William A. Rowe, Jr. wrote:
At 02:18 PM 11/10/2003, you wrote:
Bill Stoddard wrote:
[...]
Any ideas how to avoid the second call to initalize_module when I run as a service ?
You can't avoid the second call but there are ways to gracefully handle it. Here is a snip of code from a module I
Jim Jagielski wrote:
On Nov 13, 2003, at 2:43 PM, Jeff Trawick wrote:
ap_mpm_query(), implemented by each MPM, would need some help from
core to determine which pass of the pre/post-config hook it is, since
that is out of the MPM's domain.
It seems to me that the proposed patch (for modules
Joe Schaefer wrote:
Another approach would be to do nothing.
don't tell me people have been reading the wheel of httpd-ev... thread :)
I'd be happy to
prepare a patch if this approach - to do nothing - is
unacceptable. Of course, suggestions to make the patch
as robust as possible would be
Glenn wrote:
- lack of clear leadership and even basic direction
At present I see most of the time volunteered by developers to be spent
communicating with users on the bug db and trying to fix bugs. That sounds all
well and good to me.
If somebody wants something big implemented that they
Rasmus Lerdorf wrote:
*** We need to get back many of the disenfranchised Apache 1.3 developers
Who are these people?
/me raises a hand
Just compare the list of contributors today to 4 years ago if you want a
list.
diff knows no reasons.
Too bad all these supposedly-disenfranchised people
Rasmus Lerdorf wrote:
On Sun, 16 Nov 2003, Jeff Trawick wrote:
Too bad all these supposedly-disenfranchised people aren't around to review 1.3
fixes. 1.3 would be healthier if they were.
And it is the reason for why they are not around that is in question here.
Why wouldn't there be plenty
If Sander hadn't gone awol this wouldn't be so fubar. Any comments?
ap_server_state_t {
enum {AP_STARTING, AP_STARTED, AP_STOPPING} state;
enum {AP_FIRST_START, AP_SUBSEQUENT_START} start_type; /* if AP_STARTING */
enum {AP_GRACEFUL_STOP, AP_HARD_STOP} stop_type; /* if
some state transition is made.
I put the declarations below on the list because we had some conversations on
the list stall last week, and Sander and I didn't get through discussing it,
and I thought it would be nice to show some code.
At 05:17 PM 11/16/2003, Jeff Trawick wrote:
If Sander hadn't
André Malo wrote:
* Ben Collins-Sussman [EMAIL PROTECTED] wrote:
FWIW: in the Subversion project, we've assigned the hat of patch
manager to a volunteer in the community. He watches patches come in.
If any patch goes unanswered for a week or more, the patch manager files
it in the
Marc G. Fournier wrote:
I've just spent the past little while searching Google for anything that
might shed some light on this, and am drawing a blank ... from what I can
tell, it isn't a configuration issue (unless I'm missing a configuration
setting), but I don't know how to debug it from the
Andre Schild wrote:
what would be the best way to log all error 500 (all status 50x responses
in fact) into a separate logfile ?
One way could be a piped log, but depending of the format
the user has configured the output can be very different.
use piped log and tell your users to put the
Geoffrey Young wrote:
Server version: Apache/2.1.0-dev
Server built: Aug 12 2003 02:25:22
Server's Module Magic Number: 20030213:1
Architecture: 32-bit
Server MPM: Prefork
too bad ap_show_mpm doesn't list this like other modules :( (prefork.c or
worker.c)... the major source file name is
Aaron Bannert wrote:
This seems to work fine on my Mac OS X (10.3 Panther) box, my linux
2.4 x86 box, and my linux x86_64 (amd64 opteron) box. Let me know
if there are any problems (and fixes) so I can encorporate the fixes
and reroll in a few days. My goal is to churn out updated -rc tarballs
ranier wrote:
--- sdbm-old.c 2003-07-08 19:55:04.0 -0300
+++ sdbm.c 2003-07-08 19:57:52.0 -0300
@@ -495,7 +495,7 @@
while (dbit db-maxbno getdbit(db, dbit))
dbit = 2 * dbit + ((hash (1 hbit++)) ? 2 : 1);
-debug((dbit: %d..., dbit));
+debug((dbit: %lu...,
Igor Kovalenko wrote:
I am just lurking here really... but a while back I did try to post some
patches dealing with QNX support to both Apache and PHP. In my humble
opinion, there was a world of difference in attitude that I have encountered
in those two projects.
