Re: Visibility bug with prepared transaction with subtransactions on standby

2024-12-16 Thread Heikki Linnakangas
On 16/12/2024 10:00, Alexander Lakhin wrote: Recently hornet made some noise too: [1], by failing on the test modification added with e9c8747ee (in REL_13_STABLE): # issuing query via background psql: SELECT count(*) FROM t_009_tbl_standby_mvcc # pump_until: process terminated unexpectedly when

Re: Visibility bug with prepared transaction with subtransactions on standby

2024-12-16 Thread Alexander Lakhin
Hello Heikki, 27.06.2024 21:35, Heikki Linnakangas wrote: All I heard is crickets, so committed and backported to all supported versions. Recently hornet made some noise too: [1], by failing on the test modification added with e9c8747ee (in REL_13_STABLE): # issuing query via background psql:

Re: Visibility bug with prepared transaction with subtransactions on standby

2024-07-29 Thread a.kozhemyakin
Hi, I'm trying to run REL_13_STABLE recovey tests for windows and I get the error waiting for server to shut down.. failed pg_ctl: server does not shut down # pg_ctl stop failed: 256 # Stale postmaster.pid file for node "paris": PID 1868 no longer exists Bail out! pg_ctl stop failed

Re: Visibility bug with prepared transaction with subtransactions on standby

2024-06-27 Thread Heikki Linnakangas
On 20/06/2024 17:10, Heikki Linnakangas wrote: On 20/06/2024 16:41, Heikki Linnakangas wrote: Attached is a patch to fix this, with a test case. The patch did not compile, thanks to a last-minute change in a field name. Here's a fixed version. All I heard is crickets, so committed and backpo

Re: Visibility bug with prepared transaction with subtransactions on standby

2024-06-20 Thread Heikki Linnakangas
On 20/06/2024 16:41, Heikki Linnakangas wrote: Attached is a patch to fix this, with a test case. The patch did not compile, thanks to a last-minute change in a field name. Here's a fixed version. -- Heikki Linnakangas Neon (https://neon.tech) From ccb0bd36619f03055382205eddd7a0e9b64bba95 Mo

Visibility bug with prepared transaction with subtransactions on standby

2024-06-20 Thread Heikki Linnakangas
Hi, Konstantin and I found an MVCC bug with: - a prepared transaction, - which has a subtransaction, - on a hot standby, - after starting the standby from a shutdown checkpoint. See the test case in the attached patch to demonstrate this. The last query in the new test returns incorrect result