ssl do something similiar to this?
they listen on different ports and the standard '80'
you might be able to do something similiar to this.
but you need some method of not 'ending' the connection'
On Tue, 2001-09-18 at 07:05, Ryan Bloom wrote:
> On Tuesday 18 September 2001
Not really. There are no current modules that implement a persistent
connection and use that to serve requests. mod_pop3 and mod_ssl
use the listen directive to open different sockets, but the core closes
those sockets automatically, because they are registered with the
ptrans pool. Even if th
On Tuesday 18 September 2001 08:17 am, William A. Rowe, Jr. wrote:
> Why not let the MPM register the lingerclose with APR_HOOK_MIDDLE in the
> post_connection hook? That way, if Jon's (or any other author's) intent is
> to work before the lingering close, then it can be APR_HOOK_FIRST.
> Otherw
Nope.. mod_ssl doesn't maintain such persistant connections - the
connections are alive only as long as that client connection is alive.
-Madhu
-Original Message-
From: Ian Holsman [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 18, 2001 9:14 AM
To: [EMAIL PROTECTED]
Cc: [EMAIL PROT
Here's what i did:
. created the config.m4 files within module one and two directories under
modules.
. ./config --enable-so --enable-one --enable-two
. make
. make install
. ./httpd -l
But could not see that mod_one and mod_two listed.
What did i do wrong?
Thanks
Hany
-Original Message
On Tuesday 18 September 2001 10:50 am, Farag, Hany M (Hany) wrote:
Did you re-run buildconf?
Ryan
> Here's what i did:
> . created the config.m4 files within module one and two directories under
> modules.
> . ./config --enable-so --enable-one --enable-two
> . make
> . make install
> . ./httpd
From: "Ryan Bloom" <[EMAIL PROTECTED]>
Sent: Tuesday, September 18, 2001 11:44 AM
> On Tuesday 18 September 2001 08:17 am, William A. Rowe, Jr. wrote:
> > Why not let the MPM register the lingerclose with APR_HOOK_MIDDLE in the
> > post_connection hook? That way, if Jon's (or any other author's
This patch allows modules/ssl to build if src and build dirs differ.
--
Cody Sherr
Engineer
Covalent Technologies
phone: (415)536-5292
email: [EMAIL PROTECTED]
Index: Makefile.in
===
RCS file: /home/cvspublic/httpd-2.0/modules/s
no, but now after i ran it and did configure and then make i get this error:
make[2]: Leaving directory `/usr/local/src/httpd-2_0_24/server'
make[1]: Leaving directory `/usr/local/src/httpd-2_0_24/server'
Making all in modules
make[1]: Entering directory `/usr/local/src/httpd-2_0_24/modules'
Makin
On Tue, Sep 18, 2001 at 02:20:35PM -0500, William A. Rowe, Jr. wrote:
> From: "Ryan Bloom" <[EMAIL PROTECTED]>
> Sent: Tuesday, September 18, 2001 11:44 AM
>
>
> > On Tuesday 18 September 2001 08:17 am, William A. Rowe, Jr. wrote:
> > > Why not let the MPM register the lingerclose with APR_HOOK_
On Tuesday 18 September 2001 12:59 pm, Farag, Hany M (Hany) wrote:
Did you re-run ./configure after you ran buildconf? What does your config.m4
look like?
Ryan
> no, but now after i ran it and did configure and then make i get this
> error: make[2]: Leaving directory `/usr/local/src/httpd-2_0_
'not sure if it makes sense - but, do you have the modules.mk / Makefile.in
in that directory.
Thanks
-Madhu
-Original Message-
From: Farag, Hany M (Hany) [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 18, 2001 1:00 PM
To: [EMAIL PROTECTED]
Subject: RE: How to build Apache2.0 with m
yes, it looks like this:
dnl modules enabled in this directory by default
dnl APACHE_MODULE(name, helptext[, objects[, structname[, default[,
config)
APACHE_MODPATH_INIT(one)
APACHE_MODULE(one, testing module one, , , yes)
APR_ADDTO(LT_LDFLAGS,-export-dynamic)
APACHE_MODPATH_FINISH
Than
I have modules.mk and Makefile in that directory.
