Aaron Bannert <[EMAIL PROTECTED]> writes:
> My setjmp man page on Linux says it returns zero or non-zero, so is
> this patch more correct?
My Linux setjmp() man page is pretty poor w.r.t. the return value
> Index: server/mpm/perchild/perchild.c
>
My setjmp man page on Linux says it returns zero or non-zero, so is
this patch more correct?
Index: server/mpm/perchild/perchild.c
===
RCS file: /home/cvs/httpd-2.0/server/mpm/perchild/perchild.c,v
retrieving revision 1.93
diff -u -
Hello,
I've been trying to track down a couple of stability problems in 2.0's
mod_proxy. Both problems have to do with a misbehaving downstream server.
I have seen these issues in the public 2.0.28 release as well as the
latest CVS extract. I've traced them as far as we can, but now I need
som
Is ap_error_log2stderr in use any longer?
-aaron
Ryan Bloom <[EMAIL PROTECTED]> writes:
> The point of the accept abstraction is to allow people to poll on things other
> than regular sockets. The pipe-of-death is just one such example. An SSL
> socket is another, as is a UDP socket, or a hundred other examples. We
> should just move the pi
On Tuesday 18 December 2001 12:52 pm, Jeff Trawick wrote:
> Jeff Trawick <[EMAIL PROTECTED]> writes:
>
> > It looks like listen_recs aren't allocated from the right pool with
> > worker? (at least I couldn't reproduce this with prefork)
>
> And the answer is...
>
> *Most* of the listen_recs a
Jeff Trawick <[EMAIL PROTECTED]> writes:
> It looks like listen_recs aren't allocated from the right pool with
> worker? (at least I couldn't reproduce this with prefork)
And the answer is...
*Most* of the listen_recs are allocated from the right pool. The one
that isn't is the pipe of death
On 18 Dec 2001, Jeff Trawick wrote:
> <[EMAIL PROTECTED]> writes:
>
> > I see the ap_run_create_connection interface was changed, but the proxy
> > code was not updated yet... right?
>
> ouch... can you try this?
>
that seems to work great! thanks
sterling
Aaron Bannert wrote:
>
> On Tue, Dec 18, 2001 at 02:26:25PM -0500, Jeff Trawick wrote:
> > > s/HardServerLimit/SoftServerLimit/ ? I'm only suggesting this since
> > > the semantics of HARD_SERVER_LIMIT != HardServerLimit.
> >
> > I hear what you're saying but I am not crazy about
> > "SoftServerL
> From: Aaron Bannert [mailto:[EMAIL PROTECTED]]
> On Tue, Dec 18, 2001 at 02:26:25PM -0500, Jeff Trawick wrote:
> > > s/HardServerLimit/SoftServerLimit/ ? I'm only suggesting this since
> > > the semantics of HARD_SERVER_LIMIT != HardServerLimit.
> >
> > I hear what you're saying but I am not
On Tue, Dec 18, 2001 at 02:26:25PM -0500, Jeff Trawick wrote:
> > s/HardServerLimit/SoftServerLimit/ ? I'm only suggesting this since
> > the semantics of HARD_SERVER_LIMIT != HardServerLimit.
>
> I hear what you're saying but I am not crazy about
> "SoftServerLimit"/"SoftThreadLimit". Somebody
Aaron Bannert <[EMAIL PROTECTED]> writes:
> On Tue, Dec 18, 2001 at 06:55:11AM -0500, Jeff Trawick wrote:
> > (Yes, this is without any patches :) )
> >
> > I'll take a look at this after I get to a stopping point with the
> > server-limit/thread-limit stuff.
> >
> > It looks like listen_recs a
Aaron Bannert <[EMAIL PROTECTED]> writes:
> On Tue, Dec 18, 2001 at 11:24:36AM -0500, Jeff Trawick wrote:
> ...
> > This adds a HardServerLimit directive to prefork which allows you to
> > set a really big* limit on MaxClients (if you can spare the shared
> > memory). As always, you get to play
Greg Ames <[EMAIL PROTECTED]> writes:
> Jeff Trawick wrote:
> >
> > Greg Ames <[EMAIL PROTECTED]> writes:
>
> > > hmmm, seems like we have other cases like this already without order
> > > problems... Worker has pre-config logic to swap
> > > MaxClients and ThreadsPerChild if it doesn't like t
<[EMAIL PROTECTED]> writes:
> I see the ap_run_create_connection interface was changed, but the proxy
> code was not updated yet... right?
ouch... can you try this?
