> On 19.03.2009 14:02, Jeff Fulmer wrote: >> >> If I stop the tomcat server tomcat2, there's no problem. It properly >> closes the sockets and mod_jk/apache/linux knows the server went away. >> But if I shutdown the network on tomcat2 (or yank the cable), mod_jk >> continues to send packets to it. My socket table is filled with >> connections to tomcat2:8009 in SYN_SENT. Apache continues to send every >> other request to tomcat2. Since I use session stickiness, this only >> effects half the users but still. > > This should be handled by the socket_timeout, which you set to 2 minutes. > How does it behave, if you reduce your socket timeout say to 10 seconds?
MAJOR IMPROVEMENT. Here's my revised workers.properties file: worker.list=balancer,status # Definte default worker settings worker.default.port=8009 worker.default.type=ajp13 worker.default.lbfactor=1 worker.default.socket_timeout=10 worker.default.reply_timeout=5000 worker.default.socket_keepalive=True worker.default.ping_mode=P worker.default.ping_timeout=5000 worker.default.connection_pool_timeout=600 worker.default.connection_pool_minsize=0 #------WORKER1------- worker.worker1.reference=worker.default worker.worker1.host=10.248.11.15 #------WORKER2------- worker.worker2.reference=worker.default worker.worker2.host=10.248.11.16 #------BALANCER------ worker.balancer.type=lb worker.balancer.balance_workers=worker1,worker2 # Load balancing method can be [R]equest, [S]ession, [T]raffic, or [B]usyness worker.balancer.method=R # How runtime data are sync-ed. [O]ptimistic or [P]essimistic worker.balancer.lock=O worker.balancer.sticky_session=True worker.balancer.max_reply_timeouts=5 # Status worker for managing load balancer worker.status.type=status And in my virtual host I have this: <IfModule mod_jk.c> JkOptions +JkWatchdogInterval JkMount /rp/* balancer JkMount /rp balancer JkMount /env/* balancer JkMount /env balancer JkMount /jkmanager/* status </IfModule> BTW: /jkmanager/* is at the server level, but I included it here for Rainer... Here's my remaining problem. After I shut down the network on tomcat2, I have a few 30 second request transactions but each ensuing transaction is pretty zippy. If I leave my browser sit idly for about a minute, I have two 30 second transactions and each ensuing transaction is zippy. I suspect I have two back-to-back 30 second transactions because the load balancer sends me to apache1 then apache2 and each tries to reconnect to the downed tomcat. After that fails, they keep me on the active tomcat. Maybe this one be a problem once the system is taking live transactions but it would be nice if I could increase the intervals that apache uses to reconnect to the downed tomcat. Any thoughts? Jeff --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
