Hi All, My Apache 2.2 (win32) + Tomcat 6 are running very well in my production environment but I am noticed that jk are sometime spend unexpected long time to response. (4-5 seconds, it's too long. Those pages are very static with no database connection, no file io).
Here is my environment: Apache/2.2.6 (Win32) mod_ssl/2.2.6 OpenSSL/0.9.8e mod_jk/1.2.26 with AcceptEx() WinSock2 API disabed Tomcat and Apache are in same server (Windows 2000 Server) Here is my worker: # Define 1 real worker using ajp13 worker.list=worker1 # Set properties for worker1 (ajp13) worker.worker1.type=ajp13 worker.worker1.host=localhost worker.worker1.port=8009 worker.worker1.socket_timeout=300 worker.worker1.socket_keepalive=True worker.worker1.connection_pool_timeout=600 Here are what I found in my mod_jk.log: [Mon Feb 04 10:50:12 2008] [3068:8252] [info] jk_ajp_common.c (1347): (worker1) all endpoints are disconnected, detected by connect check (5), cping (0), send (0) [Mon Feb 04 10:51:46 2008] [3068:9592] [info] jk_ajp_common.c (1347): (worker1) all endpoints are disconnected, detected by connect check (6), cping (0), send (0) [Mon Feb 04 10:52:47 2008] [3068:9804] [info] jk_ajp_common.c (1347): (worker1) all endpoints are disconnected, detected by connect check (4), cping (0), send (0) // here is source code of jk_ajp_common.c /* * If we failed to reuse a connection, try to reconnect. */ if (!IS_VALID_SOCKET(ae->sd)) { /* Could not steal any connection from an endpoint - backend is disconnected */ if (err_conn + err_cping + err_send > 0) jk_log(l, JK_LOG_INFO, "(%s) all endpoints are disconnected, " "detected by connect check (%d), cping (%d), send (%d)", ae->worker->name, err_conn, err_cping, err_send); else if (JK_IS_DEBUG_LEVEL(l)) jk_log(l, JK_LOG_DEBUG, "(%s) all endpoints are disconnected, " "detected by connect check (%d), cping (%d), send (%d)", ae->worker->name, err_conn, err_cping, err_send); /* Connect to the backend. */ } Is it mean that my jk always fail to reuse existing jk connections? if yes, why? Any solution? It's not a big problem but It will affect my bechmark. Regards, Ken -- View this message in context: http://www.nabble.com/mod_jk-all-endpoints-are-disconnected-tp15261648p15261648.html Sent from the Tomcat - User mailing list archive at Nabble.com. --------------------------------------------------------------------- To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]