Greg Ames wrote:
Sander Temme wrote:
Tested on Darwin under light load and Solaris under moderate load.
Thoughts,
opinions, comments? Or shall I just commit the thing?
I'll try it on the Linux PPC machine where I saw the post log
flakyness and report back. I wouldn't have a problem if you just
Sander Temme wrote:
All,
Months ago, Madhu posted a patch to put thread-safe locks around the Post
logging in mod_specweb99... here's an alternative using apr_global_mutex
locks:
excellent! I was running specweb99 earlier this week and was seeing
post log bogosity with worker on Linu
All,
Months ago, Madhu posted a patch to put thread-safe locks around the Post
logging in mod_specweb99... here's an alternative using apr_global_mutex
locks:
Index: mod_specweb99.c
===
RCS file: /home/cvs/httpd-test/spec
PM
>To: [EMAIL PROTECTED]
>Subject: Re: SPEC / mod_specweb99.c
>
>
>MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
>> Hi,
>> Does anybody know if the mod_specweb99.c been 'blessed'
>by the SPEC
>> committee ?..I mean, have they acknowledged that the modul
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
Hi,
Does anybody know if the mod_specweb99.c been 'blessed' by the SPEC
committee ?..I mean, have they acknowledged that the module acts in a SPEC
compliant manner?
No, they have not blessed it.
I had a doubt regarding executing C
> I had a doubt regarding executing CGI scripts in SPECweb99 -
somebody here
> told me that the SPEC mandates the web server to fork a child
process to
> execute a CGI script - is that so ?.
Speaking of cgi scripts - be sure to use the C based one for this, the
perl based reference is quite a bi
> Does anybody know if the mod_specweb99.c been 'blessed' by the SPEC
> committee ?..I mean, have they acknowledged that the module acts in a SPEC
> compliant manner?
As far as I am aware, the module has not been used for benchmarks submitted
to the SPEC organization. That is
Hi,
Does anybody know if the mod_specweb99.c been 'blessed' by the SPEC
committee ?..I mean, have they acknowledged that the module acts in a SPEC
compliant manner?
I had a doubt regarding executing CGI scripts in SPECweb99 - somebody here
told me that the SPEC mandates the web
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
Okay.. Here it comes (with the changes suggested by Greg), and a couple more
fixes that I'd overseen earlier.
Madhu,
Could you take a look at what's in CVS now and see if there are changes that
should be made to that? I'm sorry if you missed my post
Greg Ames wrote:
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
As regards the function names at the end of the function, I sure like
it -
it's a lot easier when you browse through the code.. But then, I
thought it
should be there for all of them or for none of them.. That's the reason I
decided
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
As regards the function names at the end of the function, I sure like it -
it's a lot easier when you browse through the code.. But then, I thought it
should be there for all of them or for none of them.. That's the reason I
decided to take it away..
e for mod_specweb99.c
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
> Summary of the patch :
> 1. convert tabs to spaces
> 2. try to follow apache styleguide
Overall, a big improvement...thanks much!
nitpicky comments follow:
> --- mod_specweb99.c 15 Jan 2003 16:15:0
2003 11:26 AM
To: [EMAIL PROTECTED]
Subject: Re: [PATCH] Style police for mod_specweb99.c
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
> Summary of the patch :
> 1. convert tabs to spaces
> 2. try to follow apache styleguide
Overall, a big improvement...thanks much!
nitpicky comments follo
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
Summary of the patch :
1. convert tabs to spaces
2. try to follow apache styleguide
Overall, a big improvement...thanks much!
nitpicky comments follow:
--- mod_specweb99.c 15 Jan 2003 16:15:02 - 1.18
+++ mod_specweb99.c 21 Jan 2003
On Wed, 22 Jan 2003, Greg Ames wrote:
> MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
> > Summary of the patch :
> > 1. convert tabs to spaces
>
> hmmm, I wonder how I missed 'em? IIRC I used "expand -i -t4 mod_specweb99.c"
>
> > 2. try to fol
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
Summary of the patch :
1. convert tabs to spaces
hmmm, I wonder how I missed 'em? IIRC I used "expand -i -t4
mod_specweb99.c"
2. try to follow apache styleguide
I plan on reviewing this, but I'm pretty weak on style other t
Summary of the patch :
1. convert tabs to spaces
2. try to follow apache styleguide
[ Please note that there might be long lines because of the old style
followed in mod_specweb99.c ]
Thanks
-Madhu
<>
Index: mod_specw
Sander Temme wrote:
Does anyone have a perl script etc. to automatically convert all leading tabs
to
n blanks?
Doesn't indent do the trick, with the .indent.pro from somewhere in the
httpd-2.0 tree?
Roy T. Fielding wrote:
> Most Unices have a command called "expand" that does just that.
