Details
-
Bug
-
Resolution: Won't Fix
-
Critical
-
7.6.0
-
Operating System : Operating System : Debian GNU/Linux 11 (bullseye)
Initial Version : Couchbase Enterprise Edition 7.2.4-7070
Upgrade Version : Couchbase Enterprise Edition 7.6.0-2133
-
Untriaged
-
Linux x86_64
-
-
0
-
Unknown
Description
Steps to reproduce
- Created a couchbase cluster with the following configurations
- 172.23.104.66 - kv
- 172.23.105.179 - kv, n1ql
- 172.23.105.192 - fts
- 172.23.105.199 - fts
- 172.23.105.202 - fts
- Created a magma bucket named "test_bucket" and loaded 10k documents onto it
- Throughout the process of upgrades, there was constant mutations
- Created a search index - definition attached
- Throughout the process of upgrades, there was search queries
- Upgraded node 172.23.105.192 by failing over, upgrading and adding it back
- Upgraded node 172.23.105.179 in a similar manner
- Upgraded node 172.23.105.199 in a similar manner
- Upgraded node 172.23.105.202 in a similar manner
Rebalance fails
Rebalance exited with reason {{badmatch,
|
{leader_activities_error,
|
{default,rebalance}, |
{quorum_lost,
|
{lease_lost,'ns_1@172.23.104.66'}}}}, |
[{ns_rebalancer,rebalance,7, |
[{file,"src/ns_rebalancer.erl"},{line,456}]}, |
{proc_lib,init_p_do_apply,3, |
[{file,"proc_lib.erl"},{line,240}]}]}. |
Rebalance Operation Id = 714c0fc89e4b6049c2760a2bde8e86b4
|
Retried the rebalance multiple times and it always fails, hence marking this as a blocker