Can anyone throw more light on this issue? We are being plagued by this problem and the only way we can resolve this is by HUP'ing the thread.
- Shekar On 12/19/06, Shekar Tippur <[EMAIL PROTECTED]> wrote:
Rainer, One another change we noticed was that response is cut off for a particular transaction on a thread. For every other thread, we see marshelling and un-marshelling messages in the mod_jk.log file. There is no issue of load [Fri Dec 08 01:16:25 2006] [18477:0000] [debug] ajp_connection_tcp_send_message::jk_ajp_common.c (909): 03f0 32 34 34 35 30 31 26 74 65 72 6D 49 64 3D 37 35 - 244501&termId=75 .. This is where it looks like the response was cut off. The next occurance of this process id is a new request! ... [Fri Dec 08 01:23:03 2006] [18477:0000] [debug] map_uri_to_worker::jk_uri_worker_map.c (449): Attempting to map URI '/ui/ac/loadSummary.do' from 11 maps Thanks, Shekar On 12/19/06, Shekar Tippur <[EMAIL PROTECTED]> wrote: > > Rainer, > > We upgraded mod_jk to 1.2.19 and we are still getting the same error in > mod_jk. > > [Tue Dec 19 11:35:36 2006] [31691:0000] [warn] > ajp_get_endpoint::jk_ajp_common.c (2258): Unable to get the free endpoint > for worker myWorker from 1 slots > [Tue Dec 19 11:35:36 2006] [31691:0000] [info] > ajp_get_endpoint::jk_ajp_common.c (2272): can't find free endpoint > [Tue Dec 19 11:35:36 2006] myWorker gorgon 0.000099 > [Tue Dec 19 11:35:36 2006] [31691:0000] [info] jk_handler::mod_jk.c > (1986): Service error=0 for worker=myWorker > > I would really apperciate if anyone could help me on this. > > Shekar > > On 12/17/06, Shekar Tippur <[EMAIL PROTECTED] > wrote: > > > > Hello, > > > > Thanks for replying back. > > > > I cannot see any specal requests but there maybe high load. I can see > > 193 active threads. > > > > We are in the process of upgrading mod_jk version to 1.2.19. > > > > Here is some of the information you wanted. Please let me know if you > > need more information. > > > > cat /proc/version > > > > Linux version 2.6.9-34.ELsmp ([EMAIL PROTECTED] ) > > (gcc version 3.4.5 20051201 (Red Hat 3.4.5-2 )) > > > > > > worker.properties > > > > worker.list=jkstatus,consWorker,myWorker > > > > # Configure Load Balancer status manager. > > > > worker.jkstatus.type=status > > > > worker.consWorker.port=8009 > > worker.consWorker.host=localhost > > worker.consWorker.type=ajp13 > > worker.consWorker.socket_timeout=120 > > > > # Define first worker for failover > > worker.myWorker.port=8010 > > worker.myWorker.host=localhost > > worker.myWorker.type=ajp13 > > worker.myWorker.socket_timeout=120 > > > > > > > > apache_mod_jk.conf file > > > > > > # conf/include/apache_mod_jk.conf > > # global settings files for the mod jk connector > > > > LoadModule jk_module libexec/apache_mod_jk.so > > > > # Where to find workers.properties > > JkWorkersFile /home/apache/conf/apache/mod_include/worker.properties > > > > # Where to put jk logs > > JkLogFile logs/apache/mod_jk.log > > > > # Set the jk log level [debug/error/info] > > JkLogLevel info > > > > # Select the log format > > JkLogStampFormat "[%a %b %d %H:%M:%S %Y] " > > > > # JkRequestLogFormat set the request format > > JkRequestLogFormat "%w %V %T" > > > > # JkOptions indicate to send SSL KEY SIZE, > > JkOptions +ForwardURICompat -ForwardDirectories > > > > # Configure Load Balancer status manager. > > > > #JkWorkerProperties worker.jkstatus.type=status > > > > # status for later load balancing > > > > <Location /jkmanager/> > > JkMount /jkstatus/* > > Order deny,allow > > Deny from all > > Allow from 127.0.0.1 > > </Location> > > > > JkMount /jkmanager/* jkstatus > > > > On 12/16/06, Rainer Jung < [EMAIL PROTECTED]> wrote: > > > > > > Hi, > > > > > > this looks strange. Could you please post your config and give a > > > couple > > > of details about your environment (OS+Version). Is there any pattern > > > related to the problem (special requests, high load, ...)? > > > > > > It would be really good, if you could update mod_jk to 1.2.19 or > > > 1.2.20 > > > which will most likely be available middle of next week. > > > > > > Regards, > > > > > > Rainer > > > > > > Shekar Tippur schrieb: > > > > Hello > > > > > > > > We are getting unable to get the free endpoint and eventually > > > resulting > > > > in a 500 (internal server error). > > > > > > > > We are using apache 1.3.37 and mod_jk version is 1.2.15. Due to > > > many > > > > reasons, > > > > we are not in a position to upgrade wither of these packages. > > > > > > > > [Fri Dec 08 01:34:09 2006] [18477:0000] [warn] > > > > ajp_get_endpoint::jk_ajp_common.c (2138): Unable to get the free > > > > endpoint for worker myWorker from 1 slots [Fri Dec 08 01:34:09 > > > 2006] > > > > [18477:0000] [info] ajp_get_endpoint::jk_ajp_common.c (2152): > > > can't > > > > find free endpoint [Fri Dec 08 01:34:09 2006] > > > [18477:0000] myWorker > > > > 0.000088 > > > > > > > > Under these situations, we also see that either jboss threads are > > > > processing > > > > for a long time OR a particular thread is idle for a relatively > > > long time. > > > > > > > > I would really appreciate if someone can explain why this > > > particular error > > > > occurs and how to remediate the problem. Currently we are > > > restarting both > > > > jboss and apache but we are not able to get to the root cause of > > > the > > > > problem. > > > > > > > > People who have got the same error earlier have suggested that we > > > need > > > > to increase the cachesize OR the connection_pool_size of the > > > worker. > > > > > > > > but mod_jk documentation says that it is not recommended to use > > > > cachesize more than 1 for apache version 1.3.x. > > > > > > > > I would really appreciate if anyone can suggest a remediation for > > > this > > > > issue. > > > > > > > > Shekar > > > > > > > > ----------------- > > > > cachesize > > > > > > > > This directive has been deprecated since 1.2.16. > > > > Cachesize defines the number of connections made to the AJP > > > backend > > > > that are maintained as a connection pool. It will limit the number > > > of > > > > those connection that each web server child process can make. > > > > > > > > Cachesize property is used only for multi threaded web servers > > > such as > > > > Apache 2.0 (worker), IIS and Netscape. The cachesize property > > > should > > > > reflect the number of threads per child process. JK will discover > > > the > > > > number of threads per child process on Apache 2 web server with > > > > worker-mpm and set its default value to match the ThreadsPerChild > > > > Apache directive. For IIS the default value is 10. For other web > > > > servers this value has to be set manually. > > > > > > > > Do not use cachesize with values higher then 1 on Apache 2.xprefork or > > > > Apache 1.3.x! > > > > > > > > --------------- > > > > > > > > Shekar > > > > > > > > > > > > > --------------------------------------------------------------------- > > > To start a new topic, e-mail: users@tomcat.apache.org > > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > > > > > >
