Bug 40907

Summary: mod_jk:1.2.19 because of client write error
Product: Tomcat Connectors Reporter: paul hettl <paul.hettl>
Component: CommonAssignee: Tomcat Developers Mailing List <dev>
Status: RESOLVED INVALID    
Severity: normal    
Priority: P2    
Version: unspecified   
Target Milestone: ---   
Hardware: Other   
OS: Linux   

Description paul hettl 2006-11-06 04:39:49 UTC 
I observe continous  logging of error in  mode="info" in mod_jk.log 

Server Version: Apache/2.0.59 (Unix) mod_ssl/2.0.59 OpenSSL/0.9.7d mod_jk/1.2.19

It seams, that the loadbalancer drops client connections


--------  LOG ------
Mon Nov 06 13:28:09 2006] [0345:47744] [info]  jk_handler::mod_jk.c (2056):
Aborting connection for worker=loadbalancer

[Mon Nov 06 13:36:14 2006] [0692:47744] [info] 
ajp_process_callback::jk_ajp_common.c (1410): Writing to client aborted or
client network
problems

[Mon Nov 06 13:36:14 2006] [0692:47744] [info]  ajp_service::jk_ajp_common.c
(1795): (www2) request failed, because of client write error
without recovery in send loop attempt=0

[Mon Nov 06 13:36:14 2006] [0692:47744] [info]  service::jk_lb_worker.c (873):
unrecoverable error 200, request failed. Client failed in t
he middle of request, we can't recover to another instance.

[Mon Nov 06 13:36:14 2006] [0692:47744] [info]  jk_handler::mod_jk.c (2056):
Aborting connection for worker=loadbalancer
Comment 1 paul hettl 2006-11-06 04:41:19 UTC 
Worker file:

# The workers that jk should create and work with
worker.list=loadbalancer, www1,www2,jkstatus


# Defining a worker named www1 and of type ajp13
worker.www1.type=ajp13
worker.www1.host=www1
worker.www1.port=8009
worker.www1.lbfactor=1
worker.www1.cachesize=1


# Defining a worker named www2 and of type ajp13
worker.www2.type=ajp13
worker.www2.host=www2
worker.www2.port=8009
worker.www2.lbfactor=1
worker.www2.cachesize=1


# Defining a loadbalancer
worker.loadbalancer.type=lb
worker.loadbalancer.balanced_workers=www1,www2
worker.loadbalancer.method=B
worker.loadbalancer.sticky_session=1

# status worker
worker.jkstatus.type=status
Comment 2 Mladen Turk 2006-11-06 04:55:17 UTC 
This is normal operation when the clinet closes the browser in the middle of
request.

Please use Tomcat Users list before filing and [info] or [warn] messages.
Comment 3 paul hettl 2006-11-06 05:02:46 UTC 
(In reply to comment #2)
> This is normal operation when the clinet closes the browser in the middle of
> request.
> 
> Please use Tomcat Users list before filing and [info] or [warn] messages.
>
Comment 4 paul hettl 2006-11-06 05:05:27 UTC 
Hi sorry for that,  but check the bug i found now , that seams to have the same
status:

http://issues.apache.org/bugzilla/show_bug.cgi?id=35461

I cannot reproduce that the closing of the clinet app causes this strange logging.