Horton

Re: Potential G2-item cycles under serializable isolation

Peter Geoghegan <pg@bowt.ie>

View thread View on postgr.es
From: Peter Geoghegan <pg@bowt.ie>
To: Kyle Kingsbury <aphyr@jepsen.io>
Cc: PostgreSQL mailing lists <pgsql-bugs@lists.postgresql.org>
Date: 2020-06-01T04:20:30Z
Lists: pgsql-bugs 
Hi Kyle,

On Sun, May 31, 2020 at 9:05 PM Kyle Kingsbury <aphyr@jepsen.io> wrote:
> Oh! Gosh, yes, that DOES explain it. Somehow I had it in my head that both RU and RC mapped to SI, and RR & Serializable mapped to SSI. That's the case in YugabyteDB, but not here!

It's confusing because the standard only requires that the isolation
levels avoid certain read phenomena, but implementations are free to
go above and beyond. For example, you can ask Postgres for  READ
UNCOMMITTED, but you'll get READ COMMITTED. (So RC, RR, and SI each
provide distinct behavior.)

> I'll also see about getting a version of these tests that doesn't involve ON CONFLICT, in case that's relevant.

That would be great. It could easily be relevant.

Thanks
-- 
Peter Geoghegan

Commits

  1. Doc: Add references for SI and SSI.

  2. Improve comments for [Heap]CheckForSerializableConflictOut().

  3. Avoid update conflict out serialization anomalies.

  4. Implement genuine serializable isolation level.