Details
-
Bug
-
Resolution: Fixed
-
Major
-
4.5.1, 5.0.0
-
2 x E5-2680 v3 (48 vCPU), 256 GB RAM, CentOS 7
-
Untriaged
-
Centos 64-bit
-
-
Yes
Description
Setup
- Source cluster: 5 nodes, destination cluster: 4 nodes.
- 1 node is being added to the destination cluster.
- Uni-directional XDCR.
- 10K mixed KV ops/sec on both sides.
Problem
Rebalance makes no progress for 2 hours.
Please note, our performance tests are configured to terminate the execution after 2 hours of observed idleness. It's possible that rebalance would have finished after a longer period of time.
So far this issue happened twice. The logs are provided for both occurrences.
Attachments
Issue Links
- is duplicated by
-
MB-21050 Intermittent failure in ep_engine test 'dcp cursor dropping backfill'
- Closed
-
MB-23163 Rebalance deadlock with busy writer threads and no traffic
- Closed
- mentioned in
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...
-
Page Loading...