Details
-
Bug
-
Resolution: Fixed
-
Critical
-
7.1.0
-
7.1.0-2434-enterprise
-
Untriaged
-
Centos 64-bit
-
-
1
-
No
Description
Build: 7.1.0-2434
Scenario:
- 9 node cluster (4-KV, 3-Index and 2-N1ql nodes)
- Max auto failover count=2, auto failover timeout=30
- 1 Couchbase bucket with replica=1
- Bring down 4 nodes
+----------------+----------+----------------+----------------+
| Node | Services | Node status | Failover type |
+----------------+----------+----------------+----------------+
| 172.23.105.155 | kv | inactiveFailed | stop_couchbase |
| 172.23.136.109 | n1ql | active | stop_couchbase |
| 172.23.105.245 | index | active | stop_couchbase |
| 172.23.105.244 | index | active | stop_couchbase |
+----------------+----------+----------------+----------------+
Observation:
ns_server starts failover operation for KV and Index node and only KV node was failed over
Starting failing over ['ns_1@172.23.105.155','ns_1@172.23.105.244']
|
Index node not failed over since it is reported as unsafe. But no index was created for this test.
Could not automatically fail over node ('ns_1@172.23.105.244') due to operation being unsafe for service index. Safety check failed.
|
TAF test:
failover.concurrent_failovers.ConcurrentFailoverTests:
|
test_concurrent_failover,upgrade_version=7.1.0-2434,failover_order=kv:n1ql:index:index,maxCount=2,timeout=30,GROUP=P0,services_init=kv-kv-kv-kv-index-index-index-n1ql-n1ql,bucket_spec=single_bucket.default,get-cbcollect-info=False,replicas=1,bucket_storage=couchstore,log_level=error,failover_method=stop_couchbase,nodes_init=9,num_items=100000
|