Improve performance of DELETE_OLD_MARC_INDEXERS_SQL
Description
CSP Request Details
None
CSP Rejection Details
None
Potential Workaround
None
Checklist
hideActivity
Show:
Volodymyr Rohach January 9, 2025 at 10:09 PM
Hello everyone! This is a duplication ticket for [MODSOURCE-834] MARC Indexers Clean up is Taking Too Long - FOLIO Jira. It seems like in the scope of that ticket everything was fixed, so I’ll close this ticket.
CC:
Hongwei Ji December 6, 2024 at 1:43 PM
, dev is still working on the issue. For the time being, we updated mod-srs in all clusters including bugfest through an env var, so that expensive query will run less frequently (reduced from every 30 minutes to every 8 hours).
Eldiiar Duishenaliev December 6, 2024 at 1:03 PM
Hello ,
Could you please clarify is these needs to be performed on active BugFest environment?
And we need adopted SQL queries to be executed via psql please.
Strongly suggesting making this activity after BugFest completion.
Any onjections?
cc: , , , , ,
Duplicate
Details
Details
Assignee
Unassigned
UnassignedReporter
Hongwei JiPriority
P2Development Team
Folijet
Release
Ramsons (R2 2024) Bug Fix
RCA Group
Implementation coding issue
TestRail: Cases
Open TestRail: Cases
TestRail: Runs
Open TestRail: Runs
Created November 26, 2024 at 6:52 PM
Updated January 10, 2025 at 4:10 PM
Resolved January 9, 2025 at 10:09 PM
TestRail: Cases
TestRail: Runs
We observed very slow performance of this query . Depends on tenant data size, it could take minutes to complete. The worst part is that it caused the huge spike of Aurora storage IO usage and cost.
Following are some lines from database log to show the high duration numbers.