Details
-
Bug
-
Resolution: Done
-
Critical
-
4.0.0
-
None
-
Security Level: Public
-
None
-
10.6.2.144-10.6.2.147
-
Untriaged
-
-
Unknown
Description
RESOLUTION - Will not fix in Sherlock, need to document limitation.
"In <Sherlock>, it is not recommended to create/drop secondary indexes when any node with secondary index role is down. Doing so may result in duplicate index names."
1. Create 4 node cluster (1:n1ql, kv 2: kv 3:index 4:index)
2. Create default bucket with 2016 items - Employee data-set
3. Create GSI index "test_ambiguity_in_gsi_indexes_due_to_node_down"on node 3 using deployment plan
CREATE INDEX test_ambiguity_in_gsi_indexes_due_to_node_down ON default(join_yr) USING GSI WITH
4. Stop couchbase server on node 3
5. Create CREATE INDEX test_ambiguity_in_gsi_indexes_due_to_node_down ON default(join_yr) USING GSI
Step 5 should fail but is successful. When Node 3 is brought up, we have two GSI indexes with the same name
Attachments
Issue Links
- relates to
-
DOC-664 Secondary index warning
- Resolved