Details
-
Bug
-
Resolution: Fixed
-
Critical
-
5.5.0
-
Untriaged
-
Centos 64-bit
-
-
Yes
Description
Build 5.5.0-1979 / http://perf.jenkins.couchbase.com/job/hera/1606/
Setup:
- 4 nodes
- 1 bucket (full ejection, 1 replica)
- 1B items target (~1KB each)
- 80 client threads
Data ingestion is at least 3 times slower now. The disk write queue is almost always above 100M items.
Graphs:
Attachments
Issue Links
- relates to
-
MB-28421 Rebalance in DGM scenarios is slow and unstable in recent Vulcan builds
- Closed