Index: modules/proxy/proxy_ftp.c
===
RCS file: /home/cvspubli
Jeff Trawick wrote:
>
> Greg Ames <[EMAIL PROTECTED]> writes:
> > hmmm, seems like we have other cases like this already without order
> > problems... Worker has pre-config logic to swap
> > MaxClients and ThreadsPerChild if it doesn't like the order.
>
> I'll implement such since there is pre
I see the ap_run_create_connection interface was changed, but the proxy
code was not updated yet... right?
sterling
Greg Ames <[EMAIL PROTECTED]> writes:
> Jeff Trawick wrote:
>
> > cool, but of course it is perfect and needs no additional testing :)
>
> well then, maybe the eggnog and Christmas cookies are making me
> hallucinate, but with this config:
>
>
> MinSpareServers 20
> MaxSpareServers 50
> Start
From: "Brad Nicholes" <[EMAIL PROTECTED]>
Sent: Tuesday, December 18, 2001 12:10 PM
> Bill,
> I guess I don't follow your comment. mpm_netware.c still follows
> the same scheme as other MPM's in that it respects HARD_SERVER_LIMIT, it
> just makes sure that it is never anything other than 1.
On Tue, Dec 18, 2001 at 11:54:24AM +0100, Pavel Novy wrote:
> Hi,
> I'm running Apache 2.0 (2.0.29-dev) on my test server (NetWare) and I've
> found the following line(s) in server's log file recently:
>
> xx.xx.xx.xx - - [05/Dec/2001:17:19:20 +0100] "GET http://www.s3.com
> HTTP/1.1" 200 1456
Jeff Trawick wrote:
> cool, but of course it is perfect and needs no additional testing :)
well then, maybe the eggnog and Christmas cookies are making me
hallucinate, but with this config:
MinSpareServers 20
MaxSpareServers 50
StartServers 50
MaxClients 500
MaxRequestsPerChild 0
HardServerLim
Bill,
I guess I don't follow your comment. mpm_netware.c still follows
the same scheme as other MPM's in that it respects HARD_SERVER_LIMIT, it
just makes sure that it is never anything other than 1. All references
to the scoreboard always use 0 for the server slot and a thread id for
the th
On Tue, Dec 18, 2001 at 11:24:36AM -0500, Jeff Trawick wrote:
...
> This adds a HardServerLimit directive to prefork which allows you to
> set a really big* limit on MaxClients (if you can spare the shared
> memory). As always, you get to play with MaxClients across restarts
> but you can't play
Jeff,
My mistake. I read the changes backwards. The ap_log_error()
calls should be using HARD_THREAD_LIMIT as per your change. I am in the
middle of making some other changes to mpm_netware.c, so I will be sure
to change the message text to reflect threads rather than servers also.
thanks
On Tue, Dec 18, 2001 at 06:55:11AM -0500, Jeff Trawick wrote:
> (Yes, this is without any patches :) )
>
> I'll take a look at this after I get to a stopping point with the
> server-limit/thread-limit stuff.
>
> It looks like listen_recs aren't allocated from the right pool with
> worker? (at l
From: "Mladen Turk" <[EMAIL PROTECTED]>
Sent: Tuesday, December 18, 2001 11:10 AM
> I know that it's still a beta, but ...
-1 with a capital V for introducing _another_ set of win32 build files.
If there is any mistakes that prevent our .dsps from successfully
converting, mladen, please point
Hi,
I know that it's still a beta, but ...
MT.
vc7.zip
Description: Zip compressed data
Greg Ames <[EMAIL PROTECTED]> writes:
> Jeff Trawick wrote:
> >
> > This adds a HardServerLimit directive to prefork
>
> I'll give it a shot. Thanks much!
cool, but of course it is perfect and needs no additional testing :)
the main purpose of showing the patch was to present a challenge fo
Jeff Trawick wrote:
>
> A certain person complained over the phone again this a.m. about
> having to tweak HARD_SERVER_LIMIT define for daedalus,
can't imagine who would do that...
