Question about JNDI connections
Hi Tomcat users (and developers), we're using an enterprise product which utilizes tomcat (v 4.1.29) as webserver technology. We've configured a LDAP connection for user authentication. Now when we want to authenticate against the LDAP it sometimes fails and results in a second user credentials request form. This error is caused by our loadbalancer which has an idle timeout after which the connection is terminated automatically. Having opened a support call regarding this behavior of the product resulted in the conclusion that tomcat establishes a permanent connection when started and then sits and waits for authentication requests always using this permannet connection. I don't know tomcat very much so I'd like to ask you if this is right. If tomcat starts this connection would it be possible to configure it to always use a new connection (or to send something like keepalive packets every x minutes)? The product vendor does not use the official JNDI realm as described in tomcat documentation but implemented an own one so it could be their fault, not tomcats ;-) To work against missunderstandings - I just want to know if either our vendor has to take action in this (and doesn't want to) or I have to think about a workaround for myself (e.g. sending an LDAP request every x minutes to avoid such connection terminattions). If this behavior is normal to tomcat servers and someone has a clue how to avoid it I would be appreciated to here from you. Many many thanks in advance, Marcel
Re: Tomcat 5.5 Context level Logger element
Thanks for your response Rémy, it is much appreciated. I have followed that guide step by step and another guide in Pro Jakarta Tomcat 5 Which contains details of tomcat 5.5 as well. So far, all I get in the tomcat.log with the configuration specified in that link is masses of debug with no details of the error that is causing the jsp to stop displaying half way through. In 5.0.28 it had all the details of the page and output of the jsp in the logs. Any other help would be greatly appreciated. Tom On Sat, 2007-02-24 at 01:34 +0100, Rémy Maucherat wrote: On 2/23/07, Tom Crinson [EMAIL PROTECTED] wrote: The only way I have found to get round this problem is to define my own Struts Request processor and override processException, and log the uncaught exceptions in there to a logger specified in the webapp log4j.properties. So, my question is, is this the only way to do this, or have I missed something more obvious? I can't seem to find anything more useful in the documentaion, it simple states these are now logged to stdout, but even though this is redirected by catalina.sh into the catalina.out the exception is not in there. However, writing to System.out and System.err both appear in here. The documentation about logging is there: http://tomcat.apache.org/tomcat-5.5-doc/logging.html Rémy - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Parallel Tomcat 5 and 6 services on Windows
Greetings, I would like to test Tomcat 6 on Windows, however I would still like to retain the previous Tomcat 5 installation (along with the Tomcat 5 service). I have downloaded the Tomcat 6.0.10 installer from the prerelease location. However, the installation threw an error when it was about to install the Tomcat 6 service (note that I have unchecked the Service option in the component selection dialogue). Ignoring the error allowed me to finish the installation. Is there some kind of problem with the installation of the Tomcat 6 service in the setup wizard? I have been able to resolve this problem by downloading and extracting the ZIP version of Tomcat6, and installing the Tomcat 6 service by running services.bat install tomcat6 (the CATALINA_HOME must have been temporarily set to the tomcat6 installation directory). This process allowed me to have paralelly installed two services for both Tomcats. Robert Novotny -- View this message in context: http://www.nabble.com/Parallel-Tomcat-5-and-6-services-on-Windows-tf3292307.html#a9157319 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]
tomcat: jmxremote and catalina.sh
Folks; in order to have remote management and monitoring enabled, I do start our tomcat 5.5.20 installations using CATALINA_OPTS like this: -Dcom.sun.management.jmxremote -Dcom.sun.management.jmxremote.port=8086 So far this works good, only problem left to solve: This way, I can't stop my Unix-based tomcat services using catalina.sh anymore, getting a connection refused for port 8086 (or whatever value I set jmxremote.port to). So. Does Using JMX remote access render catalina.sh useless? How can I get out of this and have both things running? Thanks in advance and bye, Kristian -- Kristian Rink * http://zimmer428.net * http://flickr.com/photos/z428/ jab: [EMAIL PROTECTED] * icq: 48874445 * fon: ++49 176 2447 2771 One dreaming alone, it will be only a dream; many dreaming together is the beginning of a new reality. (Hundertwasser) - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: reload on demand?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Kent, Kent Tong wrote: John McPeek spambomb at bellsouth.net writes: If you use eclipse and WTP it does all that for you. Then you can spend almost 0 time thinking about start/stop/reload stuff. Thanks for the reply. However, it will reload the webapp whenever a change is made, just like setting reloadable to true and will slow down the computer. Why not only copy your new WAR file (or classes, etc.) to the deployment directory when you actually want the webapp to reload? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4uVS9CaO5/Lv0PARAuoRAKCoy/a3TI6UrlFttMuH/iFsZ+N+xQCgwvc2 B3C77gSFdFfg1/x2bS9Xl68= =Fa0l -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Increasing JVM parameter in Linux
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Muruga, Muruganantham wrote: Can any one let me know how to increase the [MaxPermSize] JVM parameter in tomcat? $ export JAVA_OPTS=-XX:MaxPermSize=256m $ startup.sh [Is it] safe or it will affect the already stored web app files. Changing the MaxPermSize setting will not do anything to your webapp files at all. All it will do is change the amount of permanent generation space available to the JVM while it is running. - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4uY69CaO5/Lv0PARAlxqAKC6ANhlnIhjEqI1sj6YpLRj1Sy1VgCgvucW bu7mizCY/7K3x9SC2r1Yucs= =eSgq -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Increasing JVM parameter in Linux
Thanks a lot Christopher. On 2/26/07, Christopher Schultz [EMAIL PROTECTED] wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Muruga, Muruganantham wrote: Can any one let me know how to increase the [MaxPermSize] JVM parameter in tomcat? $ export JAVA_OPTS=-XX:MaxPermSize=256m $ startup.sh [Is it] safe or it will affect the already stored web app files. Changing the MaxPermSize setting will not do anything to your webapp files at all. All it will do is change the amount of permanent generation space available to the JVM while it is running. - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4uY69CaO5/Lv0PARAlxqAKC6ANhlnIhjEqI1sj6YpLRj1Sy1VgCgvucW bu7mizCY/7K3x9SC2r1Yucs= =eSgq -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Regards, M.Muruganantham. CheapChess Development Team, Silicon Oyster Technologies, Chennai-84. www.sot.in
Re: reload on demand?
Christopher Schultz chris at christopherschultz.net writes: Why not only copy your new WAR file (or classes, etc.) to the deployment directory when you actually want the webapp to reload? Creating a WAR file and copying it takes quite a lot of time. So, the webapp is setup as an exploded directory. What's missing is reload it on demand. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
ANN Tomcat Service Manager for Windows
This is a complete rewrite using .Net 2.0. Here's a overview of the major changes. More details and downloads available at http://web.bvu.edu/staff/david/index.jsp?section=softwaresubsection=tcservcfgpage=overview -now uses the Procrun-based service binaries included with Tomcat -Works with Tomcat 4.x, 5.x, and 6.x -uses XML files for configuration storage rather than the registry. -uses Procrun binaries for service configuration rather then directly access in the registry -much easier troubleshooting of service installation problems This is a beta, but should work for most typical situations. Please feel free to email me with comments, suggestions, or to report bugs. david at bvu dot edu
Re: reload on demand?
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Kent, Kent Tong wrote: Christopher Schultz chris at christopherschultz.net writes: Why not only copy your new WAR file (or classes, etc.) to the deployment directory when you actually want the webapp to reload? Creating a WAR file and copying it takes quite a lot of time. So, the webapp is setup as an exploded directory. What's missing is reload it on demand. I'm suggesting that you only deploy when you need to (even just copying files into your exploded WAR area). Are you updating your app many times more often than you want to reload it? Why? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4va/9CaO5/Lv0PARApVBAKCBVYXvB7VYXouInqM6/PpW+A407QCghgcB E0KskEqZpq4BEfFxMtsR31A= =7hpm -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Quality check mod_jk 1.2.21-dev
Hi all, the next version of mod_jk is approaching its release. A code snapshot is available at http://people.apache.org/~rjung/mod_jk-dev/ It is in the same format as a release download, so easy to build. Under the same URL you can find the updated documentation. It would be nice, if we could get some testing feedback from the community. Feel free to post any observations. Depending on your feedback we plan to tag the release at the end of the week. The most important changes are: Configuration parsing: - Check worker attributes for unknown names during startup - Add warning if duplicate map keys are read and are not allowed, e.g. when parsing uriworkermap.properties. - Don't concat worker names, if uriworkermap.properties has a duplicate pattern, instead overwrite the worker. - Fix uriworkermap.properties: Fix off-by-one problem when deleting URL mapping during reloading of uriworkermap.properties. Features - Optional session id stripping for URLs, that are *not* being forwarded (Apache, IIS) Apache - Changed semantics of empty defaults for JkEnvVar variables. - BZ 41610: Fix incorrect detection of missing Content-Length header leading to duplicate headers. IIS Plugin - Refactoring isapi_plugin configuration reading. Sun Web Server - Better build support for SunONE (Netscape/iPlanet) webservers. Status Worker - JkStatus: Allow lb sub workers in error state to be marked for recovery administratively from the status worker. - Add configuration to make status worker user name checks case insensitive. - JkStatus: Show estimated time until next global maintenance. - Change displayed time until next recovery to a min/max pair. - JkStatus: Allow a user of a read/write status worker to switch it to and from read_only mode temporarily. - JkStatus: Do not show read/write commands in a read_only status worker. - JkStatus: Improved logging by adding status worker name to messages. Added messages to the recover worker action. - Load Balancer: Do not try to recover multiple times in parallel. Use additional runtime states PROBE and FORCED. - JkStatus: Improve data synchronization between different processes. - JkStatus: Refactoring searching for workers and sub workers. Stability - 41381: Fix segfault in feature fail_on_status. Miscellaneous - Use correct windows line endings for log file on WIN32 platform. - Add some more errno logging and unify the format. The full change log is available under http://people.apache.org/~rjung/mod_jk-dev/docs/miscellaneous/changelog.html Thanks for your valuable time! Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
... and now you can find windows binaries there too (thanks Mladen) ... http://people.apache.org/~rjung/mod_jk-dev/ - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
So I tried out this new version from your site. The first problem happened in the previous version, but still is troubling. #2 is a new error I have never seen. Any ideas on that one? I am going to download SunStudio compiler like you recommended. One of my problems is that I am a newbie to compiling on UNIX. Hopefully my comments are not too basic. 1) jk_types.h is still not available in the common directory. I have to run configure to get it, but configure is supposed to be for apache loads I thought and I normally do not have apache installed on my servers. Is there a .h file you can include in the netscape directory that one can then copy to the common directory so I do not have to run configure? Is there any other file like this that needs to be unique to compiling this on solaris for sunone? 2) gcc -DNET_SSL -DSOLARIS -D_REENTRANT -DXP_UNIX -DMCC_HTTPD -DSPAPI20 -fPIC -I../common -I/wamu/hr/sunone/include -I/wamu/hr/sunone/include/base -I/wamu/hr/sunone/include/frame -I/wamu/hr/java/include -I/wamu/hr/java/include/solaris -c ../common/jk_connect.c ../common/jk_connect.c: In function `jk_is_socket_connected': ../common/jk_connect.c:663: error: `FIONREAD' undeclared (first use in this function) ../common/jk_connect.c:663: error: (Each undeclared identifier is reported only once ../common/jk_connect.c:663: error: for each function it appears in.) make: *** [jk_connect.o] Error 1 Thanks, Kirk On 2/26/07, Rainer Jung [EMAIL PROTECTED] wrote: Hi all, the next version of mod_jk is approaching its release. A code snapshot is available at http://people.apache.org/~rjung/mod_jk-dev/ It is in the same format as a release download, so easy to build. Under the same URL you can find the updated documentation. It would be nice, if we could get some testing feedback from the community. Feel free to post any observations. Depending on your feedback we plan to tag the release at the end of the week. The most important changes are: Configuration parsing: - Check worker attributes for unknown names during startup - Add warning if duplicate map keys are read and are not allowed, e.g. when parsing uriworkermap.properties. - Don't concat worker names, if uriworkermap.properties has a duplicate pattern, instead overwrite the worker. - Fix uriworkermap.properties: Fix off-by-one problem when deleting URL mapping during reloading of uriworkermap.properties. Features - Optional session id stripping for URLs, that are *not* being forwarded (Apache, IIS) Apache - Changed semantics of empty defaults for JkEnvVar variables. - BZ 41610: Fix incorrect detection of missing Content-Length header leading to duplicate headers. IIS Plugin - Refactoring isapi_plugin configuration reading. Sun Web Server - Better build support for SunONE (Netscape/iPlanet) webservers. Status Worker - JkStatus: Allow lb sub workers in error state to be marked for recovery administratively from the status worker. - Add configuration to make status worker user name checks case insensitive. - JkStatus: Show estimated time until next global maintenance. - Change displayed time until next recovery to a min/max pair. - JkStatus: Allow a user of a read/write status worker to switch it to and from read_only mode temporarily. - JkStatus: Do not show read/write commands in a read_only status worker. - JkStatus: Improved logging by adding status worker name to messages. Added messages to the recover worker action. - Load Balancer: Do not try to recover multiple times in parallel. Use additional runtime states PROBE and FORCED. - JkStatus: Improve data synchronization between different processes. - JkStatus: Refactoring searching for workers and sub workers. Stability - 41381: Fix segfault in feature fail_on_status. Miscellaneous - Use correct windows line endings for log file on WIN32 platform. - Add some more errno logging and unify the format. The full change log is available under http://people.apache.org/~rjung/mod_jk-dev/docs/miscellaneous/changelog.html Thanks for your valuable time! Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Test, ignore
