Hi, JB Thank you very much for detailed answer. KARAF-376 looks great. I hope it will not be something like websphere clustering ;) I'm going to start from ssh script, as KISS solution for my requirements now. Lookig forward for the news about ACE.
Regards, Boris Jean-Baptiste Onofré <[email protected]> 13.01.2011 17:25 Please respond to [email protected] To [email protected] cc Subject Re: multi instance deployment (v2.0.0) Hi Boris, Your mail covers several points in fact: 1/ Artifact deployment By artifact, I mean bundle, feature or kar. Basically, you can the deploy folder using NFS to sync the bundle and feature. The problem is that if you install a bundle or feature using the shell, it won't be deployed on instances. I think that the easiest way is a kind of script, connecting using ssh and performing commands to replicate state. Another solution is to "code" a kind of deployment client using JMX layer of each karaf instances. Finally, we're working on Apache ACE/Apache Kalumet/AutoDeploy to provide a tooling for that. The console lets you manage the configuration and artifacts. These resources are "pushing" to an agent local to the Karaf instance. 2/ Karaf instances state sync After the deployment phase, you should expect that your Karaf instances are exactly in the same state (if a blueprint bundle is in GracePeriod state, we expect the same state on others karaf instances). We begun to deal with that yesterday, including maybe the usage of a Karaf manager, etc. This discussion is in KARAF-376: https://issues.apache.org/jira/browse/KARAF-376 I hope it helps. Regards JB On 01/13/2011 03:00 PM, [email protected] wrote: > > Greetings, folks. > > I'm looking for the solution for my failover requirements for bundles > run under Karaf 2.0.0. I would like to manage active and passive > instances of Karaf container. > My question is: what is the best way to perform deployment process on > several distributed Karaf container instances? > For example, I would like to install or update my software, how can I > achieve it for 2,3,4 container instances (nodes)? > > There are several ways I find: > 1. Use custom script with ssh, which will copy deployment archive in > auto deploy folder. > 2. Mount autodeploy folder on each node to the same location by NFS. > 3. Use feature url from mvn and build custom script, which do "refresh > url" for all instances. > > Please, advice. > Could you please share best practice with me or point me to some success > stories with such use case. > > Regards, > Boris Ilyushonak > > _______________________________________________________ > > The information contained in this message may be privileged and conf > idential and protected from disclosure. If you are not the original > intended recipient, you are hereby notified that any review, > retransmission, dissemination, or other use of, or taking of any action > in reliance upon, this information is prohibited. If you have received > this communication in error, please notify the sender immediately by > replying to this message and delete it from your computer. Thank you for > your cooperation. Troika Dialog, Russia. > > If you need assistance please contact our Contact Center (+7495) 258 > 0500 or go to www.troika.ru/eng/Contacts/system.wbp >
