On Thursday, November 5, 2015 12:16 AM, Jiří Hlinka <jiri(dot)hlinka(at)gmail(dot)com> wrote:
> My opinion is, that pg_repack should guarantee a consistent,
> deadlock-free behaviour via proper locking policy
I would be very interesting in seeing a description of what locking
policy would guarantee deadlock-free behavior when run concurrently
with unknown software. If you have a link to a paper on the topic,
that would serve as well as a description here.
--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company