We have multiple instances of JBoss 5.1.0 running on the same physical server, using multiple instances of the same VM:
<JBOSS_HOME>/server/instance1
<JBOSS_HOME>/server/instance2
with the appropriate configuration to avoid port clashes etc.
For each instance there is the tmp folder:
<JBOSS_HOME>/server/instance1/tmp
<JBOSS_HOME>/server/instance2/tmp
To keep things separate, is it ok to set the java.io.tmpdir variable to <JBOSS_HOME>/server/<INSTANCE>/tmp?
Thanks in advance
Rich
Probably, but JBoss might be assuming it has full control over the contents of that directory. It might be a good idea to create a directory under <INSTANCE>/tmp, and set java.io.tmpdir to that, it would likely reduce the risk of any collisions.
Related
I need to append certain JARs to CLASSPATH env variable only for managed servers and not admin server. Also need to add -D arguments for managed server only.
What would be the best place to do this? I am thinking of doing it in "setDomainEnv.sh" but I guess it is called for both admin and managed servers. Would I need to use $SERVER_NAME and add IF/ELSE logic to append to CLASSPATH and JAVA_OPTIONS selectively only for managed servers?
Regards
Jacob
The recommandation from Oracle is not to modify the setDomainEnv.sh script but create a setUserOverrides.sh in the bin directory. If you read the setDomainEnv you will see it calls the the setUserOverrides if it exists.
As you said, these scripts are called for both admin and managed server. So, you have to write if/else logic to set your system properties.
does any one know how to deactivate the automatic clustering in a JBoss 5.1.0?
we have a JBoss running on each developer machine and because we are all in the same network, they do an auto clustering. The problem could be solved if each of us could get its own multicast ip, but the network hardware is not capable of that.
Isn't there a switch in jboss to deactivate this?
Under Eclipse under Windows, you can run the server using the following JVM property (see Open Launch Configuration) :
-Djboss.partition.name=${env_var:COMPUTERNAME}
This way each of the developer machine will have its own cluster (with a single server if you run only one server). Under Linux, you will need to replace COMPUTERNAME by HOSTNAME.
If you run JBoss AS from the command line, you would use something like -Djboss.partition.name=%COMPUTERNAME% under Windows (not tested).
Note that using -Djgroups.udp.ip_ttl=0 (as proposed in another answer) has the following drawbacks:
server startup is slower (4 minutes instead of 1 minute in my case);
there are a lot of NAKACK warn/error logs;
the JGroups UDP multicast is limited to the local machine which could conflict with other applications based on JGroupds UDP;
other servers on the same machine with the same configuration will be in the same cluster, which may not be desired.
You can use different multicast or partition name to avoid conflict.
However, if you want to disable clustering in "production" or "all" configuration , you need to do following actions:
Remove
farm/
deploy-hasingleton/
deploy/cluster/
In deploy/messaging/*-persistence-service.xml, change Clustered to false:
<attribute name="Clustered>false</attribute>
and remove
<depends optional-attribute-name="ChannelFactoryName">jboss.jgroups:service=ChannelFactory</depends>
In conf/bootstrap/profile.xml, replace
<bean name="BootstrapProfileFactory" class="org.jboss.system.server.profileservice.StaticClusteredProfileFactory">
with
<bean name="BootstrapProfileFactory" class="org.jboss.system.server.profileservice.repository.StaticProfileFactory">
and remove the "farmURIs" property a few lines below that.
Replace deploy/httpha-invoker.sar with http-invoker.sar from the default profile
In the deployers/clustering-deployer-jboss-beans.xml, comment out WebAppClusteringDependencyDeployer.
In SOA-P, if you are removing clustering, you will need to take a few additional steps.
Copy the server/default/deploy/jbpm.esb/hibernate.cfg.xml to server//deploy/jbpm.esb/hibernate.cfg.xml
Remove server//deploy/riftsaw* and cp -R server/default/deploy/riftsaw* server//deploy/
You can do this by setting the TTL (time-to-live) on the multicast packets to zero. Clustering will still be enabled, but none of the JBoss servers running on the developer machines will be able to locate each other.
