Horton

Re: Concurrent ALTER SEQUENCE RESTART Regression

Michael Paquier <michael.paquier@gmail.com>

View thread View on postgr.es
From: Michael Paquier <michael.paquier@gmail.com>
To: Jason Petersen <jason@citusdata.com>
Cc: PostgreSQL mailing lists <pgsql-bugs@postgresql.org>, Peter Eisentraut <peter_e@gmx.net>
Date: 2017-04-25T05:43:58Z
Lists: pgsql-bugs, pgsql-hackers 
On Tue, Apr 25, 2017 at 2:18 PM, Jason Petersen <jason@citusdata.com> wrote:
> FWIW that was my gut read as well; take a slightly more restrictive lock,
> possibly blocking other ALTERs (unsure of prior behavior) to avoid ERROR but
> not at the cost of blocking readers. This seems about right to me.
>
> Haven't reported a bug before; what's next? Get a reviewer?

We have done everything that can be done, and for sure more review is
welcome. I have added an open item here:
https://wiki.postgresql.org/index.php?title=PostgreSQL_10_Open_Items
And that's a commit of Peter, who is also the author, now in CC, which
is causing the regression.
-- 
Michael

Commits

  1. Make ALTER SEQUENCE, including RESTART, fully transactional.

  2. Modify sequence catalog tuple before invoking post alter hook.

  3. Use weaker locks when updating pg_subscription_rel

  4. Add pg_sequence system catalog

  5. Modify sequence state storage to eliminate dangling-pointer problem