De-tab
Does anyone have a perl script etc. to automatically convert all
leading tabs to n blanks? I was thinking of manually de-tabifying in
the vicinity of this fix, but it would be better to do the whole thing
if I can get my hands on such a tool.
Most Unices have a command called "expand" that does
> Does anyone have a perl script etc. to automatically convert all leading tabs
> to
> n blanks? I was thinking of manually de-tabifying in the vicinity of this
> fix,
> but it would be better to do the whole thing if I can get my hands on such a
> tool.
Doesn't indent do the trick, with the .i
[EMAIL PROTECTED] wrote:
--- mod_specweb99.c 31 Oct 2002 19:39:04 - 1.16
+++ mod_specweb99.c 14 Jan 2003 15:02:00 - 1.17
@@ -765,8 +765,9 @@
line = apr_psprintf(r->pool, "%10d\n", 0);
- if (apr_file_write_full(f, line, strlen(line), NULL) !
On Friday, December 13, 2002, at 10:05 AM, <[EMAIL PROTECTED]> wrote:
They are in apr_private.h.
Yes, but they are also in apr.h.
pigmy:~/apr/include - 9:57AM% grep "_IS_GLOBAL" apr.h
#define APR_PROCESS_LOCK_IS_GLOBAL0
#define APR_PROC_MUTEX_IS_GLOBAL 0
pigmy:~/apr/include - 9:57AM
On Fri, 13 Dec 2002, Aaron Bannert wrote:
>
> On Friday, December 13, 2002, at 09:06 AM, <[EMAIL PROTECTED]> wrote:
>
> > On Fri, 13 Dec 2002, Aaron Bannert wrote:
> >
> >>
> >> On Thursday, December 12, 2002, at 12:34 PM, Greg Ames wrote:
> >>> I dug into APR locks a little bit. The apr_glob
ten those calls do not even translate
> into syscalls.
What we're pondering here is locking a logfile in mod_specweb99. During a
SPECWeb99 run, mutexes are not cheap and there is a lot of camping out going
on.
S.
--
Covalent Technologies
On Friday, December 13, 2002, at 09:06 AM, <[EMAIL PROTECTED]> wrote:
On Fri, 13 Dec 2002, Aaron Bannert wrote:
On Thursday, December 12, 2002, at 12:34 PM, Greg Ames wrote:
I dug into APR locks a little bit. The apr_global_mutex_* functions
turn into two separate syscalls, with #if APR_HAS_THRE
On Fri, 13 Dec 2002, Aaron Bannert wrote:
>
> On Thursday, December 12, 2002, at 12:34 PM, Greg Ames wrote:
> > I dug into APR locks a little bit. The apr_global_mutex_* functions
> > turn into two separate syscalls, with #if APR_HAS_THREADS around the
> > thread mutexing. So unfortunately they
On Thursday, December 12, 2002, at 12:34 PM, Greg Ames wrote:
I dug into APR locks a little bit. The apr_global_mutex_* functions
turn into two separate syscalls, with #if APR_HAS_THREADS around the
thread mutexing. So unfortunately they wouldn't save us any syscalls
:-( :-( But they might s
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
-Original Message-
From: Greg Ames [mailto:[EMAIL PROTECTED]
[SNIP]
I dug into APR locks a little bit. The apr_global_mutex_*
functions turn into
two separate syscalls, with #if APR_HAS_THREADS around the
thread mutexing. So
unfortunate
ile. That's an interesting idea. If mod_specweb99 opened the
post log during the post_config hook, or something similar, and the child
processes all inherited the post log fd, they could just write to it. The
problem would be the current record counter and the record number at the
b
>-Original Message-
>From: Greg Ames [mailto:[EMAIL PROTECTED]
[SNIP]
>I dug into APR locks a little bit. The apr_global_mutex_*
>functions turn into
>two separate syscalls, with #if APR_HAS_THREADS around the
>thread mutexing. So
>unfortunately they wouldn't save us any syscalls :-
*/
/* #undef SYSVSEM_IS_GLOBAL */
/* #undef FCNTL_IS_GLOBAL */
/* #undef FLOCK_IS_GLOBAL */
-Madhu
>-Original Message-
>From: Greg Ames [mailto:[EMAIL PROTECTED]
>Sent: Thursday, December 12, 2002 12:35 PM
>To: [EMAIL PROTECTED]
>Subject: Re: [PATCH] Use mutex locks in
Sander Temme wrote:
I started seeing the following errors in the specweb99 run output, when I
use mod_specweb99.c with Apache 2.0.43 and worker MPM. The following patch
seems to get rid of the problem. If you're thinking that it may degrade the
response - I did not find much difference though
MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1) wrote:
I started seeing the following errors in the specweb99 run output, when I
use mod_specweb99.c with Apache 2.0.43 and worker MPM. The following patch
seems to get rid of the problem. If you're thinking that it may degrade the
response - I did not
LI,MADHUSUDAN (HP-Cupertino,ex1)
>Cc: [EMAIL PROTECTED]
>Subject: Re: [PATCH] Use mutex locks in mod_specweb99.c
>
>
>> I started seeing the following errors in the specweb99 run
>output, when I
>> use mod_specweb99.c with Apache 2.0.43 and worker MPM. The
>foll
> I started seeing the following errors in the specweb99 run output, when I
> use mod_specweb99.c with Apache 2.0.43 and worker MPM. The following patch
> seems to get rid of the problem. If you're thinking that it may degrade the
> response - I did not find much difference
I started seeing the following errors in the specweb99 run output, when I
use mod_specweb99.c with Apache 2.0.43 and worker MPM. The following patch
seems to get rid of the problem. If you're thinking that it may degrade the
response - I did not find much difference though.
