Question
Truist
US
Last activity: 6 Oct 2023 6:26 EDT
How is the DB table pr_sys_statusdetails normally purged of older records in PRPC V 7.1.8?
We are observing constant deadlocks in PRPC V7.1.8 involving table pr_sys_statusdetails.
I observe the table contains 10 million rows and has not been purged since 2018.
Is this contributing to or causing the deadlocks? I'm guessing yes, please advise if otherwise.
I understand the default configuration of PRPC is to keep only the most recent data (7 days?) in pr_sys_statusdetails. Is that correct?
Which scheduled process normally removes that data? I see the Agent SystemCleaner which invokes the activity SystemCleaner, but I don't see where that activity cleans this specific data.
I also find a stored procedure, DBO.sppr_purge_table and an activity, pzSystemStatusPurge.
Both the activity and the stored proc clean old rows from pr_sys_statusdetails, but I don't see where they are invoked on a schedule. Are they?
***Edited by Moderator Marije to add Bug ID***
NCS Pte. Ltd
SG
Hi @EricSafern: Can you please check for agent "ClusterAndDBCleaner" in "Pega-RulesEngine".
Thanks.
Truist
US
@ArulDevan thanks for the suggestion. No ClusterAndDBCleaner - just SystemCleaner and StaticCleaner. Looks like ClusterAndDBCleaner was introduced after 7.1.8.
Truist
US
Hi, I'm still looking for assistance with this deadlock issue - where are these records purged? And can I / should I manually purge them?
Updated: 6 Oct 2023 6:26 EDT
Pegasystems Inc.
GB
@EricSafern Pega 7.1.8 is an old unsupported version. If other records in that table are getting purged, then it sounds like those particular rows were created in 2018 and were not cleaned properly for various reasons previously.
The deadlock issue is probably caused by the size of the table.
From a search in our archives (and a support article) I found that there was an issue with a build-up of records in pr_sys_statusdetails ( BUG-246364) and hotfixes were issued :
HFix-26018 for 7.1.8
HFix-27410 for 7.1.9
The issue was natively fixed from Pega 7.2 onwards.
Consider truncating pr_sys_statusdetails if it is very large, to prevent the backlog from causing performance issues with system cleaner agent. Also confirm that System Cleaner Agents (Node Cleaner and Cluster And DB Cleaner) are running regularly.
If you encounter deadlock issues even after fixing the table then I would suggest that you log a support incident if the log file output does not help you pinpoint the issue. (Likely you will be told to first upgrade).
We no longer have the 7.1.x purge documentation available, but if you want to read up on how purging works in later releases, take a look at the below links:
@EricSafern Pega 7.1.8 is an old unsupported version. If other records in that table are getting purged, then it sounds like those particular rows were created in 2018 and were not cleaned properly for various reasons previously.
The deadlock issue is probably caused by the size of the table.
From a search in our archives (and a support article) I found that there was an issue with a build-up of records in pr_sys_statusdetails ( BUG-246364) and hotfixes were issued :
HFix-26018 for 7.1.8
HFix-27410 for 7.1.9
The issue was natively fixed from Pega 7.2 onwards.
Consider truncating pr_sys_statusdetails if it is very large, to prevent the backlog from causing performance issues with system cleaner agent. Also confirm that System Cleaner Agents (Node Cleaner and Cluster And DB Cleaner) are running regularly.
If you encounter deadlock issues even after fixing the table then I would suggest that you log a support incident if the log file output does not help you pinpoint the issue. (Likely you will be told to first upgrade).
We no longer have the 7.1.x purge documentation available, but if you want to read up on how purging works in later releases, take a look at the below links:
Trimming, purging, and archiving tables
I would recommend that you update your environment to the latest Pega release .
Install and update Pega Platform
Installation and Update Documentation Roadmap