Hany
-Original Message-
From: MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1)
[mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 18, 2001 4:38 PM
To: '[EMAIL PROTECTED]'
Subject: RE: How to build Apache2.0 with more than one module
'not sure
On Monday 17 September 2001 11:12 am, Aaron Bannert wrote:
> After working with my two proposed worker MPM models, I've become more
> confident in the simple model. I'll continue benchmarking both designs,
> but I wanted to get this one out to fix what's in CVS right now, and
> so I can provide so
On Tuesday 18 September 2001 01:53 pm, Farag, Hany M (Hany) wrote:
Did you put a Makefile.in into the one directory?
> yes, it looks like this:
>
> dnl modules enabled in this directory by default
>
> dnl APACHE_MODULE(name, helptext[, objects[, structname[, default[,
> config)
>
> APACHE_MO
This is a rather simple patch that may improve cache-hit performance
under some conditions by changing the queue of available worker threads
from FIFO to LIFO. It also adds a tiny reduction in the arithmetic that
happens in the critical section, which will definately help if you have
a lame compi
On Tue, Sep 18, 2001 at 02:21:33PM -0700, Ryan Bloom wrote:
> I finally had time to review and commit this. Keep 'em coming.
Cool. Thanks for the quick turnaround. I just posted another improvement,
and I'll be posting a third in the next few minutes :)
-aaron
This patch fixes a nasty bug in the worker MPM where the
state of the worker threads was not being reported back to the
scoreboard, and eventually all the threads running in the children
would be reported as being in the "C -- closing connection" state.
This would reak havoc on the idle_server_mai
Thank you all for your help.
I can see the 2 modules included in the build.
I was missing the Makefile.in
Thanks
Hany
-Original Message-
From: Ryan Bloom [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 18, 2001 5:34 PM
To: [EMAIL PROTECTED]; Farag, Hany M (Hany)
Subject: Re: How to bu
Greg Stein wrote:
> I'm +1 on creating httpd-rollup, and -0.5 on putting proxy back in.
Ok - first question - what do we call the rollup release:
o Option A: apache-2.x.x.tar.gz
Combines httpd-2.0, apr, apr-util, httpd-proxy and
httpd-ldap and produces an apache rollup tree.
o Option B: a
This shouldn't be needed... isn't the whole point of VPATH to look in
$(top_srcdir) if the file isn't present in the build dir?
The value passed to flex would need the $(top_srcdir) since Make can't get
in there and fix it, but the dependency line "shouldn't" need the change.
Cheers,
-g
On Tue,
Ryan Bloom wrote:
>On Sunday 16 September 2001 01:57 pm, Brian Pane wrote:
>
>>Ryan Bloom wrote:
>>
>>>On Sunday 16 September 2001 01:34 pm, Justin Erenkrantz wrote:
>>>
>>>Yes, the code isn't ready. There is at least one seg fault that we
>>>already know about.
>>>
>>Is that the dir-merge segv
On Wed, Sep 19, 2001 at 01:05:27AM +0200, Graham Leggett wrote:
> Greg Stein wrote:
>
> > I'm +1 on creating httpd-rollup, and -0.5 on putting proxy back in.
>
> Ok - first question - what do we call the rollup release:
>
> o Option A: apache-2.x.x.tar.gz
>
> Combines httpd-2.0, apr, apr-uti
On Wed, Sep 19, 2001 at 01:05:27AM +0200, Graham Leggett wrote:
>...
> Ok - first question - what do we call the rollup release:
>...
> o Option B: apache-2.x.x.tar.gz, apache-modules-2.x.x.tar.gz
>
> Combine httpd-2.0, apr, apr-util into apache-2.x.x.tar.gz,
> and combine httpd-proxy, httpd-
On Tuesday 18 September 2001 04:24 pm, Greg Stein wrote:
> This shouldn't be needed... isn't the whole point of VPATH to look in
> $(top_srcdir) if the file isn't present in the build dir?
>
> The value passed to flex would need the $(top_srcdir) since Make can't get
> in there and fix it, but the
On Tue, 18 Sep 2001, Justin Erenkrantz wrote:
> > o Option A: apache-2.x.x.tar.gz
> >
> > Combines httpd-2.0, apr, apr-util, httpd-proxy and
> > httpd-ldap and produces an apache rollup tree.
>
> +1 on Option A. I think that anything else is going to be too
> confusing for end users.
I also
On Tuesday 18 September 2001 04:25 pm, Justin Erenkrantz wrote:
> On Wed, Sep 19, 2001 at 01:05:27AM +0200, Graham Leggett wrote:
> > Greg Stein wrote:
> > > I'm +1 on creating httpd-rollup, and -0.5 on putting proxy back in.
