I have a few long running transacrions that both select and update data. When the jobs are running, I suddently cannot run a basic SELECT + JOIN query on the table? It simply hangs forever... I thought mvcc would allow me to always read from the tables? (its fine to see the old version of the data). I'm not sure if the query is blocked or just slowed down (to the extreme) when the jobs are running.
I use the default read commited transaction level. pg 13.
Theories:
Has the query become extremly slow because I'm updating data in the table concurrently, and mvcc has to keep track of which version of each row the query should see?
Could it be because there are no more pg connections left in the connection pool for the SELECT query to run?