Can somebody p
[EMAIL PROTECTED] wrote:
sctemme 2002/11/05 16:41:38
Modified:specweb99/specweb99-1.3 mod_specweb99.c
Log:
Disable unfinished semaphore-based IPC for checking modifications to
our database files. This gets the module to compile again.
all right! It's good to see someone
"MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1)" wrote:
>
> I noticed that there were some places where u_int32_t is being used instead
> of apr_uint32_t. Is it purposefully done OR is it one of those "Oh, the apr
> interface changed" stuff ?.
>
> Anyways, I've included a patch that atleast gets the mo
> I noticed that there were some places where u_int32_t is being used instead
> of apr_uint32_t. Is it purposefully done OR is it one of those "Oh, the apr
> interface changed" stuff ?.
>
> Anyways, I've included a patch that atleast gets the module to compile
> against 2.0.43. Pl. let me know if
ow if it's okay.
Thanks,
-Madhu
Index: mod_specweb99.h
===
RCS file:
/home/cvspublic/httpd-test/specweb99/specweb99-2.0/mod_specweb99.h,v
retrieving revision 1.1
diff -u -r1.1 mod_specweb99.h
--- mod_specweb99.h 2 May 2002 20:3
Ian Holsman wrote:
> > Another area for potential improvement here is the use of file caching. I
> > tried
> > using mod_cache's fd caching, but quickly ran into problems with the Linux
> > per-process fd limits because SPECWeb99 accesses so many different files.
> hmm.. did you use a recent
Greg Ames wrote:
Ian Holsman wrote:
Thanks...
what I was after was more hints on configuring HTTP.
ie.. make sure FollowSymLinks is On & AllowOverride off
(to avoid unnessecary fileops) and things like this
Oh, OK, then http://httpd.apache.org/docs-2.0/misc/perf-tuning.html would be
relevant.
y
Ian Holsman wrote:
> Thanks...
> what I was after was more hints on configuring HTTP.
> ie.. make sure FollowSymLinks is On & AllowOverride off
> (to avoid unnessecary fileops) and things like this
Oh, OK, then http://httpd.apache.org/docs-2.0/misc/perf-tuning.html would be
relevant.
I set up
logs proportional to the load.
Thanks...
what I was after was more hints on configuring HTTP.
ie.. make sure FollowSymLinks is On & AllowOverride off
(to avoid unnessecary fileops) and things like this
* I wouldn't be surprised if there are bugs or bottlenecks in mod_specweb99
itself, sin
>> Also - hack the conf to do short runs first !
>
> Oh yeah, excellent point. That's the SPEC rc file. Look for *_TIME,
> ITERATIONS, and SIMULTANEOUS_CONNECTIONS.
My rc files usually have trial (short) values for *_TIME and a single
iteration. You can force all those in line for a compliant r
9 dynamic
content code for other web servers gives me a few ideas too. Since we have a
number of people doing SPECWeb99 benchmarking of Apache 2 at the moment, and
httpd itself is relatively stable (knock on wood), it's a good time to focus on
making mod_specweb99 better.
I'd love to
isk space for logs proportional to the load.
> * I wouldn't be surprised if there are bugs or bottlenecks in mod_specweb99
> itself, since it was only recently open sourced. Please feed back any
> info/patches to this list. But you don't need mod_specweb99 if you are only
> doing st
umber of concurrent connections. There
is a formula here: http://www.spec.org/osg/web99/docs/users_guide.html#Pre-Inst
Plus you need disk space for logs proportional to the load.
* I wouldn't be surprised if there are bugs or bottlenecks in mod_specweb99
itself, since it was only recently open so
server to saturation with this workload.
Naturally, we'd like to see any interesting user space results you come up with
on dev@httpd.apache.org (especially for 2.0) or dev@apr.apache.org, as
appropriate. Some Linux kernel hackers I work with at IBM are also doing
specweb99 benchmarks with 2.0. They
> Brian Pane wrote:
>
> > Do you have any profile data that shows where the bottlenecks are?
