mod_jk crash, please help
I am running apache 1.3 with mod_jk, my apache has apxs. I compiled and installed mod_jk from source, tag TOMCAT_4_1_12, configured it using the following options: ./configure --with-gnu-ld --with-apxs=/myapxpath/bin/apxs --enable-EAPI --with-java-home=/usr/java/jdk1.3.1 ran make installed mod_jk into the module directory, loaded fine. I created a virtual host in apache and added the following JkMount: JkMount /*.jsp wwwuser_wk My workers.properties looks as follows: workers.java_home=/usr/java/jdk1.3.1 workers.tomcat_home=/var/tomcat4 worker.list=wwwuser_wk worker.wwwuser_wk.host=localhost worker.wwwuser_wk.port=5266 worker.wwwuser_wk.type=ajp13 worker.wwwuser_wk.lbfactor=50 worker.wwwuser_wk.cachesize=10 worker.wwwuser_wk.cache_timeout=600 worker.wwwuser_wk.socket_keepalive=1 worker.wwwuser_wk.socket_timeout=300 I startup apache and tomcat, then i point my browser to xyz.com/index.jsp I get nothing back on the browser, and the following in error_log [notice] child pid 19817 exit signal Segmentation fault (11) I added debugging code to mod_jk and traced it to the following line in mod_jk.c s-remote_host = (char *)ap_get_remote_host(r-connection, r-per_dir_config, REMOTE_HOST); REMOTE_HOST seems to be NULL for that since i printed all the stuff from there. Please help. How can i trace this further? How do i get apache to generate a core file so i can see where in apache code this is? Any pointers for help here at all? Eugene -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Status of tomcat (up/down)
I am running multiple tomcat instances on one server (user TOMCAT_BASE setup). I need to know whether a particular tomcat instance is up or down for monitoring purposes. The tomcats I am running only have the shutdown port and ajp13 port bound. Is a socket connection without a command to the shutdown port enough? Is there any way to connect to the ajp13 port and do a dummy request? Is there a java api for client side ajp13 to help with this? Any help would be appreciated, Eugene -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: [POLL] aprize jk2
I am against only because many apache 1.3 users (who installed apache from RPMs, like me) will not be able to use jk2, unless there an easy way to install apr libs/includes, etc on top of apache1.3 Is there? -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: [PATCH] [JK2 docs] Minor patch to jk2 docs
Can anyone check this in?
Eugene Gluzberg wrote:
Just some additional installation info.
Index: jk/xdocs/jk2/configweb.xml
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/jk/xdocs/jk2/configweb.xml,v
retrieving revision 1.6
diff -c -r1.6 configweb.xml
*** jk/xdocs/jk2/configweb.xml20 Sep 2002 21:35:31 - 1.6
--- jk/xdocs/jk2/configweb.xml23 Sep 2002 21:20:46 -
***
*** 100,105
--- 100,115
thDescription/th
/tr
tr
+ tdfile/td
+ td${serverRoot}/conf/workers2.properties/td
+ tdLocation of the workers2.properties file/td
+ /tr
+ tr
+ tddebug/td
+ td0/td
+ tdSet the debug level of the config component/td
+ /tr
+ tr
tddebugEnv/td
td0/td
tdSet the debug level of the hidden env component /td
Index: jk/xdocs/jk2/installhowto.xml
===
RCS file: /home/cvspublic/jakarta-tomcat-connectors/jk/xdocs/jk2/installhowto.xml,v
retrieving revision 1.2
diff -c -r1.2 installhowto.xml
*** jk/xdocs/jk2/installhowto.xml 20 Sep 2002 21:35:31 - 1.2
--- jk/xdocs/jk2/installhowto.xml 23 Sep 2002 21:20:46 -
***
*** 3,8
--- 3,9
properties
titleInstallation of mod_jk2 in the Web Server/title
author email=[EMAIL PROTECTED]Jean-Frederic Clere/author
+ author email=[EMAIL PROTECTED]Eugene Gluzberg/author
date$Date: 2002/09/20 21:35:31 $/date
/properties
section name=Installation
***
*** 14,19
--- 15,34
installation.
/p
subsection name=Apache 1/
+ p
+For this example, we are running on a standard RH 7.3 system Apache 1.3 rpm
installed.
+The commands are executed in the jakarta-tomcat-connectors directory.
+These instructions should also work with other RH 7.x systems..
