Details
-
Bug
-
Status: Closed
-
Critical
-
Resolution: Duplicate
-
1.5.4, 1.6.0, 1.6.1
-
None
-
None
Description
The new metrics format for latency tracking causes huge metrics cardinality explosion due to the format and the fact that there is a metric reported for a every combination of source subtask index and operator subtask index. Yikes!
This format is actually responsible for basically taking down our metrics system due to DDOSing our metrics servers (at Lyft).
Attachments
Issue Links
- is related to
-
FLINK-10226 Latency metrics can choke job-manager
- Closed