> >
> > Ok - first question - what do we call the rollup release:
> >
>
On Tuesday 18 September 2001 04:30 pm, Greg Stein wrote:
> On Wed, Sep 19, 2001 at 01:05:27AM +0200, Graham Leggett wrote:
> >...
> > Ok - first question - what do we call the rollup release:
> >...
> > o Option B: apache-2.x.x.tar.gz, apache-modules-2.x.x.tar.gz
> >
> > Combine httpd-2.0, apr,
On Tuesday 18 September 2001 04:35 pm, Cliff Woolley wrote:
> On Tue, 18 Sep 2001, Justin Erenkrantz wrote:
> > > o Option A: apache-2.x.x.tar.gz
> > >
> > > Combines httpd-2.0, apr, apr-util, httpd-proxy and
> > > httpd-ldap and produces an apache rollup tree.
> >
> > +1 on Option A. I think
On Tue, 18 Sep 2001, Ryan Bloom wrote:
> Regardless of how we do the roll-up, nont of our builds should have the
> word Apache in them. The httpd project is the httpd project. If we use the
> word Apache, then we are co-opting the Foundation's name, instead of
> the project name.
>
Well, apach
On Tue, Sep 18, 2001 at 04:35:37PM -0700, Ryan Bloom wrote:
> Sure we do. Try --with-module. There are a few bugs in it, you can
> only add a single module right now, but it works, and you can
> build static modules into the source without re-running buildconf.
How would you go about adding mod
On Tuesday 18 September 2001 04:53 pm, Justin Erenkrantz wrote:
> On Tue, Sep 18, 2001 at 04:35:37PM -0700, Ryan Bloom wrote:
> > Sure we do. Try --with-module. There are a few bugs in it, you can
> > only add a single module right now, but it works, and you can
> > build static modules into the
On Wed, Sep 19, 2001 at 01:05:27AM +0200, Graham Leggett wrote:
> o Option B: apache-2.x.x.tar.gz, apache-modules-2.x.x.tar.gz
>
> Combine httpd-2.0, apr, apr-util into apache-2.x.x.tar.gz,
> and combine httpd-proxy, httpd-ldap into apache-modules-2.x.x.tar.gz.
>
> o Option C: apache-2.x.x.t
On Tue, Sep 18, 2001 at 11:09:12PM -, [EMAIL PROTECTED] wrote:
> rbb 01/09/18 16:09:12
>
> Modified:.CHANGES
>server/mpm/worker fdqueue.c fdqueue.h
> Log:
> Turn the worker MPM's queue into a LIFO. This may
> improve cache-hit performance under som
I agree with OtherBill.
Cleanups are not always the answer. When they are run, many things
associated with that pool could be torn down already because their cleanups
have already run.
If you need a known state to perform *operations* (as it sounds like Jon is
doing), then you can't use a cleanu
On Tue, Sep 18, 2001 at 06:52:35AM -0400, Jeff Trawick wrote:
> Greg Stein <[EMAIL PROTECTED]> writes:
>
> > 2) move the ap_lingering_close inside ap_process_connection, then call it
> >from with ap_process_connection. This *almost* works. All MPMs have a
> >call to ap_process_connection
-1 for A,B,C as currently proposed. Frankly, we've voted to put the proxy
back twice. Why are we having this vote again? The last reason I heard for
not putting the proxy back in was "we're worried about HTTP proxy standard
diverging". It sounds thin.
+1 for Option A once we cut out the knee jerk
From: "Brian Pane" <[EMAIL PROTECTED]>
Sent: Tuesday, September 18, 2001 6:19 PM
> Ryan Bloom wrote:
>
> >On Sunday 16 September 2001 01:57 pm, Brian Pane wrote:
> >
> >>Ryan Bloom wrote:
> >>
> >>>On Sunday 16 September 2001 01:34 pm, Justin Erenkrantz wrote:
> >>>
> >>>Yes, the code isn't rea
On Tue, Sep 18, 2001 at 06:01:35PM -0700, Greg Stein wrote:
> On Tue, Sep 18, 2001 at 11:09:12PM -, [EMAIL PROTECTED] wrote:
> > rbb 01/09/18 16:09:12
> >
> > Modified:.CHANGES
> >server/mpm/worker fdqueue.c fdqueue.h
> > Log:
> > Turn the worker MPM'
- Original Message -
From: "Ryan Bloom" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>; "Cliff Woolley" <[EMAIL PROTECTED]>
Sent: Tuesday, September 18, 2001 6:43 PM
Subject: Re: Q1: Rollup Release Format
> On Tuesday 18 September 2001 04:35 pm, Cliff Woolley wrote:
> > On Tue, 18 Sep 200
I'm sorry - I've transformed the entire schema.
