Thread: pg_stat_statements has duplicate entries for the same query & queryId
Hi Folks!
Let me know if there's a better mailing list to ask this in.
I have a statistics collector that collects data from various postgres statistics tables, pg_stat_statements being one of them. This is done on an entire fleet of Postgres databases. From the collected data we record the timestamp of each collection in the query itself as extract(epoch from now()) as ts. What I'm seeing is that for the same query and query id, there are two rows with different statistics data at the same time. For example one row can have 2 calls while another can have 4. Anyone else run into this or have any idea why this can occur?
Regards,
Jevon Cowell
On 5/18/25 12:20, Jevon Cowell wrote: > Hi Folks! > Let me know if there's a better mailing list to ask this in. > > I have a statistics collector that collects data from various postgres > statistics tables, pg_stat_statements being one of them. This is done on > an entire fleet of Postgres databases. From the collected data we record > the timestamp of each collection in the query itself as extract(epoch > from now()) as ts. What I'm seeing is that for the same query > /and/ query id, there are two rows with different statistics data /at > the same time/. For example one row can have 2 calls while another can > have 4. Anyone else run into this or have any idea why this can occur? From here: https://www.postgresql.org/docs/current/pgstatstatements.html queryid bigint Hash code to identify identical normalized queries. query text Text of a representative statement From here: https://www.postgresql.org/docs/current/functions-datetime.html#FUNCTIONS-DATETIME-CURRENT "now() is a traditional PostgreSQL equivalent to transaction_timestamp()" and "transaction_timestamp() is equivalent to CURRENT_TIMESTAMP, but is named to clearly reflect what it returns." Therefore now() is pinned to the time the transaction started. Consequently it is conceivable that the queries actually ran at different times but got stamped with an identical timestamp via extract(epoch from now()) as ts. > > Regards, > Jevon Cowell -- Adrian Klaver adrian.klaver@aklaver.com