Tag Info

Hot answers tagged

17

For nginx checkout their docs here. Proxy support is built in. In the example below from their site, you will see that specific port 80 traffic is being sent to a single servlet container running on port 8080. Note that if you want to run multiple backend servlet containers ( for load balancing, scaling, etc... ) you should look at the Upstream Fair ...


6

We have studied relative performance of JBoss vs. Glassfish, and found that Glassfish scales much better under high loads.


6

You can check my blog post on configuring Nginx as reverse proxy for Tomcat. It may be interesting to you, as example includes few additional tweaks as caching content only for not logged in users and redirecting for different languages.


5

you may want to try your questions over at stackoverflow.com. At my company the programming teams set up those application servers and then work with production support. We keep JBoss in our repository because of all the configuration changes and each server we deploy needs to be slightly modified. I've enjoyed playing with JBoss and the people in the ...


5

As already mentioned, you're probably experiencing leaking class loaders. For some reason your classes are not being unloaded. This can happen for two reasons Objects of said classes still exist on the heap, objects always reference their class, orThe class loader is referenced somewhere, for whatever reason, class loaders reference their classes in order to ...


5

Well first off, JDBC and ODBC are not compatible. JDBC was Java's answer to ODBC, and they fill the same niche, but you can't connect to an ODBC data source using JDBC and vice versa. Now there IS a jdbc-odbc bridge in Java, but that is a compatibility hack, not a native connector (The translation will go JDBC->ODBC->(Target Database) rather than ...


4

The default is admin/admin. You can find and modify this in: server/default/deploy/management/console-mgr.sar/web-console.war/WEB-INF/classes/web-console-users.properties Information can be found online (on page 3): JBoss Admin Console Quick Start Guide


3

You should try to run your script in a "screen". http://www.linuxmanpages.com/man1/screen.1.php screen -D -m run1.sh and then you can catch up the screen with: screen -r


3

We had the same problem for a file being written by log4j. The solution was to set the property "Append" for the FileAppender to "true". After this change, we have not seen this problem with files having NUL when rotated by an external program like logrotate.


3

From my experience - and the reason that we do not use logrotate with Log4j is that the way logrotate works is that it renames the files, then instructs the program to close its logs and reopen them with the old file name (which does not exist anymore), normally using the HUP signal. But Log4j cannot be told to reopen its log files, so I see you use ...


3

Can you upgrde to SCOM 2007. Interop gets a whole lot easier on that platform. With SCOM you can monitor JMX directly. See Introduction to SCOM 2007 Supported Platforms Cross Platform Extensions features. While there might be other options for MOM available. If I was going to spend a dime on a tool to enhance MOM, the upgrade would be where I spend it. ...


3

Personally I would never use PAE (over 4G of RAM on a 32-bit system). You'll get much better mileage running an actual 64-bit kernel and system. OOM should only trigger when a malloc might fail. (not when you have lots of swap available) The 32-bit kernel is likely to be part of the cause. PAE uses different memory zones and it may be that one zone isn't ...


3

Perhaps late but wanted to point out that EAP5 does have S3_PING support. As for community distribution JBoss AS 7 has that support (what you already know from Kaitsu's replay) but the feature is now supported. For EAP, you could go to the official EAP documentation that describes how to enable clustering with the Red Hat EAP AMIs. For JBoss Application ...


3

I'd use a connector in the virtual host instead of redirecting.


3

Modern OSes are actually pretty good about caching stuff in RAM. The best way to make sure is to force clear the cache (from another stack overflow: sync && echo 1 > /proc/sys/vm/drop_caches) and then the startup (and really only the startup) will be faster.


3

Either you're not actually using the tmpfs (probably because you're not actually running the copy of your app that you copied to the ramdisk, but there are other possibilities I'll let you explore), or your performance bottleneck wasn't disk in the first place.


3

Another way to do it like it's described in LikeApache wiki page. server { listen myhost:80; server_name myhost; location / { root /path/to/myapp/public; proxy_set_header X-Forwarded-Host $host; proxy_set_header X-Forwarded-Server $host; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; ...


3

Linux itself ran out of memory and killed Java. Your possible solutions would be to make the application use less memory (you can set a memory limit on the Java VM), or increase the memory (EC2 has a high-RAM instance).


3

you can either secure only connection from clients to your balancer and take balancer--workers network as trusted: client<--SSL-->balancer<--AJP/HTTP-->workers or you can secure the whole path (note: Balancer is actually a ManInTheMiddle attack by definition :-), so you will have to trust your balancer implicitly on workers...) ...


3

My go-to approach for things like this is to use the concat module, which will handle all of the notifications for building the config file itself, and wrap it in application-specific resource definitions. What I mean by that last part is something like this: class jboss { concat { '/path/to/jboss/config/file': owner => 'whoever', group => ...


3

According to this forum post JBoss AS7 management in OpenShift? in OpenShift site the management port is not accessible without first setting up port forwarding within OpenShift: The JBoss management console has been disabled on the JBoss cartridge embedded in OpenShift. [...] The management console is bound to an internal port (9990). To ...


2

What about logs, should they map to any particular location? Traditionally, you use /var/log/, or a directory therein. What is the best way to manage start and stop scripts on Ubuntu Server 8.10 for a Java Application Server? Ubuntu 8.10 still uses sysvinit for it's setup. Find the runlevel you need, and add a small initscript for run-parts. ...


2

The underlying cause is references to classes that have been discarded leaking outside their classloader, preventing the JVM from unloading those classes from the perm gen. Those flags you use may cause the JVM to aggressively purge classes that are unloadable, but it will not solve the underlying problem. There is a good, abeit complex explanation here


2

Any particular reason that you have an aversion to using the bin/jboss_init_redhat.sh SysV init script that comes already with the JBoss 5.1.0.GA release? I use it on several RHEL and CentOS systems and have absolutely no problem with JBoss starting up on reboot. I only had to do a minor change to get it to handle chkconfig without any problems. This change ...


2

After a long search I have found the answer to this one. The rewrite directive must be placed globally for the (virtual) host not in the or in a .htaccess. Apache appears not to actually parse those files as the files served out of mod_jk are not part of that structure; which makes sense if you think about it. It will however apply mod_rewrite rules that ...


2

I think you want log4j's SocketAppender / SocketReceiver functionality.


2

The suggested values are little more than guesses. Your best bet is to look at the values you have in your existing boot sequence and decide where you want Oracle and JBoss to fit in. e.g. start Oracle after the default OS stuff, and JBoss last.


2

The CGI script will run with the userid of the webserver, which might not be the userid of the person who was testing it. Have you considered that? The CGI script needs to print a header and some output like "Success, thanks for using auto.sh". echo "Content-type: text/plain" echo echo "Success, thanks for using auto.sh" If its too clunky ...


2

You can send a SIGQUIT signal to the running JVM to get each thread's stacktraces printed to stdout. This doesn't kill the process, though I think it does put all threads to sleep while the stack traces are being printed. Then, correlate the thread IDs listed with your preferred method of seeing thread-wise CPU utilization. prstat -L for Solaris, top -H ...



Only top voted, non community-wiki answers of a minimum length are eligible