> This adds a HardServerLimit directive to prefork
I'll give it a shot. Thanks much!
Greg
> Jeff,
> I think that the mistake I made here was not changing the error
> message text. I had changed HARD_SERVER_LIMIT to HARD_THREAD_LIMIT on
> purpose because NetWare does not have processes. Therefore we will
> never have more than one server anyway and in fact the NetWare MPM does
>
"Brad Nicholes" <[EMAIL PROTECTED]> writes:
> Jeff,
> I think that the mistake I made here was not changing the error
> message text. I had changed HARD_SERVER_LIMIT to HARD_THREAD_LIMIT on
> purpose because NetWare does not have processes. Therefore we will
> never have more than one serv
A certain person complained over the phone again this a.m. about
having to tweak HARD_SERVER_LIMIT define for daedalus, and prefork MPM
is a simpler playground than worker for figuring out the user
interface (since there is no HARD_THREAD_LIMIT), so what the heck...
This adds a HardServerLimit di
Greg Ames <[EMAIL PROTECTED]> writes:
> > Hmmm... (2nd thoughts :) )
> >
> > mpm_default.h exists so people can edit default settings in one nice
> > place... it wasn't nice for me to move these things out of
> > mpm_defaults.h... I'll move them back in there... hopefully these
> > settings
Jeff,
I think that the mistake I made here was not changing the error
message text. I had changed HARD_SERVER_LIMIT to HARD_THREAD_LIMIT on
purpose because NetWare does not have processes. Therefore we will
never have more than one server anyway and in fact the NetWare MPM does
not even pro
"William A. Rowe, Jr." <[EMAIL PROTECTED]> writes:
> From: "Jeff Trawick" <[EMAIL PROTECTED]>
> Sent: Tuesday, December 18, 2001 7:55 AM
>
>
> > mod_ssl has some questionable access to the scoreboard which doesn't
> > work correctly starting a few minutes ago because SCOREBOARD_SIZE is
> > much
Jeff Trawick wrote:
>
> [EMAIL PROTECTED] writes:
>
> > Log:
> > Change core code to allow an MPM to set hard thread/server
> > limits at startup.
++1. Thanks for doing this work, Jeff. It always seemed wrong that we
had to recompile http_protocol.c whenever we changed the scoreboard
ge
[EMAIL PROTECTED] writes:
> Log:
> Change core code to allow an MPM to set hard thread/server
> limits at startup.
> +
> +/* we only ever have 1 main process running... */
> +#define HARD_SERVER_LIMIT 1
> +
> +/* Limit on the threads per process. Clients will be locked out if mo
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 re
jean-frederic clere <[EMAIL PROTECTED]> 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 as foo.com
From: "Jeff Trawick" <[EMAIL PROTECTED]>
Sent: Tuesday, December 18, 2001 7:55 AM
> mod_ssl has some questionable access to the scoreboard which doesn't
> work correctly starting a few minutes ago because SCOREBOARD_SIZE is
> much smaller than it expects (< 1024, not indicative of the real size
mod_ssl has some questionable access to the scoreboard which doesn't
work correctly starting a few minutes ago because SCOREBOARD_SIZE is
much smaller than it expects (< 1024, not indicative of the real size
of the scoreboard).
This patch should get things going again, but I wonder if anybody has
(Yes, this is without any patches :) )
I'll take a look at this after I get to a stopping point with the
server-limit/thread-limit stuff.
It looks like listen_recs aren't allocated from the right pool with
worker? (at least I couldn't reproduce this with prefork)
Program received signal SIGHU
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 as foo.com on port 80.
>
>
> ServerName foo.com:80
>
>
> T
Hi,
I'm running Apache 2.0 (2.0.29-dev) on my test server (NetWare) and I've
found the following line(s) in server's log file recently:
xx.xx.xx.xx - - [05/Dec/2001:17:19:20 +0100] "GET http://www.s3.com
HTTP/1.1" 200 1456
I've made some work-around and found the following:
1. Server's docume
44 matches
Mail list logo
