Uploaded image for project: 'Infinispan'
  1. Infinispan
  2. ISPN-13461

Cache should stop if the initial state transfer times out

This issue belongs to an archived project. You can view it, but you can't modify it. Learn more

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Done
    • Icon: Major Major
    • 14.0.0.Final, 13.0.12.Final
    • 10.1.8.Final, 11.0.11.Final, 12.1.7.Final, 13.0.2.Final
    • Core
    • None

      When any component fails to start, AbstractComponentRegistry.start() catches the component exception and stops all the components of the cache.
      Waiting for the initial state transfer to finish is done in CacheImpl.start(), which does not stop the component registry on an exception.

      This means the cache is unusable (DefaultCacheManager.getCache(name) re-throws the timeout exception from the initial state transfer) but it is still a member of the cluster, and other nodes still send commands to it.

      It's even worse if the initial state transfer failed because of a marshalling error: the state transfer will never finish, and other nodes will not be able to join.

              wburns@redhat.com Will Burns
              dberinde@redhat.com Dan Berindei (Inactive)
              Archiver:
              rhn-support-adongare Amol Dongare

                Created:
                Updated:
                Resolved:
                Archived: