RE: mod_jk 1.2.0/1.1.0 vs ab

2001-10-05 Thread GOMEZ Henri
AIL PROTECTED] >Subject: Re: mod_jk 1.2.0/1.1.0 vs ab > > >Bojan Smojver wrote: >> >> jean-frederic clere wrote: >> >> > And in mod_jk.log file is there something? >> >> Mostly something like this: >> >>

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-05 Thread jean-frederic clere
Bojan Smojver wrote: > > jean-frederic clere wrote: > > > And in mod_jk.log file is there something? > > Mostly something like this: > > - > [Fri Oct 05 17:07:55 2001] [jk_ajp_common.c (914)]: Error > ajp_process_callback - write fai

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread Bojan Smojver
jean-frederic clere wrote: > And in mod_jk.log file is there something? Mostly something like this: - [Fri Oct 05 17:07:55 2001] [jk_ajp_common.c (914)]: Error ajp_process_callback - write failed [Fri Oct 05 17:07:55 2001] [jk_ajp_co

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread jean-frederic clere
Bojan Smojver wrote: > > Bojan Smojver wrote: > > > Since I can control the headers from my servlet, let my try with > > Content-length. Fingers crossed... > > If think this can actually qualify as a bug in ab - even when I set > Content-Length header, it still says that that there are length >

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread Bojan Smojver
Bojan Smojver wrote: > Since I can control the headers from my servlet, let my try with > Content-length. Fingers crossed... If think this can actually qualify as a bug in ab - even when I set Content-Length header, it still says that that there are length failures. Since jsessionid can sometime

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread Bojan Smojver
[EMAIL PROTECTED] wrote: > I think ( or guess ) that ab is checking the length of the first request, > and if following requests have different lengths it assumes it's a > failure. > > Could you check if your page returns the same thing ? Very strange.. I ran the thing with -v 99 and it shows o

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread cmanolache
On Fri, 5 Oct 2001, Bojan Smojver wrote: > Bojan Smojver wrote: > > > > Bojan Smojver wrote: > > > > > This goes OK on both 1.1.0 and 1.2.0, although some requests aren't > > > served (probably because no more threads are available in ajp13 > > > connector - I run a max of 50). > > > > Has pro

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread Bojan Smojver
Bojan Smojver wrote: > > Bojan Smojver wrote: > > > This goes OK on both 1.1.0 and 1.2.0, although some requests aren't > > served (probably because no more threads are available in ajp13 > > connector - I run a max of 50). > > Has probably very little to do with it. I tried more threads, but s

Re: mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread Bojan Smojver
Bojan Smojver wrote: > This goes OK on both 1.1.0 and 1.2.0, although some requests aren't > served (probably because no more threads are available in ajp13 > connector - I run a max of 50). Has probably very little to do with it. I tried more threads, but some requests still fail according to a

mod_jk 1.2.0/1.1.0 vs ab

2001-10-04 Thread Bojan Smojver
Just for fun I did this to exercise mod_jk/TC 3.3 combination: --- ab -c 1 -n 1000 http://some/velocity/page.vm --- This goes OK on both 1.1.0 and 1.2.0, although some requests aren't served (probably because no more threads