If idle_in_transaction_timeout is bigger than transaction_timeout,
the idle-in-transaction timeout don't needed, right?
Yes, I think so.
TODO: as Yuhang pointed out prepared transactions must not be killed, thus name "transaction_timeout" is not correct. I think the name must be like "session_transaction_timeout", but I'd like to have an opinion of someone more experienced in giving names to GUCs than me. Or, perhaps, a native speaker?
How about transaction_session_timeout? Similar to idle_session_timeout.
Well, Yuhang also suggested this name...
Honestly, I still have a gut feeling that transaction_timeout is a good name, despite being not exactly precise.
Thanks!
Best regards, Andrey Borodin.
PS Sorry for posting twice to the same thread, i noticed your message only after answering to Yuhang's review.