Why doesn’t pg_stat_monitor 2.2 collapse such similar queries? Under high TPS, all buckets get filled with identical queries. As a result, it starts consuming more shared memory. It can collapse similar queries. In pg_stat_monitor, during just a few seconds of testing, there are 47,000 rows, with a limit of 20 buckets and pgsm_max of 192MB.
Is there a way to collapse similar queries? Which parameter in pg_stat_monitor is needed for this? I couldn’t find it in the documentation.
Hi @ANDiv1976 , it seems that there is some difference between these queries. Could you please check queryid, planid, userid, dbid, toplevel, and top_queryidfields in pg_stat_monitor view for this queries ? Which of them hold different values?
Hi @artem_gavrilov I hope the screenshot has all the necessary information.
I think you was hit by this bug: Jira
Try to disable pg_stat_monitor.pgsm_enable_query_plan = off
Looking forward to the bug fix. Thanks!
Unanswered | Unsolved | Solved
MySQL, InnoDB, MariaDB and MongoDB are trademarks of their respective owners.
Copyright © 2006 - 2024 Percona LLC. All rights reserved.