+ /p
+ screen
+ noteCopy the dso files in the modules location:/note
+ typecp jk/build/jk2/apache2/mod_jk2.so /etc/httpd/modules/type
+ noteAdd mod_jk2 loading in the httpd.conf:/note
+ readLoadModule jk2_module modules/mod_jk2.so/read
+ readAddModule mod_jk2.c/read
+ notePoint mod_jk2 to your workers2.properties file inside
httpd.conf::/note
+ typeJkSet config.file /etc/httpd/conf/workers2.properties/type
+ /screen
subsection name=Apache 2
p
In the following example Apache-2.0 is installed in
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
Re: [JK2] new uriMap using hostname:port
Yikes! today's version no longer compiles without APR on 1.3:
[so] Output:
[so] rm -f
/home/eugene/tomcat/jakarta-tomcat-connectors/jk/build/jk2/apache13/common/.libs/jk_uriEnv.lo
[so] cc -c
-I/home/eugene/tomcat/jakarta-tomcat-connectors/jk/native2/include
-I/home/eugene/tomcat/jakarta-tomcat-connectors/jk/native2/common
-I\${build.compiler.base}/include -I/usr/include -I/usr/include/apache
-I/home/eugene/tomcat/jakarta-tomcat-connectors/jk/native2/include
-I/usr/java/j2sdk1.4.1/jre/../include
-I/usr/java/j2sdk1.4.1/jre/../include/linux -g -W -DEAPI -D_REENTRANT
-DCHUNK_SIZE=4096 -DHAVE_MMAP
/home/eugene/tomcat/jakarta-tomcat-connectors/jk/native2/common/jk_uriEnv.c
-fPIC -DPIC -o
/home/eugene/tomcat/jakarta-tomcat-connectors/jk/build/jk2/apache13/common/.libs/jk_uriEnv.lo
[so] StdErr:
[so]
/home/eugene/tomcat/jakarta-tomcat-connectors/jk/native2/common/jk_uriEnv.c:72:21:
apr_uri.h: No such file or directory
Costin Manolache wrote:
All that matter to me is to preserve the equivalence with
'native' apache config.
In other words,
[uri:HOST:PORT/URL]
properties
Must be equivalent with
VirtualHost addr:port ...
ServerAlias ...
Location /URL
JkUriSet
/Location
/VirtualHost
By this criteria, I would prefer use of '_default_' for the default
server.
For the syntax of HOST, port, URL - just make sure it matches the
documented semantics of apache VirtualHost and Location.
Costin
Mladen Turk wrote:
Think I didn't miss something :)
Here is how it works:
All the mappings are done in the following way:
hostname|*[:port]
Meaning '*' is used for the default host.
You can make now something like:
[uri:*:8080]
Info=will map the all the default servers on the port 8080
Aliases should work now:
[uri:*]
alias=localhost
alias=www.apache.org
I have dropped the reverse attribute for now. Its too clumsy, and it
will be enabled if pcre is ever adopted.
IIS (work to some extend) and Apache 1.3 need some extra tuning :-)
Please do test.
MT.
--
To unsubscribe, e-mail: mailto:[EMAIL PROTECTED]
For additional commands, e-mail: mailto:[EMAIL PROTECTED]
[PATCH] Loadbalancer (works in 3.3x)
Hi everyone, I was using the load balancer worker with Netscape and I noticed that the load balancer worker takes a long time to recover whenever all of the tomcats are shut down. I am using version 3.2.2. However I checked out the latest 3.3 code and the load balancer is exactly the same as 3.2.2. The patch I am submitting is against the latest 3.3 code. I did not have a chance to look at the 4.x code. Reason for patch: If a tomcat goes down, it gets taken out of the worker list for 60 seconds, before its retried. If all the tomcats that are being load balanced are being restarted (one at a time to upgrade the system for instance), you need to wait until the tomcat that you just brought up to get back in the list before you shut down the next one, or you may have website downtime. Same problem if any intermittent network problem would occur between Netscape and tomcat. The downtime would be at least 60 seconds. To avoid that I made a fix to the load balancer worker that would be activated only when all the tomcat workers were taken out of the list due to failure. If all the load balanced tomcat workers are out of the list, the patched load balancer would go through all of the workers again once in reverse order of their last error time until one is found. They would be retried even though 60 seconds did not pass yet. If and only if all the workers have been tried once in this particular request and all failed an error would be returned to the user. The retries would continue for every request until at least one tomcat worker recovers. After the first tomcat worker recovers, the rest would be retried once every 60 seconds as before. I found this very useful in our environment. Patch attached Thanks, Eugene Gluzberg lb.patch Description: Binary data -- To unsubscribe, e-mail: mailto:[EMAIL PROTECTED] For additional commands, e-mail: mailto:[EMAIL PROTECTED]
