Details
-
Bug
-
Resolution: Fixed
-
Major
-
Cheshire-Cat
-
7.0.0-4678-enterprise
-
Untriaged
-
Centos 64-bit
-
-
1
-
Unknown
Description
Build: 7.0.0-4678
Scenario:
- 4 node cluster, couchbase bucket (replica=2)
- Failover the orchestrator node (.12) and rebalance the cluster
- Add back the rebalanced out node (.12) back
Observation:
Step #3 fails with "Prepare join failed. Node already part of cluster" even though the same node has been rebalanced out successfully from the cluster.
TAF testcase:
guides/gradlew --refresh-dependencies testrunner -P jython=/opt/jython/bin/jython -P 'args=-i /tmp/oel6-4node-rebalance_in_jython.ini rerun=False,get-cbcollect-info=False,skip_cleanup=True,upgrade_version=7.0.0-4678 -t rebalance_new.swaprebalancetests.SwapRebalanceFailedTests.test_add_back_failed_node,standard_buckets=1,swap-orchestrator=True,num-swap=1,sdk_timeout=50,replicas=2,durability=MAJORITY_AND_PERSIST_TO_ACTIVE,log_level=error,nodes_init=4,num_items=100000'
|