LinkedIn
Copied!

PEGA0062 alert: Data flow execution time above threshold

The PEGA0062 alert is triggered when the total data flow run time has exceeded the threshold.

This article covers the following topics:

  1. Example message text for single case data flow runs
  2. Example message text for batch or real-time data flows
  3. Recommended next steps

Example message text for single case data flow runs

Data flow record processing time exceeds threshold of <threshold> ms: <observed value> ms. For data flow [<data flow>] run [<run ID>] node [<node ID>]. Slowest stage [<stage name>] with avg <stage observed value> ms over <record count>. SLA violations since last alert: <SLA violations number> (<percentage> of requests). Data flow metrics are: <data flow statistics>. Node data flow metrics: <node data flow statistics>

The message text contains the following variables:

  • <threshold> is the time limit for the data flow processing. The default setting is 10000 milliseconds. 
  • <observed value> is the actual data flow processing time.
  • <data flow> is the name of the data flow that exceeded the processing time during execution.
  • <run ID> is the run during which the processing time was exceeded.
  • <node ID> is the 32-character node identifier.
  • <stage name> is the name of the slowest stage of the data flow run.
  • <stage observed value> is the average time for processing a record in the slowest stage.
  • <record count> is the number of records that the slowest stage processed.
  • <SLA violations number> is the number of SLA violations since the last PEGA0062 alert.
  • <percentage> is the number of requests that did not meet the SLA out of all requests.
  • <data flow statistics> are the statistics for the data flow run over the specified time interval.
  • <node data flow statistics> are the aggregated statistics for the data flow on the current node since the start of the run.

Example message text for batch or real-time data flow runs

Data flow record processing time exceeds threshold of <threshold> ms: <observed value> ms with rate <records number> records per second over time interval <time interval> seconds. For data flow [<data flow>] run [<run ID>] node [<node ID>]. Slowest stage [<stage name>] with avg <stage observed value> ms over <record count>. Data flow metrics for interval: <data flow statistics>. Node data flow metrics are: <node data flow statistics>

The message text contains the following variables:

  • <threshold> is the time limit for the data flow processing. The default setting is 10000 milliseconds.
  • <observed value> is the actual data flow processing time.
  • <records number> is the average number of data flow records processed for each second.
  • <time interval> is the time during which the number of records specified in <records number> was processed. The default setting is 300 seconds.
  • <data flow> is the name of the data flow that exceeded the processing time during execution.
  • <run ID> is the run during which the processing time was exceeded.
  • <node ID> is the 32-character node identifier.
  • <stage name> is the name of the slowest stage of the data flow run.
  • <stage observed value> is the average time for processing a record in the slowest stage.
  • <record count> is the number of records that the slowest stage processed.
  • <data flow statistics> are the statistics for the data flow run over the specified time interval.
  • <node data flow statistics> are the aggregated statistics for the data flow on the current node since the start of the run.

Recommended next steps

Address the PEGA0062 alert by performing the following actions:

  1. Analyze the data flow statistics and the Data Flow Runs landing page.
    For more information, see Tips for troubleshooting data flow rules.
  2. Based on the information about the slowest stage, identify the component that causes the excessive data flow run time and modify that component.
    For more information, see Configuring data flows.
  3. Consult the PEGA0082 and PEGA0083 alerts to determine what other data flows were running at the time and if those data flows were the source of an overloaded system.
    For more information, see PEGA0082 alert: Dataflow started or Dataflow resumed and PEGA0083 alert: Dataflow completed, Dataflow paused, or Dataflow stopped.
  4. Search for other Pega alerts with the same correlation ID.
  5. If you do not identify any reasons for the exceeded run time, consider increasing the run time threshold by configuring a corresponding dynamic system setting:
    • dataflow/singlecase/alert/targetTimePerRow for single case data flows
    • dataflow/batch/alert/targetTimePerRow for batch data flows
    • dataflow/realtime/alert/targetTimePerRow for real-time data flows
      For more information, see Editing a dynamic system setting.
Suggest Edit

0% found this useful


Related Content

Have a question? Get answers now.

Visit the Collaboration Center to ask questions, engage in discussions, share ideas, and help others.