# QAN PostgresSQL pg\_stat\_monitor not working **URL:** \ **Category:** pg\_stat\_monitor\ **Created:** [April 18, 2023, 12:45am UTC](https://forums.percona.com/t/qan-postgressql-pg-stat-monitor-not-working/21447 "2023-04-18T00:45:20Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![Luke\_Miller](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/luke_miller/32/10177_2.png) [@Luke\_Miller](https://forums.percona.com/u/Luke_Miller)\ **Post date:** [April 18, 2023, 12:45am UTC](https://forums.percona.com/t/qan-postgressql-pg-stat-monitor-not-working/21447/1 "2023-04-18T00:45:20Z")
I am trying to get QAN monitoring for my Postgres database. I have the Postgres server talking to PMM and I am getting OS and Postgres metrics but not QAN. database version: PostgreSQL 12.14 # pmm-admin list Service type Service name Address and port Service ID PostgreSQL dev-rms-db1-postgresql 127.0.0.1:5432 /service\_id/b40789d4-2fa4-42ab-91ff-d1e1737a97d7 Agent type Status Metrics Mode Agent ID Service ID Port pmm\_agent Connected /agent\_id/fcae9b4e-36f1-4870-a380-28a18d933e35 0 node\_exporter Running push /agent\_id/6f27184d-2164-4f84-b4d5-cd96a79c7294 42000 postgres\_exporter Running push /agent\_id/ba4c4f4d-d3e9-40f4-83ce-e7584619af43 /service\_id/b40789d4-2fa4-42ab-91ff-d1e1737a97d7 42002 postgresql\_pgstatmonitor\_agent Waiting /agent\_id/c4a94e90-7c2d-4c46-abd1-fed90eef9d41 /service\_id/b40789d4-2fa4-42ab-91ff-d1e1737a97d7 0 vmagent Running push /agent\_id/fbe0a404-e60a-4744-b6bc-d3ec03fdd4af 42001 I am seeing this error in the logs: pmm-agent[11021]: ERRO[2023-04-17T19:42:24.687-05:00] getNewBuckets failed: failed to query pg\_stat\_monitor: pq: column pg\_stat\_monitor.plans does not exist agentID=/agent\_id/c4a94e90-7c2d-4c46-abd1-fed90eef9d41 component=agent-builtin type=qan\_postgresql\_pgstatmonitor\_agent and in the Postgres logs: 2023-04-17 19:14:24.514 CDT [11205] ERROR: column pg\_stat\_monitor.plans does not exist at character 1170 2023-04-17 19:14:24.514 CDT [11205] HINT: Perhaps you meant to reference the column “pg\_stat\_monitor.planid”. 2023-04-17 19:14:24.514 CDT [11205] STATEMENT: SELECT /\* pmm-agent:pgstatmonitor \*/ “pg\_stat\_monitor”.“bucket”, “pg\_stat\_monitor”.“client\_ip”, “pg\_stat\_monitor”.“query”, “pg\_stat\_monitor”.“calls”, “pg\_stat\_monitor”.“shared\_blks\_hit”, “pg\_stat\_monitor”.“shared\_ blks\_read”, “pg\_stat\_monitor”.“shared\_blks\_dirtied”, “pg\_stat\_monitor”.“shared\_blks\_written”, “pg\_stat\_monitor”.“local\_blks\_hit”, “pg\_stat\_monitor”.“local\_blks\_read”, “pg\_stat\_monitor”.“local\_blks\_dirtied”, “pg\_stat\_monitor”.“local\_blks\_written”, “pg\_stat\_monit or”.“temp\_blks\_read”, “pg\_stat\_monitor”.“temp\_blks\_written”, “pg\_stat\_monitor”.“blk\_read\_time”, “pg\_stat\_monitor”.“blk\_write\_time”, “pg\_stat\_monitor”.“resp\_calls”, “pg\_stat\_monitor”.“cpu\_user\_time”, “pg\_stat\_monitor”.“cpu\_sys\_time”, “pg\_stat\_monitor”.“rows”, “p g\_stat\_monitor”.“relations”, “pg\_stat\_monitor”.“datname”, “pg\_stat\_monitor”.“userid”, “pg\_stat\_monitor”.“top\_queryid”, “pg\_stat\_monitor”.“planid”, “pg\_stat\_monitor”.“query\_plan”, “pg\_stat\_monitor”.“top\_query”, “pg\_stat\_monitor”.“application\_name”, “pg\_stat\_moni tor”.“cmd\_type”, “pg\_stat\_monitor”.“cmd\_type\_text”, “pg\_stat\_monitor”.“elevel”, “pg\_stat\_monitor”.“sqlcode”, “pg\_stat\_monitor”.“message”, “pg\_stat\_monitor”.“pgsm\_query\_id”, “pg\_stat\_monitor”.“plans”, “pg\_stat\_monitor”.“dbid”, “pg\_stat\_monitor”.“total\_time”, “pg \_stat\_monitor”.