[jira] [Created] (FLINK-11887) Operator's Latency Metrics continues to increase

["Suxing Lee (JIRA)" <ji...@apache.org>]

Suxing Lee created FLINK-11887:
----------------------------------

             Summary: Operator's Latency Metrics continues to increase
                 Key: FLINK-11887
                 URL: https://issues.apache.org/jira/browse/FLINK-11887
             Project: Flink
          Issue Type: Bug
          Components: Runtime / Metrics
    Affects Versions: 1.6.3
            Reporter: Suxing Lee
         Attachments: flink_taskmanager_job_latency_source_id_operator_id_operator_subtask_index_1_latency.png

The operator's latency time is increased by approximately 2.7 minutes per day (see the attached).
We compute the latency by System.currentTimeMillis - marker.getMarkedTime.
There is no guarantee that System.currentTimeMillis and System.nanoTime don't drift apart.
If a GC pause or linux preemptive scheduling happenes, this should affect latency metrics.
Latency metrics drift away from their initial values with time(verify this result via the JVM Heap Dump).



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)