On Wed, Mar 18, 2015 at 10:56 PM, Bruce Momjian <bruce@momjian.us> wrote:
> I have researched this issue originally reported in June of 2014 and
> implemented a patch to ignore cancel while we are completing a commit.
> I am not clear if this is the proper place for this code, though a
> disable_timeout() call on the line above suggests I am close. :-)
This would also disable cancel interrupts while running AFTER
triggers, which seems almost certain to be wrong. TBH, I'm not sure
why the existing HOLD_INTERRUPTS() in CommitTransaction() isn't
already preventing this problem. Did you investigate that at all?
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company