#2975
- Created on2013-01-29 10:28
- Reported by
- StatusFixed
- PriorityNormal (3)
- Assigned to
- Milestone
- Plan Level-
- Due Date-
Backport cluster features and improvement to 2.1.1
(No description)
Leave a comment
on 2013-01-29 10:30 *
By Patrik Nordwall
Summary of new features and important bug fixes:
2547 - Cluster metrics-aware adaptive load balancing routers.
2704 - Load Cluster extension when ClusterActorRefProvider is used.
extensions = ["akka.cluster.Cluster"] is not needed any more
2779 - Publish AddressTerminated after a member is Downed/Removed,
previously this was done immediately when it was detected as
unreachable.
2306 - Configuration option (min-nr-of-members) defining how many
nodes the leader should wait for until moving nodes from
JOINING to UP. Also a registerOnMemberUp callback hook that
can for example be used for starting an actor when UP.
2907 - Detect failure when no heartbeats sent.
2895 - Cluster singleton actor pattern
Lots of small fixes and improvements, see related tickets.
Pull request here: https://github.com/akka/akka/pull/1064
2547 - Cluster metrics-aware adaptive load balancing routers.
2704 - Load Cluster extension when ClusterActorRefProvider is used.
extensions = ["akka.cluster.Cluster"] is not needed any more
2779 - Publish AddressTerminated after a member is Downed/Removed,
previously this was done immediately when it was detected as
unreachable.
2306 - Configuration option (min-nr-of-members) defining how many
nodes the leader should wait for until moving nodes from
JOINING to UP. Also a registerOnMemberUp callback hook that
can for example be used for starting an actor when UP.
2907 - Detect failure when no heartbeats sent.
2895 - Cluster singleton actor pattern
Lots of small fixes and improvements, see related tickets.
Pull request here: https://github.com/akka/akka/pull/1064