On 15.04.2009 08:19, Munkhbold.B wrote: > Hi > > > > I have a problem with tomcat connector mod_jk. > > From time to time the connector completely hangs apache. > > Tomcat alone is still alive, but apache no longer replies to requests, and I > need to restart both > > > > In mod_jk logs file I have a lot of lines like > > > > [info] init_jk::mod_jk.c (3183): mod_jk/1.2.28 initialized > > [info] jk_handler::mod_jk.c (2608): Aborting connection for worker=node1 > > [info] ajp_process_callback::jk_ajp_common.c (1788): Writing to client > aborted or client network problems > > [info] ajp_service::jk_ajp_common.c (2447): (node1) sending request to > tomcat failed (unrecoverable), because of client write error (attempt=1) > > [info] jk_handler::mod_jk.c (2608): Aborting connection for worker=node1 > > [info] ajp_process_callback::jk_ajp_common.c (1788): Writing to client > aborted or client network problems > > [info] ajp_service::jk_ajp_common.c (2447): (node1) sending request to > tomcat failed (unrecoverable), because of client write error (attempt=1) > > [info] jk_handler::mod_jk.c (2608): Aborting connection for worker=node1 > > [info] ajp_process_callback::jk_ajp_common.c (1788): Writing to client > aborted or client network problems > > [info] ajp_service::jk_ajp_common.c (2447): (node1) sending request to > tomcat failed (unrecoverable), because of client write error (attempt=1) > > [info] jk_handler::mod_jk.c (2608): Aborting connection for worker=node1 > > [info] ajp_process_callback::jk_ajp_common.c (1788): Writing to client > aborted or client network problems > > [info] ajp_service::jk_ajp_common.c (2447): (node1) sending request to > tomcat failed (unrecoverable), because of client write error (attempt=1) > > [info] jk_handler::mod_jk.c (2608): Aborting connection for worker=node1
What the other list members were answering is correct. I myself would expect you have a performance or stability problem in your webapp, so that requests queue up in front of Tomcat, fill up Apache httpd and thus you get that httpd is no longer available and requests making it to the web server take a long time. Whether this is correct or not you can check by doing thread dumps of Tomcat during this situation and before doing the restarts. Google for "java thread dump" (I hope this is a good search term). A roughly correct formula is: Concurrency = Throughput * ResponseTime I expect your Response time in the webapp inctases a lot, e.g. because some other system (DB, Mainframe, ...) is slow or your webapp even deadlocks, and as a consequence the concurrency inreases until all available threads in Tomcat and Apache get exhausted. Regards, Rainer --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
