[users@httpd] Strange problem with mod_jk2

[Praveen Alavilli <Pr...@aol.com>]

Hi,
We are seeing a strange problem with Apache/2.0.53 + Tomcat 4.1.30 
connected using mod_jk2. Occasionally all the threads in one of the 
Apache processes (using worker mpm)  gets into "*|W|*" Sending Reply 
mode (as seen on the server-status page) forever taking up a lot of 
memory and cpu. They pretty much stay in that state until that process 
is killed.

When we looked at the core image, all the threads are showing the same 
call stack as below waiting in the jk2_close_endpoint() function:

#0  0xb7492c5b in apr_pool_cleanup_kill () from libapr-0.so.0
#1  0xb7492ce3 in apr_pool_cleanup_run () from libapr-0.so.0
#2  0xb748ae7f in apr_socket_close () from libapr-0.so.0
#3  0xb71f3b99 in jk2_channel_apr_close () from mod_jk2.so
#4  0xb72134fc in __JCR_LIST__ () from mod_jk2.so
#5  0x3b3e77a8 in ?? ()
#6  0xb7203986 in jk2_close_endpoint () from mod_jk2.so


This happend more than once and the core images are always the same. Any 
one got any clues ? To me it looks like mod_jk2 is waiting for the 
Tomcat to close the connection or something. Could it be some race 
condition between the apache & tomcat connection management ?

any help would be greatly appreciated.

thanks
Praveen

---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org