Details
-
Bug
-
Resolution: Fixed
-
Critical
-
5.5.0
-
Untriaged
-
No
Description
Reported by Mihir on MB-29517.
If an index movement is in progress and the node fails over, on recovery, there can be duplicate index instances.
Logs:
https://s3.amazonaws.com/bugdb/jira/MB-29517-0516/collectinfo-2018-05-16T061335-ns_1%4010.112.180.101.zip
https://s3.amazonaws.com/bugdb/jira/MB-29517-0516/collectinfo-2018-05-16T061335-ns_1%4010.112.180.102.zip
https://s3.amazonaws.com/bugdb/jira/MB-29517-0516/collectinfo-2018-05-16T061335-ns_1%4010.112.180.103.zip
Attachments
Issue Links
- relates to
-
MB-29717 Duplicate index if node fails during index move
- Closed
For Gerrit Dashboard: MB-30576 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
97467,1 | MB-30576 block move index if failover is in progress | spock | indexing | Status: MERGED | +2 | +1 |
97468,1 | MB-30576 do not clean up orphan tokens on failover | spock | indexing | Status: MERGED | +2 | +1 |
97469,1 | MB-30576 fix log message | spock | indexing | Status: MERGED | +2 | +1 |
97470,4 | MB-30576 make transfer token cleanup more robust | spock | indexing | Status: MERGED | +2 | +1 |