Yes, I'm +1 for the 'real' option A.
My concerns about it remain - folks will download the 'lite' core version, only
to turn around and download the 'full' version.
And yes, something like httpd-complete would be a very nice name.
Bill
- Ori
Option B seems to be reasonable to me..
- apache-lite - containing httpd-2.0, apr, apr-util and
- apache-complete - containing lite(?) + http-proxy + http-ldap
Thanks
-Madhu
-Original Message-
From: William A. Rowe, Jr. [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, September 18, 2001 6:
On Tue, 2001-09-18 at 18:50, William A. Rowe, Jr. wrote:
> I'm sorry - I've transformed the entire schema.
>
> Yes, I'm +1 for the 'real' option A.
+1 for 'real' option A. as well
(here's a conundrum.. I have commit to proxy not httpd itself.. so does
this vote count?)
the reason I like it is tw
On Tue, 18 Sep 2001, MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
> Option B seems to be reasonable to me..
> - apache-lite - containing httpd-2.0, apr, apr-util and
> - apache-complete - containing lite(?) + http-proxy + http-ldap
Just to be clear, Option A assumes the pre-existence of w
From: "Ian Holsman" <[EMAIL PROTECTED]>
Sent: Tuesday, September 18, 2001 10:02 PM
> On Tue, 2001-09-18 at 18:50, William A. Rowe, Jr. wrote:
> > I'm sorry - I've transformed the entire schema.
> >
> > Yes, I'm +1 for the 'real' option A.
> +1 for 'real' option A. as well
> (here's a conundrum.
To quote our mantra;
be forgiving of the input, strict in the output ;)
If the user 'asks' for a rollup, they aught to get a complete package,
unless it is _clearly_ labeled as an incremental.
I'm leaning, more and more, to offer both options, -complete (no
questions asked, download nothing mor
On Tue, 18 Sep 2001, William A. Rowe, Jr. wrote:
> be forgiving of the input, strict in the output ;)
>
> If the user 'asks' for a rollup, they aught to get a complete package,
> unless it is _clearly_ labeled as an incremental.
Yep.
> I'm leaning, more and more, to offer both options, -complet
[This is a repost of an uncommitted patch from a couple of weeks ago.
I've updated it to work against the current code in CVS.]
This patch eliminates the wasteful run-time conversion of method names
from strings to numbers in places where the methods are known at compile
time.
--Brian
Index:
On Tue, Sep 18, 2001 at 10:34:14PM -0700, Brian Pane wrote:
> [This is a repost of an uncommitted patch from a couple of weeks ago.
> I've updated it to work against the current code in CVS.]
>
> This patch eliminates the wasteful run-time conversion of method names
> from strings to numbers in p
On Tue, Sep 18, 2001 at 03:14:58PM -0700, Aaron Bannert wrote:
> This patch fixes a nasty bug in the worker MPM where the
> state of the worker threads was not being reported back to the
> scoreboard, and eventually all the threads running in the children
> would be reported as being in the "C --
On Fri, Sep 14, 2001 at 12:40:14PM -0700, Ian Holsman wrote:
> here is a patch which does just that.
>
> Index: mod_include.c
> ===
> RCS file: /home/cvspublic/httpd-2.0/modules/filters/mod_include.c,v
> retrieving revision 1.146
> d
On Wed, Sep 19, 2001 at 06:53:26AM -, [EMAIL PROTECTED] wrote:
> jerenkrantz01/09/18 23:53:26
>
> Modified:server Makefile.in
>modules/loggers mod_log_config.c
>.CHANGES
> Added: include util_time.h
>server util_ti
This should clear out all of the patches in my inbox that were
either reviewed by multiple people or I thought were worthy for
inclusion on my own.
The only outstanding patch that I see right now is the one for
FD_SETSIZE from the IBM iSeries folks. I'll take a look at it, but
I'm not really
70 matches
Mail list logo
