Hello, We have statement_timeout, idle_in_transaction_timeout, idle_session_timeout and many more! But we have no transaction_timeout. I've skimmed thread [0,1] about existing timeouts and found no contraindications to implement transaction_timeout.
Nikolay asked me if I can prototype the feature for testing by him, and it seems straightforward. Please find attached. If it's not known to be a bad idea - we'll work on it. Thanks! Best regards, Andrey Borodin. [0] https://www.postgresql.org/message-id/flat/763A0689-F189-459E-946F-F0EC4458980B%40hotmail.com
v1-0001-Intorduce-transaction_timeout.patch
Description: Binary data