Well, that is great! And stupid Gmail still eats up my messages that include sample config and log output that illustrate the problem I am trying to solve with clustering. Ed On 2/24/07, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: yeah we all now certainly have seen your e-mail so u can post to the whole group all ya want!!! :) - Original Message From: Gmail User [EMAIL PROTECTED] To: Tomcat Users List users@tomcat.apache.org Sent: Saturday, February 24, 2007 10:05:15 AM Subject: Test, ignore Not seeing my posts. Just testing... - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Rainer, Rainer Jung wrote: Features - Optional session id stripping for URLs, that are *not* being forwarded (Apache, IIS) Woo hoo! No more crazy mod_rewrite rules! Thanks everyone! - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4xsu9CaO5/Lv0PARAtj/AKCcIQgB/oFZe+lTDma5EgBEsLbGugCfcFtE pWkktd1mRPy92wKZzex+Ccs= =EIaU -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Parallel Tomcat 5 and 6 services on Windows
Robert Novotny wrote: I would like to test Tomcat 6 on Windows, however I would still like to retain the previous Tomcat 5 installation (along with the Tomcat 5 service). I would think you can do that. You can even run them both at the same time, as long as they are configured to listen on different ports. However, the installation threw an error when it was about to install the Tomcat 6 service (note that I have unchecked the Service option in the component selection dialogue). What was the error? Was it a port conflict? If so, you know how to solve that one, right? If it was a service name conflict, you should be able to solve that one too by renaming one of the services. Did you at least stop the tomcat5 service before installing tomcat6? As long as only one of them is running at once, it should work fine. See Control Panel | Administrative Tools | Services Mojo -- Morris Jones Monrovia, CA http://www.whiteoaks.com Old Town Astronomers http://www.otastro.org - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
[mod_jk] JkStripSession directive
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 All, I decided to test out the newly available mod_jk 1.2.21 pre-release and the JkStripSession directive in order to remove mod_rewrite from my configuration. It does not appear to be working using my configuration. I'm using Apache httpd 2.0.58 with the pre-release mod_jk 1.2.21. In my Apache httpd configuration, I have: LoadModule ... JkWorkersFile ... The above is outside of my VirtualHost. Also: JkStripSession On JkMount ... [many of these] These are inside my VirtualHost. In my mod_jk.log, I can see: [Mon Feb 26 12:59:06 2007] [6999:34480] [debug] jk_map_to_storage::mod_jk.c (3025): removing session identifier [;jsessionid=A17B19517B58B2C06D8B7120D3B8503C] for non servlet url [/diagnosis/images/flags/en.png;jsessionid=A17B19517B58B2C06D8B7120D3B8503C] [Mon Feb 26 12:59:06 2007] [6999:34480] [debug] map_uri_to_worker::jk_uri_worker_map.c (589): Attempting to map URI '/diagnosis/errors/not_found.html' from 28 maps But I can still see the attempt to access the file in Apache's access log: 192.168.1.47 - - [26/Feb/2007:12:59:06 -0500] GET /diagnosis/images/flags/es.png;jsessionid=A17B19517B58B2C06D8B7120D3B8503C HTTP/1.1 404 1766 Now, Apache httpd might be logging the request before mod_jk sees if, in which case the jsessionid in the URL is no indication of mod_jk's behavior. If I attempt to use the URI without the session id (that mod_jk has appeared to correctly identify: ;jsessionid=A17B19517B58B2C06D8B7120D3B8503C), Apache serves the request as expected. Any ideas? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4yH19CaO5/Lv0PARAouDAJ4736bY2UA6cMTPAccQTb+FdXzVmgCeJMLA 0+R1IoDfqQ6WotGZrvuiwS0= =35xc -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
IIS Plugin won't load balance?
I've got jk load balancing working from Apache, but need it to work from IIS. I'm using the My workers.properties and uriworkermap.properties files are the same for Apache and IIS. #workers.properties worker.list=loadbalancer worker.tomcat1.type=ajp13 worker.tomcat1.host=192.168.1.7 worker.tomcat1.port=8009 worker.tomcat2.type=ajp13 worker.tomcat2.host=192.168.1.7 worker.tomcat2.port=9009 worker.loadbalancer.type=lb worker.loadbalancer.balance_workers=tomcat1,tomcat2 #uriworkermap.properties /servlets/*=loadbalancer When I access servlets/Build from Apache everything works fine. When I try to access servlets/Build from IIS, I get the following output in my browser window: The specified module could not be found. And the last thing in the isapi.log file is: [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_uri_worker_map.c (602): Found a wildchar match '/servlets/*=loadbalancer' [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_isapi_plugin.c (885): check if [/servlets/Build] is points to the web-inf directory [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_isapi_plugin.c (903): [/servlets/Build] is a servlet url - should redirect to loadbalancer Anybody know what The specified module could not be found. means? The note about the web-inf directory seems a bit wierd, and I don't see it in the Apache mod_jk log. -- Scott. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: IIS Plugin can't find Module ??
Maybe I should walk before running. The plugin fails the same way with a single worker (i.e., without load balancing). My current theory is that maybe the problem is that I'm using the root context. It's all I can think of as a cause for The specified module could not be found. I just love error messages that refer to specified items :-\ Anybody else have any ideas? -- Scott. Scott Danforth wrote: I've got jk load balancing working from Apache, but need it to work from IIS. I'm using the My workers.properties and uriworkermap.properties files are the same for Apache and IIS. #workers.properties worker.list=loadbalancer worker.tomcat1.type=ajp13 worker.tomcat1.host=192.168.1.7 worker.tomcat1.port=8009 worker.tomcat2.type=ajp13 worker.tomcat2.host=192.168.1.7 worker.tomcat2.port=9009 worker.loadbalancer.type=lb worker.loadbalancer.balance_workers=tomcat1,tomcat2 #uriworkermap.properties /servlets/*=loadbalancer When I access servlets/Build from Apache everything works fine. When I try to access servlets/Build from IIS, I get the following output in my browser window: The specified module could not be found. And the last thing in the isapi.log file is: [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_uri_worker_map.c (602): Found a wildchar match '/servlets/*=loadbalancer' [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_isapi_plugin.c (885): check if [/servlets/Build] is points to the web-inf directory [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_isapi_plugin.c (903): [/servlets/Build] is a servlet url - should redirect to loadbalancer Anybody know what The specified module could not be found. means? The note about the web-inf directory seems a bit wierd, and I don't see it in the Apache mod_jk log. -- Scott. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] -- Scott Danforth, Ph.D. Director of Technology ePath Learning, Inc. 70 Howard Street, Suite C New London, CT 06320 Phone: (860) 444-6989 ext. 3 Fax: (215) 243-7454 email: [EMAIL PROTECTED] - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
increase the backend idle connection timeout or the connection_pool_minsize
