Websphere FAQ : Clustering, Deployment Manager & Node Agent
How does Deployment Manager and node Agent work together? Does deployment manager send message to node agent actively or node agent sends message to deployment manage?
It’s JMX-based. I suppose it’s pull, because the time interval is specified per Node Agent. When the Node Agent is started, it will discover the Deployment Manager, so it should be pretty direct.
How can I adjust time interval between the node agent and deployment manager?
Will my application run even if the deployment manager is down.
Does cluster members maintains IPs of other cluster members and can communicate with each other for the session persistence,including the session back up and retrieval, without the help of the deployment manager?
No, routing to cluster members is done by the plugin at the HTTP server. It has nothing to do with the deployment manager.
Who participate in clustering udder websphere?
For WAS 5 : DM and Node Agent participate in high-availability under WAS 5. Websphere cluster has capability of session maintenance, which means that normally if a cluster member is down, the request session it deals with can be forwarded to other cluster members,which have knowledge of this request session. Hence DM is much needed, If deployment manager do not know cluster member B to which the request is forwarded, cluster member B will not have knowledge of the request session and it might deal with the request incorrectly. So deployment manager should know its cluster members as soon as possible to distribute the session state.
But under WAS 6, DMGR can be stopped. Only admin tasks are not possible anymore if DMGR is down.
What happens if a cluster member is down? The session backup on this cluster member is lost for sure. How long it takes for the owner to detect the failure of the backup server? How can I adjust this time interval?
If a cluster member is down, the plugin routes to another member. As there is no in memory copy of the session, the new server will attempt to retrieve, either from the database or from a replica, depending on how you configured it. There are 2 options, memory to memory and database persistence to store session details.
What heppen if cluster member becomes alive again? How long it takes for other cluster members to detect it and how can I adjust this time?
Other cluster members don’t care.
If deployment manager is down,How can the session backup information transfered to other cluster members? Is web server necessary?
Not necessarily, no. The WAS plugin will automatically fail over requests for a down server to some other server in the same cluster. It’s up to you to configure session persistence so that the session is available to any other server in the cluster. You can use peer clustering or client server clustering, as described here: http://publib.boulder.ibm.com/infoce…ry2memory.html
How cluster session replication is done
Cluster members do not know each other, only web server and plugin know all cluster members. When web server receives a new requests, it will forward it to a cluster member WAS. For the session backup replication and only one replica is used), the web server or the plugin will choose another cluster member WAS and copy and update session state to that cluster member at the same time.
When the owner of the session is down, the web server will forward requests belonging to that session to the a new cluster member. If the new cluster member is the backup cluster member,then it already have the knowledge of the session; if not, then the new cluster member will ask the web server or the plugin to get the information of the session back up cluster member and finally get the session knowledge from that backup cluster member.
Note ;Session replication is not done by the plugin, that is done the data replication service.