When starting JBoss, set the jgroups.udp.ip_ttl system property, e.g.
-Djgroups.udp.ip_ttl=0
You'll need to hack that into the JBoss startup script, most likely.
I have a web app I'm writing in mod_perl 2. (It's a custom handler module, not registry or perlrun scripts.) There are several configuration options I'd like to have set at server initialization, preferably from a configuration file. The problem I'm having is that I haven't found a good place to pass a filename for my app's config file.
I first tried loading "./app.conf" but the current directory isn't the location of the modules, so it's unpredictable and error-prone. Or, I have to assume some path -- relative or absolute. This is inflexible and could be problematic if the host OS distribution is changed. I don't want to hard-code a path (though, something in /etc may be acceptable if there's just no better way).
I also tried PerlSetVar, but the value isn't available until request time. While this is workable, it means I'm potentially reading a config file from disk at least once per child (thread) init. I would rather load at server init and have an immutable static hash that is part of the spawned environment when a child is created.
I considered using a config.pl, but this means I either have a config.pl with one option to configure where to find the app.conf file, or I move the options themselves into config.pl and require end-users to respect Perl syntax when setting options. Future users will be internal admins, so that's not unreasonable, but it's more complicated than I'd like.
So what am I missing? Any good alternatives?
Usually a top priority is to avoid having configuration files amongst your executables. Otherwise a server misconfiguration could accidentally show your private configuration info to the world. I put everything the app needs under /srv/app0, with subdir cfg which is a sibling to the dirs containing executables. (More detail.)
If you're pre-loading modules via PerlPostConfigRequire startup.pl to access mod/startup.pl then that's the best place to put the configuration file location ../cfg/app.cnf and you have complete flexibility re how to store the configuration in memory. An alternative is to PerlModule your modules and load the configuration (with a relative path as above) in a BEGIN block within one of them.
Usually processing a configuration file doesn't take appreciable time, so a popular option is to lazy-load: if the code detects the configuration is missing it loads it before continuing. That's no use if the code needed to know the configuration earlier than that, but it avoids lots of problems, especially when migrating code to a non-modperl env.
I found couple of tutorials how to run multiple instances of JBoss on the same machine.
All of them mention uncommenting Service Binder and having separate service-binding.xml files for each server.
The question is why it's done like that? Is there any reason except adding additional layer of indirection?
It looks the same could be done by modification of ports in jboss-service.xml for each server. The only restriction would be that there won't be easy way to switch which instance of JBoss uses which set of ports.
You are right with modifying the ports in jboss-service.xml. This is the straightforward and genuine way to change the ports.
Unfortunately, ports are not only defined in that file, but also in other places like jboss-web's configuration etc.
Catching all those places can be error prone.
So the idea was to have a central file (service-binding.xml) that lives in the root of a server installation. You basically copy the 'default' config to server1, server2 etc and then via command line pass in the server name when starting so that the correct port-offset for all of the services is taken from service-bindings.xml and applied to the resulting runtime configuration.
JBossAS 7 takes this concept one step further to the ServiceBindingGroups, where the base ports are defined on a domain level and then per server you pick a basic group + just a port offset by name, so that there is even less work needed than in as4
Calling the JBoss command line tool Twiddle, located in $JBOSS_HOME\bin directory, can give us the port number on which JBoss is listening for HTTP requests (see Q2366489). That's smashing :)
However, we can't rely on the fact that the system administrator has set the JBOSS_HOME path variable, and it would be useful to know if we can obtain this directory programatically from within the JBoss environment. In my case the project is deployed as a WAR file.
TIA
This, and various other path-related values, are set by JBoss as system properties, which you should use to obtain them.
For example, the system property jboss.home.dir will be set to the equivalent of $JBOSS_HOME, regardless of how the server is actually started.
The various system properties are defined in org.jboss.system.server.ServerConfig.
So:
String jbossHomeDir = System.getProperty("jboss.home.dir");