Details
Description
1) 6 nodes cluster. Half a dozen of couchbase buckets, some of them had up to 400.000.000 items.
2) Data loaded via XDCR from the same Couchbase cluster in different AZ.
3) XDCR is removed.
4) Several test workloads run.
5) One of the nodes is stopped by EC2 management console (failover scenario test)
6) Node is failovered in CB web UI.
7) New node was started and configured. Added to cluster.
8) Rebalanced via CB web UI - 3 of all the buckets are rebalanced.
9) Continuous failing of reballance, log attached.
I've also downloaded the Diagnostic Report Log, but I'm not sure if it contains private info.