Re: simple update query stuck

Поиск
Список
Период
Сортировка
Искать
От
Andrew Sullivan
Тема
Re: simple update query stuck
Дата
Msg-id
20140402024518.GA90415@crankycanuck.ca
Ответ на
Список
Дерево обсуждения
simple update query stuck Si Chen <sichen@opensourcestrategies.com>
Re: simple update query stuck Igor Neyman <ineyman@perceptron.com>
Re: simple update query stuck Si Chen <sichen@opensourcestrategies.com>
Re: simple update query stuck Andrew Sullivan <ajs@crankycanuck.ca>
Re: simple update query stuck David Johnston <polobo@yahoo.com>
Re: simple update query stuck Tom Lane <tgl@sss.pgh.pa.us>
Re: simple update query stuck Andrew Sullivan <ajs@crankycanuck.ca>
Re: simple update query stuck Si Chen <sichen@opensourcestrategies.com>
Re: simple update query stuck Paul Jungwirth <pj@illuminatedcomputing.com>
On Tue, Apr 01, 2014 at 07:00:16PM -0400, Tom Lane wrote:

> one of the clients, in a way that isn't visible to the deadlock detector.
> One way for that to happen without any external interconnections is if the
> client is waiting for a NOTIFY that will never arrive because the would-be
> sender is blocked.

I bet the case I was thinking of was the NOTIFY example.  That never
occurred to me as an explanation, but now that you mention it, it
seems quite likely to me.  

More generally (and for the OP's problem), my experience is that lots
of updates against the same rows in an unpredictable order is an
excellent way to run into trouble, and long-running transactions are a
major source of these problems.  Without a more detailed report about
what is going on in the present case, I don't think it's going to be
possible to diagnose better than has been done.

A

-- 
Andrew Sullivan
ajs@crankycanuck.ca

В списке pgsql-general по дате отправления
От: Adrian Klaver
Дата:
От: Bui, Michelle P
Дата:
FAQ