tree b193dbc7550c79b9b16d09f6f79e57f3e8c04d75
parent c318249efe354a3657186a6ebecdfa8513ed22c9
author Murtadha Hubail <mhubail@uci.edu> 1455781989 -0800
committer Murtadha Hubail <hubailmor@gmail.com> 1455788973 -0800

Asterix NCs Failback Support

- Allow Failed NCs to failback and takeover their partitions.
- New cluster API servlet for cluster state description.
- Remove nodeId from txn logs except remote FLUSH_LOG.
- Add partition id in UPDATE and Entity_COMMIT logs.
- Adapt remote recovery to new logs format.
- Refactor RecoveryManager and split Analysis and Redo phase.
- Spill remote recover logs to temporary file.
- Replicate files to remote replicas for partitions of interest only.
- Introduce NC active/inactive partitions concept.
- Test case for failback.

Change-Id: Id17819542d6b9c4e32647e64737c4a467b630f24
Reviewed-on: https://asterix-gerrit.ics.uci.edu/613
Tested-by: Jenkins <jenkins@fulliautomatix.ics.uci.edu>
Reviewed-by: abdullah alamoudi <bamousaa@gmail.com>