The attitude of PHP folks was
Joshua Slive wrote:
On Mon, 17 Nov 2003, Jeff Trawick wrote:
What must be done to the bug tracker we have now to track the patches better?
(I'm not against having something better, but I don't want to see a delay in
waiting for some magic tool.)
We can already assign the keyword PatchAvailable
Jeff Trawick wrote:
I'd like to write up some notes tomorrow as a draft of a new how to
submit a patch, pointing to existing info on httpd patches and
describing the current bugzilla capability.
With the help of Glenn (gs-apache-dev at gluelogic.com, there is a first draft
at http
Paul J. Reder wrote:
You could have something like:
An important benefit of Apache httpd is that you can modify the source as
you require. While we do understand the desire to avoid re-applying patches
to each new Apache httpd release, that ability allows users with unique
requirements to tailor
Geoffrey Young wrote:
Jeff Trawick wrote:
Geoffrey Young wrote:
wow, I didn't expect to see this followed up upon. thanks.
maybe httpd developers should be stranded in airports more often... also,
several weeks ago somebody was complaining to me about various things they
didn't like about
[EMAIL PROTECTED] wrote:
ianh2003/11/19 19:45:23
Modified:.CHANGES
docs/manual/mod mod_autoindex.xml
modules/generators mod_autoindex.c
which prompts me to add a section on special documentation issues to my
submitting-your-patch changes,
Folks with outstanding patches or new patches in the works:
Please follow the guidelines at http://httpd.apache.org/dev/patches.html
The biggest change is that we ask people to submit patches via the bug
database. Some people have been doing this already, but many patches were
posted only to
Jim Jagielski wrote:
This has been itching me for awhile... I don't like loosing
%c whenever mod_ssl is in the mix. This is basically a
feature (yes, I said it, a feature)
You may be on shaky ground there, Jim. At the hackathon, I suggested an
interesting feature for 1.3 to one of those
William A. Rowe, Jr. wrote:
We need to axe or decorate the symbol NO_USE_SIGACTION in our
ongoing effort to prevent namespace clashes.
sounds good
We do have a flag APR_HAVE_SIGACTION which is tested and
configured for, and the attached patch to the Apache MPMs
presumes that this was the intent
Stas Bekman wrote:
Jeff Trawick wrote:
- make sure that the PR # is in the subject of any on-list discussion
of the patch
Why not automate this process. Change bugzilla to do the job and
autogenerate the right link:
a hred=mailto:[EMAIL PROTECTED]Post
the patch to the httpd-dev list
William A. Rowe, Jr. wrote:
Well here's the patch... enjoy
+1
Haskell Curry wrote:
How much diferent connections can a child handle !?
one per MPM-created thread
for prefork, that is 1 per child since there is only one MPM-created thread
for threaded MPMs, the number of threads (and thus concurrent connections) is
set by the ThreadsPerChild directive
Haskell Curry wrote:
does the child not schedule the request of a same connection !?
A subsequent request on the same connection will be handled by the same thread
that handled the first request on that connection.
like
[computer A] connection--- [computer B]
[computer A]
Glenn wrote:
Should the filters mentioned above use apr_brigade_cleanup() instead?
e.g. what happens if a handler uses ap_brigade_pass() to pass a brigade
to a filter, say mod_ext_filter, and then calls apr_brigade_cleanup()
and reuses the brigade? The brigade will already have been destroyed
by
Glenn wrote:
On Sun, Nov 23, 2003 at 11:09:47AM -0500, Jeff Trawick wrote:
Glenn wrote:
Does that mean that it is incorrect to call apr_brigade_destroy() on
a brigade that you have passed? There are a number of places in the
code that do this.
yes; where is that code?