“min\_time”, “pg\_stat\_monitor”.“max\_time”, “pg\_stat\_monitor”.“mean\_time”, “pg\_stat\_monitor”.“stddev\_time”, “pg\_stat\_monitor”.“bucket\_start\_time” FROM “pg\_stat\_monitor” WHERE queryid IS NOT NULL AND query IS NOT NULL AND bucket\_done This does not show that column: \d pg\_stat\_monitor; View “public.pg\_stat\_monitor” Column | Type | Collation | Nullable | Default ---------------------±-------------------------±----------±---------±-------- bucket | bigint | | | bucket\_start\_time | timestamp with time zone | | | userid | oid | | | username | text | | | dbid | oid | | | datname | text | | | client\_ip | inet | | | pgsm\_query\_id | bigint | | | queryid | bigint | | | top\_queryid | bigint | | | query | text | | | comments | text | | | planid | bigint | | | query\_plan | text | | | top\_query | text | | | application\_name | text | | | relations | text | | | cmd\_type | integer | | | cmd\_type\_text | text | | | elevel | integer | | | sqlcode | text | | | message | text | | | calls | bigint | | | total\_time | double precision | | | min\_time | double precision | | | max\_time | double precision | | | mean\_time | double precision | | | stddev\_time | double precision | | | rows | bigint | | | shared\_blks\_hit | bigint | | | shared\_blks\_read | bigint | | | shared\_blks\_dirtied | bigint | | | shared\_blks\_written | bigint | | | local\_blks\_hit | bigint | | | local\_blks\_read | bigint | | | local\_blks\_dirtied | bigint | | | local\_blks\_written | bigint | | | temp\_blks\_read | bigint | | | temp\_blks\_written | bigint | | | blk\_read\_time | double precision | | | blk\_write\_time | double precision | | | resp\_calls | text | | | cpu\_user\_time | double precision | | | cpu\_sys\_time | double precision | | | bucket\_done | boolean | | | Do you have any ideas? Should I try adding the column? Did something not get installed correctly? Thanks, Luke ---
**Author:** ![browseman](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/browseman/32/98_2.png) [@browseman](https://forums.percona.com/u/browseman)\ **Post date:** [May 24, 2023, 2:45pm UTC](https://forums.percona.com/t/qan-postgressql-pg-stat-monitor-not-working/21447/2 "2023-05-24T14:45:38Z")
Hey Luke I think I had somewhat similar problem to the one described by you. In my case I solve it with multiple steps: - re-installing the packages `pmm-agent` and `percona-pg_stat_monitor12` (pmm-agent was updated but not the pg\_stat\_monitor) - reconfigure the extension based on the latest config from `https://docs.percona.com/pg-stat-monitor/` - `DROP/CREATE EXTENSION pg_stat_monitor;` - double check the permissions of the dedicated monitoring user and make sure it’s working - delete and re-add the pmm service - `pmm-admin remove/add postgresql [SERVICE_NAME]` Probably all of this wasn’t going to be needed if I simply execute the update procedure of the PG extension - [Upgrade - pg\_stat\_monitor Documentation](https://docs.percona.com/pg-stat-monitor/upgrade.html). May be you could give it a shot. Hope this is somewhat helpful. Cheers ---
**Author:** ![Jorge\_Torralba](https://sea1.discourse-cdn.com/flex019/user_avatar/forums.percona.com/jorge_torralba/32/3566_2.png) [@Jorge\_Torralba](https://forums.percona.com/u/Jorge_Torralba)\ **Post date:** [January 27, 2024, 6:51pm UTC](https://forums.percona.com/t/qan-postgressql-pg-stat-monitor-not-working/21447/3 "2024-01-27T18:51:07Z")
Make sure you connect to your postgres database and create the pg\_stat\_monitor extension in there.