Details
-
Bug
-
Resolution: Fixed
-
Critical
-
6.6.0, 6.6.1, 6.6.2, Cheshire-Cat
-
None
-
Untriaged
-
1
-
Unknown
Description
The most common scenario or restart vbucket getting triggered is kv rebalance. In that case, it will be helpful to use latest vbucket map, as compared to the cached one.
With all the cluster info cache optimisations, it should not be very costly to fetch vbmap only for one bucket.
Attachments
Issue Links
- backports to
-
MB-45541 [BP 6.6.2 MB-45358] - [CBSE] stale vbmap in projector causes stale=false scans to timeout during KV rebalance
- Closed