DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUGĀ· RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT <http://issues.apache.org/bugzilla/show_bug.cgi?id=38869>. ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED ANDĀ· INSERTED IN THE BUG DATABASE.
http://issues.apache.org/bugzilla/show_bug.cgi?id=38869 [EMAIL PROTECTED] changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |INVALID ------- Additional Comments From [EMAIL PROTECTED] 2006-03-06 21:41 ------- It works. I tried with Solaris 10, mod_jk 1.2.15 (!) and apache 2.0.55. The name of the attribute for a worker in workers.properties file (set by JkWorkersFile) is socket_keepalive, like in worker.list=single worker.single.type=ajp13 worker.single.host=mytomcat worker.single.port=21109 worker.single.socket_keepalive=1 If I set it like that and connect a httpd process (apache I used 2.0.55) to tomcat, then I can see in Solaris pfiles: ... 17: S_IFSOCK mode:0666 dev:294,0 ino:47164 uid:0 gid:0 size:0 O_RDWR SOCK_STREAM SO_KEEPALIVE,SO_SNDBUF(49152),SO_RCVBUF(49152) sockname: AF_INET 192.168.0.220 port: 33689 peername: AF_INET 192.168.0.220 port: 21109 I use Solaris 10. Please note "SO_KEEPALIVE", so the socket option is set correctly. Please very for your setup using pfiles. Also note, that you should check in your mod_jk.log using JkLogLevel debug or better trace, that during apache startup you find: [Mon Mar 06 22:08:43 2006] [3832:0001] [trace] init::jk_ajp13_worker.c (47): enter [Mon Mar 06 22:08:43 2006] [3832:0001] [trace] ajp_init::jk_ajp_common.c (1879): enter [Mon Mar 06 22:08:43 2006] [3832:0001] [debug] ajp_init::jk_ajp_common.c (1895): setting socket keepalive to 1 ... and during connect: [Mon Mar 06 22:11:14 2006] [3838:0001] [trace] ajp_send_request::jk_ajp_common.c (1166): enter [Mon Mar 06 22:11:14 2006] [3838:0001] [trace] ajp_connect_to_endpoint::jk_ajp_common.c (856): enter [Mon Mar 06 22:11:14 2006] [3838:0001] [trace] jk_open_socket::jk_connect.c (308): enter [Mon Mar 06 22:11:14 2006] [3838:0001] [debug] jk_open_socket::jk_connect.c (328): socket TCP_NODELAY set to On [Mon Mar 06 22:11:14 2006] [3838:0001] [debug] jk_open_socket::jk_connect.c (341): socket SO_KEEPALIVE set to On You can check the keep alive interval by: myapache% ndd -get /dev/tcp tcp_keepalive_interval 7200000 This is the default for Solaris 8, 9 and 10. The unit is milliseconds, so by default keep alive packets are not send before expiration of 7200 seconds = 2 hours. Maybe someone tuned that parameter on your old system. I manually set the interval to a very short value to check: # ndd -set /dev/tcp tcp_keepalive_interval 20000 # ndd -get /dev/tcp tcp_keepalive_interval 20000 After restarting apache and conecting to tomcat I sniffed the connection and voila: 1 22:36:7.66269 myapache -> mytomcat TCP D=21109 S=33730 Push Ack=1181356454 Seq=256221097 Len=1 Win=49640 2 22:36:7.66288 mytomcat -> myapache TCP D=33730 S=21109 Ack=256221098 Seq=1181356454 Len=0 Win=49561 3 22:36:27.67494 myapache -> mytomcat TCP D=21109 S=33730 Push Ack=1181356454 Seq=256221097 Len=1 Win=49640 4 22:36:27.67514 mytomcat -> myapache TCP D=33730 S=21109 Ack=256221098 Seq=1181356454 Len=0 Win=49561 5 22:36:47.68717 myapache -> mytomcat TCP D=21109 S=33730 Push Ack=1181356454 Seq=256221097 Len=1 Win=49640 6 22:36:47.68737 mytomcat -> myapache TCP D=33730 S=21109 Ack=256221098 Seq=1181356454 Len=0 Win=49561 If you need to discuss mod_jk configuration further please use tomcat-users. Concerning Solaris and TCP-Tuning you should use another appropriate forum. -- Configure bugmail: http://issues.apache.org/bugzilla/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug, or are watching the assignee. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]