Hi, This is a followup to my Feb14 post: Re: socket is not connected any more (errno=11) I've just upgraded from 1.2.19 to 1.2.20 and the new log messages are: increase the backend idle connection timeout or the connection_pool_minsize I could not find much on which parameters this is referring to? On the mod_jk side I have: worker.tomcat-publicweb-01.type=ajp13 worker.tomcat-publicweb-01.host=app-01 worker.tomcat-publicweb-01.port=12345 worker.tomcat-publicweb-01.lbfactor=50 worker.tomcat-publicweb-01.cachesize=10 worker.tomcat-publicweb-01.cache_timeout=600 worker.tomcat-publicweb-01.socket_keepalive=1 worker.tomcat-publicweb-01.retries=2 worker.tomcat-publicweb-01.socket_timeout=30 worker.tomcat-publicweb-01.connect_timeout=2000 worker.tomcat-publicweb-01.prepost_timeout=5000 worker.tomcat-publicweb-01.reply_timeout=30 On the tomcat side I have: parameter namemaxActive/name value20/value /parameter parameter namemaxIdle/name value2/value /parameter parameter namemaxWait/name value1000/value /parameter parameter namelogAbandoned/name valuetrue/value /parameter parameter nameremoveAbandoned/name valuetrue/value /parameter parameter nameremoveAbandonedTimeout/name value1200/value /parameter Which parameters are these log messages referring to? Many thanks, Fletcher. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [mod_jk] JkStripSession directive
Christopher Schultz wrote: JkStripSession On JkMount ... [many of these] So far so good :) 192.168.1.47 - - [26/Feb/2007:12:59:06 -0500] GET /diagnosis/images/flags/es.png;jsessionid=A17B19517B58B2C06D8B7120D3B8503C HTTP/1.1 404 1766 Right, the ;jsessionid=... will be shown in the logs. However, you are right, I made a bug in that code, it should trip the session from r-filename not r-uri. Will be fixed in the next tryout. Regards, Mladen. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: increase the backend idle connection timeout or the connection_pool_minsize
on that connection pool size, these two: parameter namemaxActive/name value20/value /parameter parameter namemaxIdle/name value2/value /parameter You're also only allowing 1 sec to connect it appears. Might ratchet it up a little more. -Original Message- From: news [mailto:[EMAIL PROTECTED] Behalf Of Fletcher Cocquyt Sent: Monday, February 26, 2007 12:32 PM To: users@tomcat.apache.org Subject: increase the backend idle connection timeout or the connection_pool_minsize Hi, This is a followup to my Feb14 post: Re: socket is not connected any more (errno=11) I've just upgraded from 1.2.19 to 1.2.20 and the new log messages are: increase the backend idle connection timeout or the connection_pool_minsize I could not find much on which parameters this is referring to? On the mod_jk side I have: worker.tomcat-publicweb-01.type=ajp13 worker.tomcat-publicweb-01.host=app-01 worker.tomcat-publicweb-01.port=12345 worker.tomcat-publicweb-01.lbfactor=50 worker.tomcat-publicweb-01.cachesize=10 worker.tomcat-publicweb-01.cache_timeout=600 worker.tomcat-publicweb-01.socket_keepalive=1 worker.tomcat-publicweb-01.retries=2 worker.tomcat-publicweb-01.socket_timeout=30 worker.tomcat-publicweb-01.connect_timeout=2000 worker.tomcat-publicweb-01.prepost_timeout=5000 worker.tomcat-publicweb-01.reply_timeout=30 On the tomcat side I have: parameter namemaxActive/name value20/value /parameter parameter namemaxIdle/name value2/value /parameter parameter namemaxWait/name value1000/value /parameter parameter namelogAbandoned/name valuetrue/value /parameter parameter nameremoveAbandoned/name valuetrue/value /parameter parameter nameremoveAbandonedTimeout/name value1200/value /parameter Which parameters are these log messages referring to? Many thanks, Fletcher. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
On Feb 26, 2007, at 11:11 AM, Rainer Jung wrote: Hi all, the next version of mod_jk is approaching its release. A code snapshot is available at http://people.apache.org/~rjung/mod_jk-dev/ It is in the same format as a release download, so easy to build. Under the same URL you can find the updated documentation. It would be nice, if we could get some testing feedback from the community. Feel free to post any observations. Depending on your feedback we plan to tag the release at the end of the week. So far so good... I haven't seen any problems with the snap. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [mod_jk] JkStripSession directive
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Mladen, Mladen Turk wrote: Right, the ;jsessionid=... will be shown in the logs. However, you are right, I made a bug in that code, it should trip the session from r-filename not r-uri. Will be fixed in the next tryout. Sweet. When should I check back for an update? - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF4zoA9CaO5/Lv0PARAoo0AKCxf4a19ougghcNvAq0FU6fDGNakACfVMQu j6AmFloB2lZD0wk08hy+RRc= =AWdv -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: IIS Plugin can't find Module ??
Apparently, The specified module could not be found is a Windows Message. Although the ISAPI_redirect dll is being loaded, it must be using COM/ActiveX/.NET to ask for something else that isn't available on my system once it decides to forward a page request to Tomcat.. What module would that be? Beats me. I'm using the version of IIS 5.0 that comes with XP. As before, suggestions would be appreciated. -- Scott. Scott Danforth wrote: Nope. It fails the same way for HelloWordExample in the servlets-examples context. Any ideas appreciated. Anybody ever seen The specified module could not be found when using isaspi_redirect.dll? -- Scott. [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_uri_worker_map.c (575): Attempting to map URI '/localhost/servlets-examples/' from 2 maps [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets-examples/servlet/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_isapi_plugin.c (875): Default redirection of /servlets-examples/ [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_uri_worker_map.c (575): Attempting to map URI '/servlets-examples/' from 2 maps [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets-examples/servlet/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_isapi_plugin.c (885): check if [/servlets-examples/] is points to the web-inf directory [Mon Feb 26 14:10:58 2007] [2020:1756] [debug] jk_isapi_plugin.c (1004): [/servlets-examples/] is not a servlet url [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_isapi_plugin.c (801): Filter started [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_isapi_plugin.c (868): Virtual Host redirection of /192.168.1.50/servlets-examples/servlet/HelloWordExample [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_uri_worker_map.c (575): Attempting to map URI '/192.168.1.50/servlets-examples/servlet/HelloWordExample' from 2 maps [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets-examples/servlet/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_isapi_plugin.c (875): Default redirection of /servlets-examples/servlet/HelloWordExample [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_uri_worker_map.c (575): Attempting to map URI '/servlets-examples/servlet/HelloWordExample' from 2 maps [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_uri_worker_map.c (587): Attempting to map context URI '/servlets-examples/servlet/*=tomcat1' source 'uriworkermap' [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_uri_worker_map.c (602): Found a wildchar match '/servlets-examples/servlet/*=tomcat1' [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_isapi_plugin.c (885): check if [/servlets-examples/servlet/HelloWordExample] is