I can not really find any conclusive documentation on the different avenues for JMX invokation on a JBoss 6.0.0.Final.
The jboss.org guide SecureTheJmxConsole only specifies how to set an authentication realm for the jmx-console.war web app and JMX-connector. I feel like there is probably other ways of accessing JMX. What about RMI, HttpInvoker?
Ideally, I would like to know the best way to limit all JMX access to localhost.
So, how to go about making sure all avenues of JMX are accessible from localhost only?
There's an official documentation repository for Red Hat products, which includes JBoss EAP (the Red Hat product based on JBoss AS). This documentation is for EAP 5.x, but should be similar to JBoss AS 6 (EAP 5.x is based on JBoss AS 5.1).
http://docs.redhat.com/docs/en-US/JBoss_Enterprise_Application_Platform/5/html/Administration_Console_Quick_Start_Guide/ch04s02.html
In short, you'll have to change two files:
server/$PROFILE/conf/jboss-service.xml
server/$PROFILE/deploy/jmx-invoker-service.xml
Related
I am new to Solr world. I got a core indexed by Solr 4.6 and I want Solr 5.2 to take the existing core for searching. I have spent hours to figure out auto-discovery feature but no documentation exists.
My effort has been done here:
1) Quick start guide only tells you how to create one in 5.2
2) Upgrading a Solr 4.x Cluster to Solr 5.0 in Official 5.2 documentation touches the topic but does not provide useful hints (what is ZK_HOST? Why do I need it by the way?) . Plus, I don't want to make it a service without knowing it will work via command line.
I believe there must be a command option to set the core location and let Solr finds it. Could you share some useful hints?
Thanks
The way that you manage your Solr service changed quite a bit between 4.6 and 5.2.
First off, you now have a solr script in bin directory to manage the instance. You also have a solr.in.sh file for configuring your solr instance and you now stick all the solr configs in there, things like port, jvm parameters, etc...
Anyway, down to your question. Core Auto discovery scans your SOLR_HOME directory (specified in solr.in.sh) In that directory it expects to find directories that have core.properties files in them. Solr will try to attach any core it finds in there.
Other parameters like ZK_HOST are related to zookeeper. If you are running Solr Classic (with basic replication), you don't need to worry about that. However, if you are moving to SolrCloud, you will need to learn a bit about zookeeper.
Does anyone know if the Enterprise version of JBoss supports the concurrent utilities defined in JSR-236 and, if not, whether there is any roadmap for their support?
I couldn't find any info online and the only related RH knowledge base post is not accessible without a paid subscription.
JBoss EAP 7 does support the JSR-236 Concurrency Utilities. As far as I know this is the first release that supports them.
I've went through several examples (shown below) based on the hopes that I could get WCF to work on the 4.0 framework in Mono.
Installing a Mono-Server for WCF
Indigo Infocard
However, the link on http://www.mono-project.com/WCF didn't seem to contain a "guide" so to speak as to "here's how to setup wcf on linux with mono". On a side note. I think I've checked the first two pages of google for most wcf/mono keyword combinations to no avail.
Is it possible to easily get a WCF service up and going on linux? I presume there may be some configuration on the apache2 side? (I'm running a lamp server with Ubuntu 11.10).
I took the stock 4.0 WCF Service Library and tested deploying it to my server and accessing the *.svc to see if I got the wsdl, however, I did get a status code of 500.
As far as alternatives. I'm thinking that http://www.servicestack.net/ is probably the best?
So there are a couple of questions:
Can you get WCF to work on Linux easily? If so, how? (Windows was easily done)
SOA Alternatives. Should I look at asmx (wcf did replace this) or go with Service Stack.
My goal (in a basic sense) is to pass data to a service and have it compute work. Nothing is returned.
servicestack.net itself (and all the live examples) runs on an Ubuntu 12.04 on a Hetzner vServer. All our live demos run on Nginx/MonoFastCgi.
Here is how to run a ServiceStack self-hosted service inside a Linux daemon which also includes configuration of hosting it behind an Apache or Nginx reverse proxy. This post was written by a team who moved to ServiceStack to fix their memory and performance issues with MVC on Mono.
I generally avoid trying to run any heavy Microsoft web frameworks on Mono unless the Mono team has expressed supported configuration. For WCF they've stated only the "Silverlight subset" of WCF is supported.
Microsoft themselves don't support or test on Linux so you will generally have a better experience with others that do.
I've also answered what I believe are the main differences between ServiceStack, WebApi and WCF in an earlier question here.
Ubuntu 11.10 has a very old version of Mono.
If you use Ubuntu 12.04, as it includes a newer version, it may just work.
How to make acces to neo4j REST standalone server by LDAP? By default, there is no any security things in neo4j. In my opinion, i should run Apache Http server over neo4j, which use jetty inside. But I also know, that jetty can do LDAP, but it is part of neo4j, so its hard to configure. Wich way should I go?
Right now I think there are two possibilities. The first, as you mention, is to front Neo4j with Apache and let Apache take on the security workload.
The other is much more invasive, and that's to write a filter for JAX-RS (or a servlet filter) and get that registered with Jersey. If you're comfortable with hacking a bit of code, the second gives you a single box solution.
I have been able to configure Hyperic HQ to identify servicemix 4, however, It could only see the "ServiceMix 4.x Garbage Collector" service. I would like to be able to monitor activemq Queues, Endpoints, and Camel Applications (if possible)
Has anyone been able to do this?
I ran into the same issue at a client recently. We didn't find any support for doing this with HypericHQ, so we built our own basic web app using JSP, Ajax/jQuery and JMX to interface with AMQ/Camel. Another option is to customize the ActiveMQ and Camel web consoles...
Here are some notes on using JMX to monitor/manage both ActiveMQ and Camel...
http://benoday.blogspot.com/2010/08/managing-camel-routes-with-jmx-apis.html
http://benoday.blogspot.com/2010/08/monitoring-and-managing-activemq-with.html
Also, I know the Camel team is looking to enhance the web console in future releases (see the 3.0 roadmap)
good luck...