missing ddl for metrics

[BrixSat]

Hello we are running postgrest 12.2.6 on rocky linux 9 with postgresql 16 and we are having a lot of errors in the postgresql logs like the next one and we cannot find the ddl for the table anywhere. Even postgrest running as a superuser is not enough for it to create the table, what am i missing here?

{
  "timestamp": "2025-01-31 12:18:41.950 WET",
  "user": "",
  "dbname": "mydb",
  "pid": 530986,
  "remote_host": "localhost",
  "remote_port": 33416,
  "session_id": "679cbf88.81a2a",
  "line_num": 2,
  "ps": "PARSE",
  "session_start": "2025-01-31 12:18:16 WET",
  "vxid": "4/408",
  "txid": 0,
  "error_severity": "ERROR",
  "state_code": "42P01",
  "message": "relation \"public.metrics\" does not exist",
  "statement": "WITH pgrst_source AS ( SELECT \"public\".\"metrics\".* FROM \"public\".\"metrics\"     LIMIT $1 OFFSET $2 )  SELECT null::bigint AS total_result_set, pg_catalog.count(_postgrest_t) AS page_total, coalesce(json_agg(_postgrest_t), '[]') AS body, nullif(current_setting('response.headers', true), '') AS response_headers, nullif(current_setting('response.status', true), '') AS response_status, '' AS response_inserted FROM ( SELECT * FROM pgrst_source ) _postgrest_t",
  "cursor_position": 57,
  "func_name": "parserOpenTable",
  "file_name": "parse_relation.c",
  "file_line_num": 1428,
  "application_name": "PostgREST 12.2.6",
  "backend_type": "client backend",
  "query_id": 0
}

This logs every 20 seconds without doing anything at all.  What do i need to do to fix this?

[laurenceisla]

Hi,

I'm assuming you're trying to access the /metrics endpoint of the admin server? Verify that the admin-server-port you're using to query the admin server is correct. It looks like you're using the server-port instead (the one that connects to the database).

This logs every 20 seconds

This is a cron job that is executed right? If so, you should verify the correct admin server there.