points to the web-inf directory [Mon Feb 26 14:11:28 2007] [2020:1756] [debug] jk_isapi_plugin.c (903): [/servlets-examples/servlet/HelloWordExample] is a servlet url - should redirect to tomcat1 Scott Danforth wrote: Maybe I should walk before running. The plugin fails the same way with a single worker (i.e., without load balancing). My current theory is that maybe the problem is that I'm using the root context. It's all I can think of as a cause for The specified module could not be found. I just love error messages that refer to specified items :-\ Anybody else have any ideas? -- Scott. Scott Danforth wrote: I've got jk load balancing working from Apache, but need it to work from IIS. I'm using the My workers.properties and uriworkermap.properties files are the same for Apache and IIS. #workers.properties worker.list=loadbalancer worker.tomcat1.type=ajp13 worker.tomcat1.host=192.168.1.7 worker.tomcat1.port=8009 worker.tomcat2.type=ajp13 worker.tomcat2.host=192.168.1.7 worker.tomcat2.port=9009 worker.loadbalancer.type=lb worker.loadbalancer.balance_workers=tomcat1,tomcat2 #uriworkermap.properties /servlets/*=loadbalancer When I access servlets/Build from Apache everything works fine. When I try to access servlets/Build from IIS, I get the following output in my browser window: The specified module could not be found. And the last thing in the isapi.log file is: [Mon Feb 26 12:14:29 2007] [2036:3200] [debug] jk_uri_worker_map.c (602): Found a wildchar match '/servlets/*=loadbalancer' [Mon Feb 26 12:14:29 2007] [2036:3200]
Re: increase the backend idle connection timeout or the connection_pool_minsize
Fletcher Cocquyt wrote: Hi, This is a followup to my Feb14 post: Re: socket is not connected any more (errno=11) I've just upgraded from 1.2.19 to 1.2.20 and the new log messages are: increase the backend idle connection timeout or the connection_pool_minsize I could not find much on which parameters this is referring to? On the mod_jk side I have: worker.tomcat-publicweb-01.type=ajp13 worker.tomcat-publicweb-01.host=app-01 worker.tomcat-publicweb-01.port=12345 worker.tomcat-publicweb-01.lbfactor=50 worker.tomcat-publicweb-01.cachesize=10 worker.tomcat-publicweb-01.cache_timeout=600 worker.tomcat-publicweb-01.socket_keepalive=1 worker.tomcat-publicweb-01.retries=2 worker.tomcat-publicweb-01.socket_timeout=30 worker.tomcat-publicweb-01.connect_timeout=2000 worker.tomcat-publicweb-01.prepost_timeout=5000 worker.tomcat-publicweb-01.reply_timeout=30 You can add connection_pool_minsize. See the workers.properties pages in the reference guide. On the tomcat side I have: parameter namemaxActive/name value20/value /parameter parameter namemaxIdle/name value2/value /parameter parameter namemaxWait/name value1000/value /parameter parameter namelogAbandoned/name valuetrue/value /parameter parameter nameremoveAbandoned/name valuetrue/value /parameter parameter nameremoveAbandonedTimeout/name value1200/value /parameter This looks like a DataSource definition (Tomcat to database). The idle connection timeout in the case of tomcat is the value of conncetiontimeout in the Connector element of server.xml. You need to find the AJP/1.3 connector. Which parameters are these log messages referring to? Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
Hi Kirk, Number 2 should be a consequence of number 1. configure detects a header file related to FIONREAD. Concerning 1), we would need to rework configure to allow to use it in the Sun Web Server case. Until now it forces you to provide a --with-apxs. For your testing I uploaded two header files on the test URL (people.apache.org...) containing the configure results I produced on another Solaris system. I hope they will work for you. I'm not sure, if we can fix configure for 1.2.21, because we want to provide all the other fixes in there soon. But I'm confident, that configure will handle your case starting with 1.2.22 in a couple of weeks. Thanks for testing and your feedback. Let us know, if the header files work for you. Regards, Rainer Kirk wrote: So I tried out this new version from your site. The first problem happened in the previous version, but still is troubling. #2 is a new error I have never seen. Any ideas on that one? I am going to download SunStudio compiler like you recommended. One of my problems is that I am a newbie to compiling on UNIX. Hopefully my comments are not too basic. 1) jk_types.h is still not available in the common directory. I have to run configure to get it, but configure is supposed to be for apache loads I thought and I normally do not have apache installed on my servers. Is there a .h file you can include in the netscape directory that one can then copy to the common directory so I do not have to run configure? Is there any other file like this that needs to be unique to compiling this on solaris for sunone? 2) gcc -DNET_SSL -DSOLARIS -D_REENTRANT -DXP_UNIX -DMCC_HTTPD -DSPAPI20 -fPIC -I../common -I/wamu/hr/sunone/include -I/wamu/hr/sunone/include/base -I/wamu/hr/sunone/include/frame -I/wamu/hr/java/include -I/wamu/hr/java/include/solaris -c ../common/jk_connect.c ../common/jk_connect.c: In function `jk_is_socket_connected': ../common/jk_connect.c:663: error: `FIONREAD' undeclared (first use in this function) ../common/jk_connect.c:663: error: (Each undeclared identifier is reported only once ../common/jk_connect.c:663: error: for each function it appears in.) make: *** [jk_connect.o] Error 1 Thanks, Kirk - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
Hi Kirk, I never built for Sun Web Server, but I just saw the configure flag: --enable-netscape Did you ever try running configure with that one before doing the make? Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [mod_jk] JkStripSession directive
Mladen Turk wrote: Right, the ;jsessionid=... will be shown in the logs. However, you are right, I made a bug in that code, it should trip the session from r-filename not r-uri. Will be fixed in the next tryout. ... which is now available for testing under the same URL. Since this change was Apache only, docs and Windows binaries didn't change. The old tarball has been deleted and only the new one is on people.apache.org. Thanks to Christopher for testing and reporting, and of course to Mladen for fixing. Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
Due to a bug one user found (thanks) in the new JkStripSession feature (only in the Apache httpd implementation of it), there is now a new tarball available at the old address. http://people.apache.org/~rjung/mod_jk-dev/ Happy testing! Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Can tomcat send out event about persistent connection being closed?
I am using Tomcat 5.5 and I would like to bind the HttpSession lifecycle with its underlying persistent connection lifecycle. Partically, if the persistent connection is closed due to connectionTimeout or client program exits, I would like to invalid the client's HttpSession, and if the HttpSession timeout, I would like to close the persistent connection too. I know the APIs to control the HttpSession lifecycle, but I am wondering if there is any API in Tomcat to allow me to control the lifecycle of the underlying persistent connection. Your help is greatly appreciated. Monica Expecting? Get great news right away with email Auto-Check. Try the Yahoo! Mail Beta. http://advision.webevents.yahoo.com/mailbeta/newmail_tools.html - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: Can tomcat send out event about persistent connection being closed?
