Uploaded image for project: 'WildFly'
  1. WildFly
  2. WFLY-19682

Improve client side node discovery process in a clustered environmnet

XMLWordPrintable

    • Icon: Enhancement Enhancement
    • Resolution: Unresolved
    • Icon: Major Major
    • None
    • 33.0.1.Final
    • EJB
    • None
    • ---
    • ---

      Given situation:

      • the client side node registry contains already a ConcurrentHashMap<String, Set<String>> for cluster nodes where the key is the cluster name with entries for multiple clusters
      • a client is invoking an EJB with a proxy having an affinity of Cluster "<cluster-name>"
      • when doing the cluster node discovery the RemotingEJBDiscoveryProvider.discover(ServiceType, FilterSpec, DiscoveryResult) method tries to connect to all nodes stored in the above mentioned ConcurrentHashMap<String, Set<String>> ignoring the cluster name. This lead to connection attempts against nodes not relevant for this specific EJB invocation.

      Optimization suggestion:

      • instead of connecting to all nodes of the ConcurrentHashMap<String, Set<String>> only the nodes of the cluster name given via the FilterSpec parameter should be contacted.

              tadamski@redhat.com Tomasz Adamski
              rhn-support-jbaesner Joerg Baesner
              Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

                Created:
                Updated: