Horton

Re: [PATCH] Clarify the behavior of the system when approaching XID wraparound

Peter Eisentraut <peter@eisentraut.org>

View thread View on postgr.es
From: Peter Eisentraut <peter@eisentraut.org>
To: Peter Geoghegan <pg@bowt.ie>, John Naylor <john.naylor@enterprisedb.com>
Cc: Aleksander Alekseev <aleksander@timescale.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Pavel Borisov <pashkin.elfe@gmail.com>
Date: 2023-10-01T18:46:02Z
Lists: pgsql-hackers 
On 20.09.23 05:41, Peter Geoghegan wrote:
> On Sun, May 14, 2023 at 6:06 PM Peter Geoghegan <pg@bowt.ie> wrote:
>> BTW, Google cloud already just instruct their users to ignore the
>> xidStopLimit HINT about single user mode:
>>
>> https://cloud.google.com/sql/docs/postgres/txid-wraparound
> 
> I read this just today, and was reminded of this thread:
> 
> https://cloud.google.com/blog/products/databases/alloydb-for-postgresql-under-the-hood-adaptive-autovacuum
> 
> It reads:
> 
> "1. Transaction ID wraparound: PostgreSQL transaction IDs or XIDs are
> 32-bit unsigned integers that are assigned to each transaction and
> also get incremented. When they reach their maximum value, it would
> wrap around to zero (similar to a ring buffer) and can lead to data
> corruption."

What is the status of this patch discussion now?  It had been set as 
Ready for Committer for some months.  Do these recent discussions 
invalidate that?  Does it need more discussion?

Commits

  1. Reword messages about impending (M)XID exhaustion.

  2. Talk about assigning, rather than generating, new MultiXactIds.

  3. Update the documentation on recovering from (M)XID exhaustion.