Details
-
Bug
-
Resolution: Duplicate
-
Major
-
5.5.0
-
Untriaged
-
Centos 64-bit
-
-
Yes
Description
A separate ticket for rebalance failures observed in MB-29217 and MB-26791.
Test scenario:
- 9 nodes
- 1 bucket, 1 replica, full eviction
- 1B items (~1KB), 5-10% resident ratio
- 15K ops/sec (90% read, 10% update), 10% cache miss ratio (before rebalance)
- Swap rebalance of one node (172.23.96.108 -> 172.23.96.109)
[user:error,2018-05-06T09:34:17.933-07:00,ns_1@172.23.96.100:<0.2270.0>:ns_orchestrator:do_log_rebalance_completion:1122]Rebalance exited with reason {child_died,bad_replicas}
|
[user:info,2018-05-06T09:34:17.932-07:00,ns_1@172.23.96.100:<0.241.36>:ns_rebalancer:verify_replication:985]Bad replicators after rebalance:
|
Missing = [{'ns_1@172.23.96.109','ns_1@172.23.96.107',1023}]
|
Previously:
2018-05-06T08:07:22.818569Z INFO (bucket-1) DCP (Consumer) eq_dcpq:replication:ns_1@172.23.96.109->ns_1@172.23.96.107:bucket-1 - Disconnecting because a message has not been received for 360s. lastMessageTime:361
|