What is the benefit
Lars Hecking wrote:
# ./apachectl configtest
Syntax error on line 205 of /usr/local/apache/conf/httpd.conf:
Cannot load /usr/local/apache/libexec/mod_mime_magic.so into server: ld.so.1:
/usr/local/apache/bin/httpd: fatal: relocation error: file
/usr/local/apache/libexec/mod_mime_magic.so:
Lars Hecking wrote:
Plain, unpatched 1.3.29 without mod_ssl and mm gives the same error.
# ./apachectl configtest
Syntax error on line 205 of /WWWserv/conf/httpd.conf:
Cannot load /WWWserv/libexec/mod_mime_magic.so into server: ld.so.1:
/WWWserv/bin/httpd: fatal: relocation error: file
Lars Hecking wrote
:
gcc 2.95.3, no GNU binutils. apache is configured with
OPTIM=-O2 \
./configure \
I tweaked only the prefix option and it appears to work fine for me. I too
have gcc 2.95.3 (from Sun freeware CD I'd guess) and AFAIK I'm not using GNU
binutils either.
What does this show
Lars Hecking wrote:
$ nm /usr/local/apache/bin/httpd | grep alloc
[403] | 0| 0|FILE |LOCL |0|ABS|alloc.c
[406] |518516| 4|OBJT |LOCL |0|13 |alloc_mutex
[1675] |156472| 8|FUNC |GLOB |0|10 |ap_cleanup_alloc
[1332] |156420|
Lars Hecking wrote:
Jeff, one more thing: Could you email me, off-list, the output of
truss /usr/local/apache/bin/httpd -t? You can use -o to capture the
output into a file.
done
the worker MPM is under load yields
no surprises.
--
Jeff Trawick | [EMAIL PROTECTED]
Born in Roswell... married an alien...
Bastiaan van der Put wrote:
Been using the patch for a while now.
But it seems not only some notice/error in the error logs
piped log program '/usr/local/apache2/bin/logresolve
/home/accounts/x/xx/logs/access_log' failed unexpectedly
Does that happen when you stop/restart apache, or can it
Joe Schaefer wrote:
For kicks I just tried to build httpd from HEAD in httpd-2.0
and apr* using Visual Studio .NET, but ran into a problem
with server/mpm/winnt/child.c. Unfortunately the build barfs
on a deprecated function - ap_proc_other_child_check - that's
no longer present in libapr API:
FWIW, it segfaults on a jsp request... I suppose that this is handled by a
third party module such as mod_jk? See the final snippet:
[pid 32119] read(11, GET /messaging/businessObject.js..., 4096) = 775
[pid 32119] rt_sigaction(SIGUSR1, {SIG_IGN}, {0x4002127c, [],
SA_INTERRUPT|0x400}, 8) =
Jeff Trawick wrote:
FWIW, it segfaults on a jsp request... I suppose that this is handled
by a third party module such as mod_jk? See the final snippet:
I was reminded by a little bird that you had mentioned in your original message
that you were using the WebLogic plug-in. Maybe some hints
Geoffrey Young wrote:
contrary to your advice, I took the initiative and coded IfThreaded
against 2.1. basically, this would allow you to group threaded directives
together. so, instead of this
IfModule worker.c
ThreadsPerChild 5
/IfModule
and so on for every threaded mpm, we could
Geoffrey Young wrote:
here is the latest patch. basically, it's the same as what I submitted
before. the differences are those suggested by stas and jeff - make the
AP_MPMQ_STATIC/DYNAMIC wording a bit better and axe the -D APACHE_MPM_DIR=
stuff.
commited to 2.1-dev... thanks!
Gagan Puri wrote:
I have installed Apache 2.0.48 with enable-ssl option on Solaris 8.
The server starts and works fine in non-ssl mode however on starting in ssl mode it
gives the following exception:
# ../bin/apachectl startssl
[Thu Dec 04 16:27:04 2003] [crit] [Thu Dec 04 16:27:04 2003] file
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
Here is the fix for AIX:
http://cvs.apache.org/viewcvs.cgi/httpd-2.0/modules/generators/
mod_cgid.c.diff?r1=1.158r2=1.159
Thanks - I'll try it out right away.
