Article Number: 528701

printer Print mail Email

Data Protection Advisor - Report began running slowly and now takes 22-23 hours

Primary Product: Data Protection Advisor 6.5

Product: Data Protection Advisor 6.5

Last Published: 20 Mar 2020

Article Type: Break Fix

Published Status: Online

Version: 4

Data Protection Advisor - Report began running slowly and now takes 22-23 hours

Article Content

Issue


This custom report started taking a much longer time to run than it had before. It used to complete in few minutes and now takes 22-23 hours scoped to a parent group of 20 backup servers. As per customer nothing has changed in the configuration of the report.
 
Troubleshooting path consisted of: 
  1. Timing the report scoped to lesser number of nodes in order to narrow down the issue.
  2. Gathering the report debug logs using both single node scope and parent group scope to examine the queries that take most of the time.
  3. The .wds export of the report.
The report completed in couple of minutes when scoped to either the 20 individually selected nodes or to any single node but when scoped to the entire parent group it took a very long time.
 
Cause
It was identified that for some reason PostgreSQL database is using a poor query plan for the backupjob table on the 20-node parent group.
 
Resolution
Workaround was running the report on the 20 individually selected child nodes.
 
More investigation is needed to determine why the entire 20-node group query started performing poor suddenly. It has occurred twice in few months.
 
 Contact Dell EMC Data Protection Advisor support team for further information.
Notes

Issue


This custom report started taking a much longer time to run than it had before. It used to complete in few minutes and now takes 22-23 hours scoped to a parent group of 20 backup servers. As per customer nothing has changed in the configuration of the report.
 
Troubleshooting path consisted of: 
  1. Timing the report scoped to lesser number of nodes in order to narrow down the issue.
  2. Gathering the report debug logs using both single node scope and parent group scope to examine the queries that take most of the time.
  3. The .wds export of the report.
The report completed in couple of minutes when scoped to either the 20 individually selected nodes or to any single node but when scoped to the entire parent group it took a very long time.
 
Cause
It was identified that for some reason PostgreSQL database is using a poor query plan for the backupjob table on the 20-node parent group.
 
Resolution

Workaround was running the report on the 20 individually selected child nodes.
 
More investigation is needed to determine why the entire 20-node group query started performing poor suddenly. It has occurred twice in few months.
 
 Contact Dell EMC Data Protection Advisor support team for further information.

Notes

Article Attachments

Attachments

Attachments

Article Properties

First Published

Thu Dec 20 2018 18:55:38 GMT

First Published

Thu Dec 20 2018 18:55:38 GMT

Rate this article

Accurate
Useful
Easy to understand
Was this article helpful?
0/3000 characters