Re: Undetected Deadlock

Rob Sargent <robjsargent@gmail.com>

From: Rob Sargent <robjsargent@gmail.com>
To: Michael Harris <harmic@gmail.com>
Cc: pgsql-general@lists.postgresql.org
Date: 2022-01-25T05:06:09Z
Lists: pgsql-general

> On Jan 24, 2022, at 10:02 PM, Michael Harris <harmic@gmail.com> wrote:
> 
> My apologies,
> 
> After posting this and looking at how it appears I realised that line
> wrapping makes the tables totally illegible.
> 
> Here they are again with all unnecessary columns removed and others shortened.
> 
> locktype | database |  relation  |   pid   |        mode         |
> granted |           waitstart           |     relation
> ----------+----------+------------+---------+---------------------+---------+-------------------------------+--------------------
> relation |   529986 | 1842228045 | 2130531 | AccessShareLock     | f
>     | 2022-01-19 00:32:32.626152+01 | st.ctr_table_efr_oa
> relation |   529986 | 1842228045 | 2128603 | AccessExclusiveLock | t
>     |                               | st.ctr_table_efr_oa
> relation |   529986 | 1842231489 | 2128603 | AccessExclusiveLock | f
>     | 2022-01-19 00:32:32.924694+01 | st.tpd_oa
> relation |   529986 | 1842231489 | 2130531 | AccessShareLock     | t
>     |                               | st.tpd_oa
> 
> 
>   pid   |    query_start     |   state_change     | wait_event_type |
> wait_event |                       query
> ---------+--------------------+--------------------+-----------------+------------+-------------------------------------------------------------
> 2128603 | 00:32:32.924413+01 | 00:32:32.924413+01 | Lock            |
> relation   | DROP TABLE st.tpd_oa_18929
> 2130531 | 00:32:32.625706+01 | 00:32:32.625708+01 | Lock            |
> relation   | DELETE FROM st.ctr_table_efr_oa_19010 WHERE ropid =
> 44788868
> 
> I hope this version is easier to read.
> 
> Cheers
> Mike
> On Tue, 25 Jan 2022 at 15:49, Michael Harris <harmic@gmail.com> wrote:
It’s definitely tidier, but on my mac Mail.app and on Thunderbird there is still a line wrap problem.  I don’t think that will prevent the experts from helping out.  I’m not one of them.




Commits

  1. Make timeout.c more robust against missed timer interrupts.

  2. Improve timeout.c's handling of repeated timeout set/cancel.