While researching the AIX issue affecting mod_cgid, in which kill() would not
report
Geoffrey Young wrote:
I've griped a bit before about having default_handler make conditional GET
decisions, and this is probably another instance where having
ap_meets_conditions in it's own filter could avoid inevitable problems.
I'm up for laying out my issues if the list is interested in them
I put some rough notes at http://www.apache.org/~trawick/servbench.html
I'm curous about the perceived usefulness of it as well as who would be
interested in putting some effort into definition of the benchmark and/or
working on reports for specific platforms.
Personally I'd love to see a
[EMAIL PROTECTED] wrote:
complain via error_log when mod_include's INCLUDES filter is
enabled, but the relevant Options flag allowing the filter to run
for the specific resource wasn't set, so that the filter won't
silently get skipped.
Index: mod_include.c
Stas Bekman wrote:
- httpd 2.0 doesn't maintain start/stop times in the extended status mode
add it
- while vhost info was added, it doesn't contain the port info,
rendering the vhost record useless for the non named vhosts.
any objections to adding it?
none from me
Should it be in a separate
Currently when Apache httpd accepts a new connection via APR, it compares the
fd with FD_SETSIZE and bombs if fd = FD_SETSIZE.
The limited value of this check is on platforms such as OS X 10.3 with no
poll(), where APR has to use select(). Unfortunately, use 1K threads with
worker MPM on
Brian Akins wrote:
Backported from 2.1. Stable for me in various loads.
without the non-portable atomic code available and enabled in APR, this is
going to hurt performance, right? (more mutex operations performed in mainline
path for the unlucky who use the non-portable atomics???) is it
Stas Bekman wrote:
Does this look good?
Index: modules/filters/mod_include.c
===
RCS file: /home/cvs/httpd-2.0/modules/filters/mod_include.c,v
retrieving revision 1.291
diff -u -r1.291 mod_include.c
--- modules/filters/mod_include.c
Cliff Woolley wrote:
On Tue, 9 Dec 2003, Greg Stein wrote:
This is where process gets in the way of just doing the right thing.
Backport it for chrissakes.
amen.
The process requires getting 3 +1s. Anywhere (list, irc, phonecall, STATUS) is
okay. I assume what you meant was that you
Sander Striker wrote:
On Wed, 2003-12-10 at 12:29, Jeff Trawick wrote:
Cliff Woolley wrote:
On Tue, 9 Dec 2003, Greg Stein wrote:
This is where process gets in the way of just doing the right thing.
Backport it for chrissakes.
amen.
The process requires getting 3 +1s. Anywhere (list, irc
Marc Giger wrote:
I'm not sure but I think there is a bug in the sdbm module.
The same code runs fine on intel hardware but not on my alpha.
[EMAIL PROTECTED] is the proper mailing list for apr and apr-util discussions.
I found this by running subversion which calls apr_dbm_* and like
Larry Toppi wrote:
I found the culprit. The EOS bucket was being removed from the brigade
but not destroyed. I'm going to submit the following patch to fix this bug.
patch mangled just a teeny weeny bit to fit with changes to that code in
2.1-dev, then committed to 2.1-dev... I'll suggest
[EMAIL PROTECTED] wrote:
Jeff Trawick wrote:
Also, should we be allowing modules to look at the scoreboard
structures directly, or indexing into the scoreboard arrays directly?
If the latter were avoided, at least we could add fields to the end of
process_score or worker_score without
Matthieu Estrade wrote:
Here is a little patch, fixing the ldap cache using shared memory.
After the fix for plateform using SHM or not, it was a problem with
cache init, unable to get it's rmm address to alloc memory.
why is st-cache_rmm not filled out (or invalid)?
(gdb 5.0 or 6.0 on Solaris 9 x86, at least; not sure if it affects Solaris/Sparc)
If you're having problems viewing threads after attaching to a threaded child
process, hack the start_threads() function in worker.c to sleep instead of
exit, then see if you can attach and debug successfully.
