Hi,

I'm running tomcat 5.0.28 with mod_jk 1.2.10 and
apache 2.0.46 with the worker MPM. This is a new
setup, and for the most part everything works fine,
but I've had a couple reports of truncated pages being
returned to customers. After turning on debug logging
for jk, I see the below sequence often in the logs,
but from what I've read elsewhere it's most likely
just a browser being closed or a new request being
sent from the browser before the previous one
completes. Can anyone who's seen this before verify
whether this could be indicative of a real problem? 
My workers.properties and AJP connector from
server.xml are below as well. 

Thanks,
Lou

>From mod_jk.log:

[Wed May 11 12:30:37 2005] [2244:39856] [debug]
ajp_connection_tcp_get_message::jk_ajp_common.c
(1024): received from ajp13 pos=0 len=8188 max=8192
[Wed May 11 12:30:37 2005] [2244:39856] [debug]
ws_write::mod_jk.c (380): writing 4096 (4096) out of
8184
[Wed May 11 12:30:37 2005] [2244:39856] [debug]
ws_write::mod_jk.c (380): writing 4088 (4088) out of
4088
[Wed May 11 12:30:37 2005] [2244:39856] [debug]
ajp_connection_tcp_get_message::jk_ajp_common.c
(1024): received from ajp13 pos=0 len=12 max=8192
[Wed May 11 12:30:37 2005] [2244:39856] [debug]
ws_write::mod_jk.c (380): writing 8 (-1) out of 8
[Wed May 11 12:30:37 2005] [2244:39856] [info] 
ajp_process_callback::jk_ajp_common.c (1363):
Connection aborted or network problems
[Wed May 11 12:30:37 2005] [2244:39856] [info] 
ajp_service::jk_ajp_common.c (1706): Receiving from
tomcat failed, because of client error without
recovery in send loop 0
[Wed May 11 12:30:37 2005] [2244:39856] [debug]
ajp_done::jk_ajp_common.c (2046): recycling connection
cache slot=4 for worker local
[Wed May 11 12:30:37 2005] [2244:39856] [info] 
jk_handler::mod_jk.c (1968): Aborting connection for
worker=local


workers.properties:

worker.list=local
worker.local.type=ajp13
worker.local.host=localhost
worker.local.port=8009
                                                      
                         
                                                      
                         
# Number of connections made to the AJP backend that
are maintained as a connection pool. It does not limit
the number of those connection.
# Should reflect the number of threads per child
process in Apache2 conf.
worker.local.cachesize=50
                                                      
                         
                                                      
                         
# Used with cachesize to specify how long JK should
keep an open socket in cache before closing it.
# recycle_timeout similar to cache_timeout but works
for uncached connections also
worker.local.cache_timeout=300
worker.local.recycle_timeout=300
                                                      
                         
                                                      
                         
# Socket timeout in seconds used for communication
channel between JK and remote host.
worker.local.socket_timeout=300


server.xml AJP connector:

                <Connector port="8009"
protocol="AJP/1.3" enableLookups="false"
                        redirectPort="8443"
acceptCount="100" maxThreads="900"
                        minSpareThreads="25"
maxSpareThreads="200"
                        maxKeepAliveRequests="1"
connectionTimeout="300000"
                        disableUploadTimeout="true"
debug="0"/>




---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to