From: M M [mailto:[EMAIL PROTECTED] Subject: Can tomcat send out event about persistent connection being closed? I am using Tomcat 5.5 and I would like to bind the HttpSession lifecycle with its underlying persistent connection lifecycle. A couple of issues: 1) There may not be an underlying persistent connection, if the client chooses not to use keep-alives. 2) There will likely be more than one connection associated with a given HttpSession. For example, content embedded in a page will usually result in multiple simultaneous connections being establshed by the browser to retrieve said content as quickly as possible. - Chuck THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus for use only by the intended recipient. If you received this in error, please contact the sender and delete the e-mail and its attachments from all computers. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Bootstrap stop()
I've been working on a class that will start and stop Tomcat from via a JMenu
item. It seems I have tomcat starting correctly with the following code:
class StartListener implements ActionListener {
public void actionPerformed(ActionEvent event) {
try {
bootStrap.setCatalinaHome(C:\\Varin\\Java_Dev\\Servers\\Tomcat\\apache-tomcat-5.5.20);
bootStrap.start();
} catch (Throwable e ) {
System.out.println(Unable to start Tomcat);
e.printStackTrace();
return;
}
System.out.println(Tomcat Started...);
}
} //End Class SaveAsListener
...and I'm using the following to stop the Tomcat instance:
class StopListener implements ActionListener {
public void actionPerformed(ActionEvent event) {
try {
bootStrap.stop();
bootStrap.destroy();
} catch (Throwable e ) {
System.out.println(Unable to stop Tomcat);
e.printStackTrace();
return;
}
System.out.println(Tomcat stopped...);
}
} //End Class SaveAsListener
The issue that I am having is that when I select the start a second time
Tomcat does not seem to start. Also, when I select stop and then hit the
Tomcat server I receive a 503 message as opposed to a server not found.
So, this leads me to believe that the server is in fact not actually
stopping.
The documentation is really spartan in this area so, I took a look at the
source for Bootstrap...there is a destroy method that according to the
javadoc is supposed to end the daemon but, it has a single comment in it
with no body that just says fix me. So, I'm thinking maybe there really is
no way to bring the daemon down so that it can be recycled.
Any help would be appreciated.
Fran
--
View this message in context:
http://www.nabble.com/Bootstrap-stop%28%29-tf3296281.html#a9169712
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]
Re: Quality check mod_jk 1.2.21-dev
I will take a look at it. I did not see that this option, so I never knew to use it. If I were to try and document the process I am going through to compile, would that be of help to put as a solaris sunone build doc? finding information on the web is few and far between. Most people probably run mod_jk under apache from what I have noticed and all of the docs are geared towards that. The ones I can find for sunone are geared towards windows. I am sure between the emails you and I have traded in the past few weeks on my trials and tribulations, if I can get everything to work, I can write up some docs to contribute. Kirk On 2/26/07, Rainer Jung [EMAIL PROTECTED] wrote: Hi Kirk, I never built for Sun Web Server, but I just saw the configure flag: --enable-netscape Did you ever try running configure with that one before doing the make? Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
connection pool - DBCP error
I am getting a slight error with regards to my connection pool configuration. I may not have totally closed the connection properly, but shouldn't the following account for it? AbandonedObjectPool is used ( [EMAIL PROTECTED]) LogAbandoned: true RemoveAbandoned: true RemoveAbandonedTimeout: 60 in other words, if I have the following configured like so, parameter namemaxWait/name value7000/value /parameter parameter nameremoveAbandoned/name valuetrue/value /parameter parameter nameremoveAbandonedTimeout/name value60/value /parameter parameter namelogAbandoned/name valuetrue/value /parameter shouldn't the RemoveAbandoned params account for an open or vagrant connection and kill it off properly? Am I wrong in the assumption this attribute should have? Thanks, Barry
Re: Can tomcat send out event about persistent connection being closed?
Thanks for your reply. Our client application will always use keep-alive which is the default beahvior of HTTP/1.1 and it does not do simultaneous connections to the web server. Thanks. Monica - Original Message From: Caldarale, Charles R [EMAIL PROTECTED] To: Tomcat Users List users@tomcat.apache.org Sent: Monday, February 26, 2007 4:28:58 PM Subject: RE: Can tomcat send out event about persistent connection being closed? From: M M [mailto:[EMAIL PROTECTED] Subject: Can tomcat send out event about persistent connection being closed? I am using Tomcat 5.5 and I would like to bind the HttpSession lifecycle with its underlying persistent connection lifecycle. A couple of issues: 1) There may not be an underlying persistent connection, if the client chooses not to use keep-alives. 2) There will likely be more than one connection associated with a given HttpSession. For example, content embedded in a page will usually result in multiple simultaneous connections being establshed by the browser to retrieve said content as quickly as possible. - Chuck THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY MATERIAL and is thus for use only by the intended recipient. If you received this in error, please contact the sender and delete the e-mail and its attachments from all computers. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] TV dinner still cooling? Check out Tonight's Picks on Yahoo! TV. http://tv.yahoo.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]
RE: connection pool - DBCP error
I had to add the following before I saw that behavior. You'll have to check the API, to make sure the minEvictableIdleTimeMillis and timeBetweenEvictionRunsMillis are set correctly for your pool. But once I added these, I would see abandoned connections get dropped. I put these settings in, because I found that the DBCP pool only evalutes connections on a check-in/out basis. parameternameminEvictableIdleTimeMillis/namevalue5000/value/param eter parameternametimeBetweenEvictionRunsMillis/namevalue1/value/p arameter parameternametestWhileIdle/namevaluetrue/value/parameter parameternamevalidationQuery/namevalueselect count(*) from dual/value/parameter -Original Message- From: Propes, Barry L [GCG-NAOT] [mailto:[EMAIL PROTECTED] Sent: Monday, February 26, 2007 3:48 PM To: Tomcat Users List Subject: connection pool - DBCP error I am getting a slight error with regards to my connection pool configuration. I may not have totally closed the connection properly, but shouldn't the following account for it? AbandonedObjectPool is used ( [EMAIL PROTECTED]) LogAbandoned: true RemoveAbandoned: true RemoveAbandonedTimeout: 60 in other words, if I have the following configured like so, parameter namemaxWait/name value7000/value /parameter parameter nameremoveAbandoned/name valuetrue/value /parameter parameter nameremoveAbandonedTimeout/name value60/value /parameter parameter namelogAbandoned/name valuetrue/value /parameter shouldn't the RemoveAbandoned params account for an open or vagrant connection and kill it off properly? Am I wrong in the assumption this attribute should have? Thanks, Barry - The information in this message may be proprietary and/or confidential, and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify First Data immediately by replying to this message and deleting it from your computer.
Re: increase the backend idle connection timeout or the connection_pool_minsize
Rainer Jung rainer.jung at kippdata.de writes: You can add connection_pool_minsize. See the workers.properties pages in the reference guide. I'm using apache 2.0 prefork - the docs: http://tomcat.apache.org/connectors-doc/reference/workers.html say Do not use connection_pool_size with values higher then 1 on Apache 2.x prefork I've tried tuning all the other suggested parameters without a noticable effect on the rate of these info messages (2000-3000/hour) I tried: maxActive 40, maxIdle 5, maxWait 3000 no discernable difference in error rate. Maybe I need to modify them with bigger deltas to see a difference? Is there a way to tell which parameter would have the most impact? This looks like a DataSource definition (Tomcat to database). The idle connection timeout in the case of tomcat is the value of conncetiontimeout in the Connector element of server.xml. You need to find the AJP/1.3 connector. I tried connectionTimeout = 60 (10 times the default of 60 seconds) - no reduction in error rate thanks - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
RE: mod_jk replacement?