[EMAIL PROTECTED] wrote:
bnicholes2003/12/16 12:55:04
Modified:server/mpm/netware mpm_netware.c
Log:
Added mpm_state to ap_mpm_query() for the NetWare MPM
Thanks so much! I'm not even sure I have it working right for WinNT MPM yet,
and that's even something I can test :)
short description:
. implements a single process/single thread daemon running alongside Apache MPM
(on Unix) to enable requests to be handled by an alternate mini-MPM
. this can be useful for diagnosing web server problems
. this can be useful for providing a simpler (unthreaded, predictable)
We use hsregex on older Solaris (2.0-2.5 or something like that).
Theoretically maybe something stops working or starts working when the switch
is made, but beyond that theoretical possibility does anyone have real
knowledge that there is a non-trivial likelihood of that occurring or that
[EMAIL PROTECTED] wrote:
+ PR 24469, broken reverse lookups with IPv4-mapped addrs on old OS X
+The autoconf check added between 2.0.47 and 2.0.48 isn't
+sufficient for catching the OS X problem, because building the
+IPv4 numeric address string from IPv4-mapped address would
# apachectl start
(some error message)
Unable to open logs\n
(that \n shouldn't be escaped)
[EMAIL PROTECTED] wrote:
jerenkrantz2003/12/26 23:41:28
Modified:.Tag: APACHE_2_0_BRANCH STATUS
Log:
Reflect merged backports (those that I casted at least the 3rd +1 for), and
cast some votes on those with less than 3 +1s.
Happy holidays!!!
Index: STATUS
@@ -209,7
On Dec 28, 2003, at 6:55 PM, xmb wrote:
Morning, i just saw something weird while the server had heavy load
for a day
Server uptime: 28 days 3 hours 33 minutes 54 seconds
Total accesses: 5327861 - Total Traffic: 102.0 GB
Server uptime: 28 days 3 hours 45 minutes 44 seconds
Total
On Dec 28, 2003, at 7:03 PM, Colm MacCarthaigh wrote:
On Fri, Dec 26, 2003 at 04:10:04PM -0800, Justin Erenkrantz wrote:
--On Tuesday, December 23, 2003 2:59 PM + [EMAIL PROTECTED]
wrote:
+ PR 24469, broken reverse lookups with IPv4-mapped addrs on old
OS X
+The autoconf check added
Ben Laurie wrote:
One of the problems that crops up depressingly often is that someone
gets owned, and they can't find out why. This is generally because the
offending request didn't get logged, because the server died before it
logged it.
far more often than getting owned are the
Ben Laurie wrote:
If it does nothing unless a file is specified, why not enable by default?
to avoid silent growth in the set of code built into somebody's server... when
does somebody have to add --disable-foo to create a build compatible with
what they had with the 1.3.(n-1) release?
(I
Brian Akins wrote:
Call me stupid, put why in various places does Apache do things like this:
if (csd = FD_SETSIZE) {
ap_log_error(APLOG_MARK, APLOG_WARNING, 0, NULL,
new file descriptor %d is too large; you probably
need
to rebuild Apache with a
Jeff Trawick wrote:
We use hsregex on older Solaris (2.0-2.5 or something like that).
Theoretically maybe something stops working or starts working when the
switch is made, but beyond that theoretical possibility does anyone have
real knowledge that there is a non-trivial likelihood
[EMAIL PROTECTED] wrote:
stoddard2003/12/15 15:19:14
Modified:server/mpm/winnt child.c mpm_winnt.c mpm_winnt.h
Log:
Win32: Rename WindowsSocketsWorkaround directive to Win32DisableAcceptEx.
Clean up code paths.
Index: mpm_winnt.c
-static const char *set_sockets_workaround
configure-time checks would used to check for availability (certain
Linux only); the syscall would be made in the same situations as 2.x (if
CoredumpDirectory has been set and we're starting as root)
no patch yet, just wondering if anybody objects for some reason and/or
there is a chance of
(originally posted to wrong mailing list)
---BeginMessage---
2.x already does this
Index: src/modules/standard/mod_mime_magic.c
===
RCS file: /home/cvs/apache-1.3/src/modules/standard/mod_mime_magic.c,v
retrieving revision 1.51
diff
Rather than using multiple symbols (HAVE_SYS_PRCTL_H, HAVE_PRCTL), which
would add to the CFLAGS, there is a single symbol HAVE_SET_DUMPABLE
which is defined via CFLAGS if all prerequisites are met.
testing:
Fedora Core 1: verified that feature was recognized and that the new
code was
Bill Stoddard wrote:
Do you know of any cases that actually require mpm_state to be updated
in ap_signal_parent()? Setting winnt_mpm_state to AP_MPMQ_STOPPING in
child main should be sufficient unless I am missing something.
the code in service.c which shuts down the server for other reasons?