>
> No, sorry. At the moment I'm focusing on mod_specweb99.
>
> > From recent tests with other workloads, I anticipate that the most
> > expensive operations are
Brian Pane wrote:
> Do you have any profile data that shows where the bottlenecks are?
No, sorry. At the moment I'm focusing on mod_specweb99.
> From recent tests with other workloads, I anticipate that the most
> expensive operations are likely to be: reading th
Greg Ames wrote:
But I can mention that my very unofficial mini-SPECweb99 runs with the client
and server both on my ThinkPad with 100% "standard dynamic GETs"* show that
prefork is the fastest, worker is about 1% slower, and leader is about another
1.5% slower. This is a noticeable improvement fr
Brian Pane wrote:
>
> [EMAIL PROTECTED] wrote:
>
> >gregames2002/06/03 11:05:50
> >
> > Modified: specweb99/specweb99-2.0 mod_specweb99.c
> >
>
> BTW, does anyone have SPECweb results for 2.0 that they're
> able to discuss?
Not that can b
[EMAIL PROTECTED] wrote:
gregames2002/06/03 11:05:50
Modified:specweb99/specweb99-2.0 mod_specweb99.c
BTW, does anyone have SPECweb results for 2.0 that they're
able to discuss?
Thanks,
--Brian
On Thu, 2 May 2002, Rodent of Unusual Size wrote:
> > What do folks think about adding mod_specweb99 (attached) as
> > an Apache httpd-test component? It is a module which allows
> > you to benchmark Apache 2.0 or 1.3 using the SPECweb99
> > benchmarking suite, describe
Greg Ames wrote:
>
> What do folks think about adding mod_specweb99 (attached) as
> an Apache httpd-test component? It is a module which allows
> you to benchmark Apache 2.0 or 1.3 using the SPECweb99
> benchmarking suite, described at http://www.spec.org/osg/web99/ .
+1
--
#k
Justin Erenkrantz wrote:
>
> On Wed, May 01, 2002 at 06:45:58PM -0400, Greg Ames wrote:
> > Do you mean Dirk and I automagically have httpd-test commit
> > access by virtue of being httpd committers? If so, great!
>
> AIUI, yup.
Yes, that is correct. Other committers may be added to httpd-test
On Wed, 1 May 2002, Greg Ames wrote:
> > If we need to arrange for people without httpd commit to get commit
> > to just httpd-test, we can probably do so... -- justin
>
> Do you mean Dirk and I automagically have httpd-test commit access by
> virtue of being httpd committers? If so, great!
Yep
On Wed, 1 May 2002, Greg Ames wrote:
> KRISTI FOREE wrote:
> >
> > i would like more info about this please
>
> There is information about how to install mod_specweb99 in the README files
> inside the tarball. Is that the kind of info you're looking for?
Bear in
Going out on a limb - I've just submitted to the Secretariat of the ASF
the contribution paperwork. So should this be accepted by the ASF then
know that all paperwork has been filed - and you are free to import it
into CVS.
Dw
--
Dirk-Willem van Gulik
all
> the work he did to get mod_specweb99 working on 2.0.35. I imagine Sander
> would
> be interested as well.
I probably agree. I'm not sure if we need to bring this up on pmc@
or whether we deal with it here on [EMAIL PROTECTED] I think in the past we
handled it here on test-de
KRISTI FOREE wrote:
>
> i would like more info about this please
There is information about how to install mod_specweb99 in the README files
inside the tarball. Is that the kind of info you're looking for?
Greg
> Greg Ames wrote:
> > What do folks think about adding mod
Justin Erenkrantz wrote:
>
> On Wed, May 01, 2002 at 05:54:00PM -0400, Greg Ames wrote:
> > What do folks think about adding mod_specweb99 (attached) as an Apache
> > httpd-test component? It is a module which allows you to benchmark Apache
> > 2.0
> > or 1.3 us
On Wed, May 01, 2002 at 05:54:00PM -0400, Greg Ames wrote:
> What do folks think about adding mod_specweb99 (attached) as an Apache
> httpd-test component? It is a module which allows you to benchmark Apache 2.0
> or 1.3 using the SPECweb99 benchmarking suite, described at
> http://
i would like more info about this please
Greg Ames wrote:
> What do folks think about adding mod_specweb99 (attached) as an Apache
> httpd-test component? It is a module which allows you to benchmark Apache 2.0
> or 1.3 using the SPECweb99 benchmarking suite, described at
> http://
What do folks think about adding mod_specweb99 (attached) as an Apache
httpd-test component? It is a module which allows you to benchmark Apache 2.0
or 1.3 using the SPECweb99 benchmarking suite, described at
http://www.spec.org/osg/web99/ .
While there certainly are improvements that could be
66 matches
Mail list logo