Hi, how to put wildcards into ProxyPass, for example if I want to send only jsp requests to tomcat and not html like below. ProxyPass /*.jsp ajp://127.0.0.1:8009/ the above config is not working, I have jsp files spread across multiple folders on tomcat and the same folder structure on apache has html pages, so I want to forward only jsp requests to tomcat thanks Raj Please Note: Due to Florida's very broad public records law, most written communications to or from College employees regarding College business are public records, available to the public and media upon request. Therefore, this email communication may be subject to public disclosure [EMAIL PROTECTED] 1/22/2007 4:14:54 AM Recently i'm using mod_proxy_ajp, as mod_jk replacement with apache. I don't have this in an production environment, but along this week I plan to do it. This is an extract of my apache config: ProxyPass /app/img/ ! ProxyPass /app/js/ ! ProxyPass /app/doc/ ! ProxyPass /app/css/ ! ProxyPass /app/ ajp://127.0.0.1:8009/app/ Alias /app /srv/www/tomcat5/base/webapps/app Directory /srv/www/tomcat5/base/webapps/app Options None allow from all /Directory -Mensaje original- De: JNeuhoff [mailto:[EMAIL PROTECTED] Enviado el: lunes, 22 de enero de 2007 0:27 Para: users@tomcat.apache.org Asunto: mod_jk replacement? Is there another connector software available between Apache 2.0.59 and Tomcat 5.5.17 on Windows 2003 which is more stable and suitable for a live web service? mod_jk 1.2.20 (which I believe is the latest version) appears to have some problems with managing TCP connections (see http://www.nabble.com/Apache-mod_jk-memory-leak--tf3023318.html for details). We are desparately looking for a working connection software between Apache and Tomcat. Alternatively, we are thinking of running a standalone Tomcat in which case we'd find to find a way of mapping some web-enabled directories with static resources (mainly PNG images and some simple PHP scripts) to Tomcat in addition to our Servlet. Is that possible? -- View this message in context: http://www.nabble.com/mod_jk-replacement--tf3050993.html#a8481132 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] -- La Información incluida en el presente correo electrónico es SECRETO PROFESIONAL Y CONFIDENCIAL, siendo para el uso exclusivo del destinatario arriba mencionado. Si usted lee este mensaje y no es el destinatario señalado, el empleado o el agente responsable de entregar el mensaje al destinatario, o ha recibido esta comunicación por error, le informamos que esta totalmente prohibida cualquier divulgación, distribución o reproducción de esta comunicación, y le rogamos que nos lo notifique inmediatamente y nos devuelva el mensaje original a la dirección arriba mencionada. Gracias. The information contained in this e-mail is LEGALLY PRIVILEDGED AND CONFIDENTIAL and is intended only for the use of the addressee named above. If the reader of this message is not the intended recipient or the employee or agent responsible for delivering the message to the intended recipient, or you have received this communication in error, please be aware that any dissemination, distribution or duplication of this communication is strictly prohibited, and please notify us immediately and return the original message to us at the address above. Thank you. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Can tomcat send out event about persistent connection being closed?
On 2/26/07, M M [EMAIL PROTECTED] wrote: Thanks for your reply. Our client application will always use keep-alive which is the default beahvior of HTTP/1.1 and it does not do simultaneous connections to the web server. Thanks. Monica Why use HTTP for a connection that is persistent? Wouldn't you be better off using RMI or some similar technology? HTTP was designed to be connectionless. The keepalive is an optimization trick, and is specifically layered beneath the ability of the programmer to utilize or control. -- Gary Evesson
RE: connection pool - DBCP error
thanks, Daniel!! Ill try to plug that in and see if my version will accept those params! -Original Message- From: Stephens, Daniel [mailto:[EMAIL PROTECTED] Sent: Monday, February 26, 2007 4:55 PM To: Tomcat Users List Subject: RE: connection pool - DBCP error I had to add the following before I saw that behavior. You'll have to check the API, to make sure the minEvictableIdleTimeMillis and timeBetweenEvictionRunsMillis are set correctly for your pool. But once I added these, I would see abandoned connections get dropped. I put these settings in, because I found that the DBCP pool only evalutes connections on a check-in/out basis. parameternameminEvictableIdleTimeMillis/namevalue5000/value/param eter parameternametimeBetweenEvictionRunsMillis/namevalue1/value/p arameter parameternametestWhileIdle/namevaluetrue/value/parameter parameternamevalidationQuery/namevalueselect count(*) from dual/value/parameter -Original Message- From: Propes, Barry L [GCG-NAOT] [mailto:[EMAIL PROTECTED] Sent: Monday, February 26, 2007 3:48 PM To: Tomcat Users List Subject: connection pool - DBCP error I am getting a slight error with regards to my connection pool configuration. I may not have totally closed the connection properly, but shouldn't the following account for it? AbandonedObjectPool is used ( [EMAIL PROTECTED]) LogAbandoned: true RemoveAbandoned: true RemoveAbandonedTimeout: 60 in other words, if I have the following configured like so, parameter namemaxWait/name value7000/value /parameter parameter nameremoveAbandoned/name valuetrue/value /parameter parameter nameremoveAbandonedTimeout/name value60/value /parameter parameter namelogAbandoned/name valuetrue/value /parameter shouldn't the RemoveAbandoned params account for an open or vagrant connection and kill it off properly? Am I wrong in the assumption this attribute should have? Thanks, Barry - The information in this message may be proprietary and/or confidential, and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please notify First Data immediately by replying to this message and deleting it from your computer.