(or consider this a bitch at the mess we call ab; just cleaning out some
of my source trees of potentially useful changes)
if errno != 0, err() prints the errno string... yet err() is called
sometimes when errno is irrelevant; large-scale rework of error handling
would be nice, but this at
Bill Stoddard wrote:
Checkout this patch. I -know- state AP_MPMQ_RESTARTING does not exist in
ap_mpm.h (so this patch will not compile) but consider it for purposes
of discussion. The parent process always goes through master_main() and
master_main() can distinguish between stopping and
Min Xu(Hsu) wrote:
On Thu, 15 Jan 2004 Erik Abele wrote :
Have a look at the old gnats bug-archive (ranging back to '96):
http://archive.apache.org/gnats/
Thanks! I was initially confused by the big number of modules. I guess
the module general should be my focus.
On the other hand, is it true
Brian Akins wrote:
Rodent of Unusual Size wrote:
* Replace some of the mutex locking in the worker MPM with
atomic operations for higher concurrency.
server/mpm/worker/fdqueue.c 1.24, 1.25
+1: brianp, ianh
I submitted a patch some time ago to do this and got little
Min Xu(Hsu) wrote:
I did a little more research, it seems the data race bugs show up usually
in some specific module with specific conditions. I wonder how can I setup
my server to test those unusual conditions quickly? Or, even better, are
there any known crashes caused by unknown race
Henri Gomez wrote:
I'd like to use an Apache 2.0.x one (for mod_deflate) and wonder
if there is such RPM / SPEC available somewhere (I've got
problem with SUSE .spec).
no idea
Alternativly, did there is some specific stuff for this PowerPC
box and what's the prefered mode (worker or prefork ?)
See http://www.apache.org/~trawick/exception_hook_13.html
There is a small patch to Apache 1.3 required to make the sample modules
work. This is analogous to the toys using the Apache 2.1 exception hook
which are described at http://www.apache.org/~trawick/exception_hook.html.
Ben Laurie wrote:
Jeff Trawick wrote:
See http://www.apache.org/~trawick/exception_hook_13.html
You should make the logged strings safe, like mod_log_forensic does, and
I think the format should be compatible (which means no space after the
colon).
Thanks for taking a look!
I removed the space
Ben Laurie wrote:
Jeff Trawick wrote:
Ben Laurie wrote:
You should make the logged strings safe, like mod_log_forensic does,
and I think the format should be compatible (which means no space
after the colon).
Thanks for taking a look!
I removed the space after the colon, but at present am
Jim Jagielski wrote:
Anyway +1 (untested) for the core patch.
+1 (tested) on the core-patch... I'm mulling over whether
it should be included by default or, at least, runtime configurable :)
personally I would like to see one of those two so that users who are
experiencing a problem that could
Bill Stoddard wrote:
What is the purpose of the geteuid() call in run_fatal_exception_hook
and when might it return 0?
geteuid() returns 0 when running with root privileges
the purpose of not running the hook as root is a nod to the possibility
that the hook could be hijacked (dunno how since
Mads Toftum wrote:
On Thu, Jan 22, 2004 at 06:48:46AM -0500, Jeff Trawick wrote:
anyone else lurking?
A very nice idea - seen with my #apache hat on, it'd be really
nice to get this when people turn up complaining about segfaults.
With my old admin and security paranoid hats on, it would
Stas Bekman wrote:
Jeff Trawick wrote:
I don't know much about mod_ssl, but I'd guess the issue is in
ssl_engine_io.c:ssl_io_filter_output().
Other filters such as content-length filter know that it is nice to
flush data already read to the network when a bucket read would block.
