Details
-
Bug
-
Resolution: Fixed
-
Major
-
3.0
-
Security Level: Public
-
None
-
3.0.0-973-rel
Centos 6.4
-
Untriaged
-
Unknown
Description
Setup:
1. Cluster 7 nodes, 2 buckets, 1 design doc X 2 views
2. Load 120M, 99M items on both the buckets, dgm state of 70% active resident.
3. Do a graceful failover on 1 node
4. Choose delta recovery, add back node and rebalance
5. I tried to stop the rebalance a couple of times ( about 10 times) — Unsuccessful on a number of attempts.
Rebalance eventually failed with reason " Rebalance exited with reason stop" — Rebalance stop is not working as expected.
- Attaching logs
Attachments
For Gerrit Dashboard: MB-11775 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
39625,2 | MB-11775: log all process backtraces on slow rebalance stop | master | ns_server | Status: MERGED | +2 | +1 |
39692,2 | MB-11775: added more logging around stopping rebalance | master | ns_server | Status: MERGED | +2 | +1 |
39694,2 | MB-11775: anticipate rebalance stop while verifying replications | master | ns_server | Status: MERGED | +2 | +1 |
39707,2 | Revert "MB-11775: anticipate rebalance stop while verifying..." | master | ns_server | Status: MERGED | +2 | +1 |
39756,2 | MB-11775: anticipate rebalance stop while verifying replicas | master | ns_server | Status: MERGED | +2 | +1 |