On Mon, Oct 9, 2017 at 2:29 AM, Peter Geoghegan <p...@bowt.ie> wrote: > On Sat, Oct 7, 2017 at 4:25 PM, Alvaro Herrera <alvhe...@alvh.no-ip.org> > wrote: >> Hmm, I think I added a random sleep (max. 100ms) right after the >> HeapTupleSatisfiesVacuum call in vacuumlazy.c (lazy_scan_heap), and that >> makes the race easier to hit. > > I still cannot reproduce. Perhaps you can be more specific?
I have been trying to reproduce things for a total of 4 hours, testing various variations of the proposed test cases (killed Postgres, changed fillfactor, manual sleep calls), but I am proving unable to see a regression as well. I would not think that the OS matters here, all my attempts were on macos with assertions and debugging enabled. At least the code is now more stable, which is definitely a good thing. -- Michael -- Sent via pgsql-hackers mailing list (email@example.com) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers