From: | Simon Riggs <simon(at)2ndQuadrant(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, Tomas Vondra <tv(at)fuzzy(dot)cz>, Thom Brown <thom(at)linux(dot)com>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: change in LOCK behavior |
Date: | 2012-10-11 18:54:47 |
Message-ID: | CA+U5nM+uftwXBoXXLr_d2-nuh6pgDJJg81U+fjLyyksHzL_eDQ@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 11 October 2012 19:41, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Simon Riggs <simon(at)2ndQuadrant(dot)com> writes:
>> On 11 October 2012 18:22, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>>> If it worked, I might be amenable to that, but it doesn't. You can't
>>> trigger taking a new snapshot off whether we waited for a lock; that
>>> still has race conditions, just ones that are not so trivial to
>>> demonstrate manually. (The other transaction might have committed
>>> microseconds before you reach the point of waiting for the lock.)
>
>> So where's the race?
>
> Same example as before, except that the exclusive-lock-holding
> transaction commits (and releases its lock) between the time that the
> other transaction takes its parse/plan snapshot and the time that it
> takes AccessShare lock on the table.
A cache invalidation could also set the need-second-snapshot flag.
--
Simon Riggs http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2012-10-11 18:59:57 | Re: Truncate if exists |
Previous Message | Simon Riggs | 2012-10-11 18:48:45 | Re: change in LOCK behavior |