However
Joe Orton wrote:
Index: ssl_engine_io.c
===
RCS file: /home/cvs/httpd-2.0/modules/ssl/ssl_engine_io.c,v
retrieving revision 1.113
diff -u -r1.113 ssl_engine_io.c
--- ssl_engine_io.c 1 Jan 2004 13:26:21 - 1.113
+++
Jeff Trawick wrote:
See http://www.apache.org/~trawick/exception_hook_13.html
There is a small patch to Apache 1.3 required to make the sample modules
work. This is analogous to the toys using the Apache 2.1 exception hook
which are described at http://www.apache.org/~trawick
Jeff Trawick wrote:
Jeff Trawick wrote:
See http://www.apache.org/~trawick/exception_hook_13.html
1) automatically define AP_ENABLE_EXCEPTION_HOOK in ap_config.h for
platforms that have been tested (AIX, Linux, Solaris already and I'll
test HP-UX before long)
2) user can still enable
Mads Toftum wrote:
On Sat, Jan 24, 2004 at 07:23:24AM -0500, Jeff Trawick wrote:
url above now points to patch which implements these changes
vote added to apache-1.3/STATUS
Segfaults here (linux-2.4.23_pre8 - --enable-module=most
--enable-shared=max)
With mod_backtrace_13 enabled:
#0
Mads Toftum wrote:
On Sat, Jan 24, 2004 at 07:23:24AM -0500, Jeff Trawick wrote:
url above now points to patch which implements these changes
vote added to apache-1.3/STATUS
Segfaults here (linux-2.4.23_pre8 - --enable-module=most
--enable-shared=max)
With mod_backtrace_13 enabled:
#0
Min Xu(Hsu) wrote:
On Thu, 15 Jan 2004 Jeff Trawick wrote :
data race? consider http://nagoya.apache.org/bugzilla/show_bug.cgi?id=25520
Thanks. I was able to reproduce this one.
hopefully without the fix which I subsequently committed to 2.1-dev :)
(gotta propose that one for backport I
Colm MacCarthaigh wrote:
On Mon, Jan 26, 2004 at 06:28:03PM +, Colm MacCarthaigh wrote:
I'd love to find out what's causing your worker failures. Are you using
any thread-unsafe modules or libraries?
Not to my knowledge, I wasn't planning to do this till later, but
I've bumped to 2.1, I'll
Aryeh Katz wrote:
# diff -u http_protocol.old.h http_protocol.h
--- http_protocol.old.h
+++ http_protocol.h
@@ -528,7 +528,7 @@
* @param r The current request
* @param pw The password as set in the headers
* @return 0 (OK) if it set the 'pw' argument (and assured
- * a correct value
Tikka, Sami wrote:
[error] (OS 10048)Only one usage of each socket address (protocol/network
address/port) is normally permitted. : proxy: HTTP: attempt to connect to
1.2.3.4:80 (1.2.3.4) failed
... hundreds of megabytes of these errors.
10048 is WSAEADDRINUSE. It seems to happen in
worker MPM stack corruption in parent:
int free_slots[MAX_SPAWN_RATE];
...
/* great! we prefer these, because the new process can
* start more threads sooner. So prioritize this slot
* by putting it ahead of any slots with active threads.
Jeff Trawick wrote:
worker MPM stack corruption in parent:
int free_slots[MAX_SPAWN_RATE];
...
/* great! we prefer these, because the new process can
* start more threads sooner. So prioritize this slot
* by putting it ahead of any
Colm MacCarthaigh wrote:
On Mon, Jan 26, 2004 at 10:05:33PM +, Colm MacCarthaigh wrote:
disable the check for geteuid()==0 and see if you get backtrace?
exception hook purposefully doesn't run as root (I assume your parent is
running as root)
No problem, first thing tomorrow :)
O.k.,
Colm MacCarthaigh wrote:
On Tue, Jan 27, 2004 at 02:24:46PM -0500, Jeff Trawick wrote:
I'm testing with this patch currently (so far so good):
Same here, I've applied the patch, and right now have 1 hours uptime,
which is 12 times more than I've ever had with worker before.
Looks like
1 - 100 of 3135 matches
Mail list logo