Tomcat Service Manager for windows
Hello, I just joined this mailing list to plug a piece of software that has been enormously helpful to the company I work for. I have no idea if its been mentioned on this mailing list or not, so excuse my redundancy. Its called Tomcat Service Manager. it enables one to very easily make multiple instances of Tomcat 4, 5, 5.5, and now 6 on Windows Operating Systems. The author is David Boyer. If you require multiple instances of Tomcat on Windows servers, I highly recommend it. http://web.bvu.edu/staff/david/index.jsp?section=softwaresubsection=tcservc fgpage=overview - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
HTTP plus
Was this last message of mine just too naive :) Should I break it into smaller parts? Still curious :) Peter K. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Quality check mod_jk 1.2.21-dev
Kirk schrieb: I will take a look at it. I did not see that this option, so I never knew to use it. If I were to try and document the process I am going through to compile, would that be of help to put as a solaris sunone build doc? finding information on the web is few and far between. Most people probably run mod_jk under apache from what I have noticed and all of the docs are geared towards that. The ones I can find for sunone are geared towards windows. I am sure between the emails you and I have traded in the past few weeks on my trials and tribulations, if I can get everything to work, I can write up some docs to contribute. That would be very nice and useful contribution. Before you start writing HTML, please notice, that the docs are managed in a special format. Contributing docs is explained under http://tomcat.apache.org/connectors-doc/miscellaneous/doccontrib.html Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: increase the backend idle connection timeout or the connection_pool_minsize
Fletcher Cocquyt schrieb: Rainer Jung rainer.jung at kippdata.de writes: You can add connection_pool_minsize. See the workers.properties pages in the reference guide. I'm using apache 2.0 prefork - the docs: http://tomcat.apache.org/connectors-doc/reference/workers.html say Do not use connection_pool_size with values higher then 1 on Apache 2.x prefork Yes, but the log message is talking about the min size. The default min size is size/2, but we changed it between 1.2.19 and 1.2.20 to (size+1)/2. So for prefork the min size was 0 until 1.2.19 and is 1 starting with 1.2.20. I've tried tuning all the other suggested parameters without a noticable effect on the rate of these info messages (2000-3000/hour) I tried: maxActive 40, maxIdle 5, maxWait 3000 no discernable difference in error rate. Maybe I need to modify them with bigger deltas to see a difference? Is there a way to tell which parameter would have the most impact? If those are for a DataSource, they have no relation to your mod_jk messages. This looks like a DataSource definition (Tomcat to database). The idle connection timeout in the case of tomcat is the value of conncetiontimeout in the Connector element of server.xml. You need to find the AJP/1.3 connector. I tried connectionTimeout = 60 (10 times the default of 60 seconds) - no reduction in error rate Are you sure, you put it into the correct connector? Please post your version numbers, platform details, log example and relevant apache and tomcat configuration items. Regards, Rainer - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: Parallel Tomcat 5 and 6 services on Windows
Robert Novotny wrote: Greetings, I would like to test Tomcat 6 on Windows, however I would still like to retain the previous Tomcat 5 installation (along with the Tomcat 5 service). I don't think the windows installer isn't set up to do this. I can't remember what the issue was but have a vague recollection of a bug report in this area. Using services.bat is the best work-around. Mark - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: increase the backend idle connection timeout or the connection_pool_minsize
Rainer Jung rainer.jung at kippdata.de writes: Yes, but the log message is talking about the min size. The default min size is size/2, but we changed it between 1.2.19 and 1.2.20 to (size+1)/2. So for prefork the min size was 0 until 1.2.19 and is 1 starting with 1.2.20. So I tried connection_pool_minsize=1 - no discernable effect on error rate I've tried tuning all the other suggested parameters without a noticable effect on the rate of these info messages (2000-3000/hour) I tried: maxActive 40, maxIdle 5, maxWait 3000 no discernable difference in error rate. Maybe I need to modify them with bigger deltas to see a difference? Is there a way to tell which parameter would have the most impact? If those are for a DataSource, they have no relation to your mod_jk messages. Yes, maxActive 40, maxIdle 5, maxWait 3000 are all in the context.xml file in the datasource definition - so these would not affect mod_jk messages - I get that now! I tried connectionTimeout = 60 (10 times the default of 60 seconds) - no reduction in error rate Are you sure, you put it into the correct connector? I put it into the server.xml AJP connector - actually I'm not sure now if I had the value in quotes: Server port=13135 shutdown=SHUTDOWN Service name=Catalina Connector port=13138 / Connector port=13139 protocol=AJP/1.3 connectionTimeout=60 / Engine name=Catalina defaultHost=localhost jvmRoute=tomcat-publicweb-01 Logger className=org.apache.catalina.logger.FileLogger / Host name=localhost appBase=webapps / /Engine /Service /Server Please post your version numbers, platform details, log example and relevant apache and tomcat configuration items. Web servers: Solaris 9 x86, Apache 2.0.52, mod_jk 1.2.20 App servers: Solaris 9 Sparc, tomcat 5.0 worker definition: worker.tomcat-publicweb-01.type=ajp13 worker.tomcat-publicweb-01.host=irt-app-01 worker.tomcat-publicweb-01.port=13139 worker.tomcat-publicweb-01.lbfactor=50 worker.tomcat-publicweb-01.cachesize=10 worker.tomcat-publicweb-01.cache_timeout=600 worker.tomcat-publicweb-01.socket_keepalive=1 worker.tomcat-publicweb-01.retries=2 worker.tomcat-publicweb-01.socket_timeout=30 worker.tomcat-publicweb-01.connect_timeout=2000 worker.tomcat-publicweb-01.prepost_timeout=5000 worker.tomcat-publicweb-01.reply_timeout=30 mod_jk errors: ajp_send_request::jk_ajp_common.c (1201): (tomcat-publicweb-01) error sending request. Will try another pooled connection ajp_send_request::jk_ajp_common.c (1227): (tomcat-publicweb-01) all endpoints are disconnected ajp_send_request::jk_ajp_common.c (1230): (tomcat-publicweb-01) increase the backend idle connection timeout or the connection_pool_minsize ajp_service::jk_ajp_common.c (1916): (tomcat-publicweb-01) sending request to tomcat failed, recoverable operation attempt=1 - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: [mod_jk] JkStripSession directive
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Rainer and Mladen, Rainer Jung wrote: Mladen Turk wrote: However, you are right, I made a bug in that code, it should trip the session from r-filename not r-uri. Will be fixed in the next tryout. ... which is now available for testing under the same URL. Looks like everything is working as expected. This seems much nicer than my old hack of using mod_rewrite. Perhaps this is actually more of a hack ;) Thanks for the new feature, guys! - -chris -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFF44EX9CaO5/Lv0PARApjDAJ0UeRalx6IsM0HDuXI6v+FUSsnYcwCgsWfZ 2nzLxqmiyLfFm/TLouKYMio= =lIdC -END PGP SIGNATURE- - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
Re: reload on demand?
Christopher Schultz chris at christopherschultz.net writes: I'm suggesting that you only deploy when you need to (even just copying files into your exploded WAR area). Are you updating your app many times more often than you want to reload it? Why? Because the project output folder in Eclipse is set to context/WEB-INF/classes, so every time a Java file is saved, the .class file is updated and the webapp is reloaded. What's I'd like is to not reload it then, but only when I access the webapp. I could set up the output folder to somewhere else and copy it to the exploded area only when I'd like to test run it. But this takes an extra manual step (copy) in addition to accessing the webapp in a browser. - To start a new topic, e-mail: users@tomcat.apache.org To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
.java(Permission denied) error in tomcat
hi, When i tried to access a jsp page the tomcat shows the following error message org.apache.jasper.JasperException: Unable to compile class for JSP and the root cause is java.io.FileNotFoundException: /usr/local/jakarta/jakarta-tomcat-5.5.9/work/Catalina/localhost/_/org/apache/jsp/Serv_jsp.java (Permission denied) Please any one help me. -- Regards, M.Muruganantham. CheapChess Development Team, Silicon Oyster Technologies, Chennai-84. www.sot.in
Tomcat monitoring
Does anyone know whats the best way to monitor a tomcat application remotely - I have seen the mailing lists and there are some tools that provide monitoring. However I need a way to programatically ping Tomcat and collect performance statistics - Any language would do : perl/java, I need to integrate the monitoring to another application. I am looking to collect the foll stats: - JVM Heap usage - Number of active threads - JDBC connection pool stats - Active thread list - Hits/Sec (running average) - TV dinner still cooling? Check out Tonight's Picks on Yahoo! TV.
