Realtime Compute offers job diagnosis to help you troubleshoot job issues.

Procedure

Note Only jobs that are in the running state can be diagnosed.
  1. Log on to the Realtime Compute development platform.
  2. In the top navigation bar, click Administration.
  3. On the Administration page that appears, find the target job, move the pointer over the More icon in the Actions column, and click Check.

Check metrics

  • Failover
    • Job failover: Check whether the job encountered a failover within the last 30 minutes.
    • Application Master (AM) failover: Check whether a failover is detected in AM.
  • Blink MetricJob latency: Check the job latency. If a latency occurs, the nodes with backpressure are displayed.
    • High latency: The latency is longer than 100 seconds and shorter than 200 seconds.
    • Excessively high latency: The latency is 200 seconds or longer.
  • This tab displays the Yarn check result.
  • This tab displays the OS check result.