| From: | guntiso(at)latnet(dot)lv |
|---|---|
| To: | pgsql-general(at)lists(dot)postgresql(dot)org |
| Subject: | single-row deadlock |
| Date: | 2021-07-27 00:53:25 |
| Message-ID: | 2fcd64476bfaf1ca62f768508325825f@latnet.lv |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Hi,
After upgrading from PostgreSQL 9.6 to 13 we started to experience
locking issues once per week or two.
Replacing exclusive table lock with helper table row lock did not help.
Now it seems like a single-row deadlocks can occur, which should not be
possible (audit_data_lock table contains one row):
blocked_pid,blocking_pid,blocked_statement,current_statement_in_blocking_process,blocked_application,blocking_application
6752,6577,update audit_data_lock set locks_used = locks_used + 1,update
audit_data_lock set locks_used = locks_used + 1,PostgreSQL JDBC
Driver,PostgreSQL JDBC Driver
6577,6752,update audit_data_lock set locks_used = locks_used + 1,update
audit_data_lock set locks_used = locks_used + 1,PostgreSQL JDBC
Driver,PostgreSQL JDBC Driver
Is this a known issue? It seems like
https://postgrespro.com/list/thread-id/1231581. Are there any
workarounds?
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Ben Madin | 2021-07-27 01:43:45 | Re: pg_restore (fromuser -> touser) |
| Previous Message | Cory Nemelka | 2021-07-26 22:42:26 | Re: Logical replication from Rds into on-premise |