From: | Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com> |
---|---|
To: | Simon Riggs <simon(at)2ndQuadrant(dot)com> |
Cc: | Ryan Kelly <rpkelly22(at)gmail(dot)com>, tom Tom <tom(at)tomforb(dot)es>, pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: BUG #6629: Creating a gist index fails with "too many LWLocks taken" |
Date: | 2012-05-11 14:14:25 |
Message-ID: | 4FAD1EC1.2000502@enterprisedb.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
On 11.05.2012 16:56, Simon Riggs wrote:
> On 11 May 2012 11:07, Heikki Linnakangas
> <heikki(dot)linnakangas(at)enterprisedb(dot)com> wrote:
>
>> I wonder if we should reserve a few of the lwlock "slots" for critical
>> sections, to make this less likely to happen. Not only in this case, but in
>> general. We haven't seen this problem often, but it would be quite trivial
>> to reserve a few slots.
>
> Why reserve them solely for critical sections?
Because if you run out of lwlocks in a critical section, you get a PANIC.
> What is the downside from having>100 slots for general use.
>
> ISTM we should have 250 slots and log a warning if we ever hit 50 or more.
Then we would be back to square one, if a piece of code acquires 250
locks, then enters a critical section, and tries to acquire one more
lock -> PANIC.
--
Heikki Linnakangas
EnterpriseDB http://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2012-05-11 14:19:06 | Re: BUG #6624: Tab completion of identifier containing single backslash triggers warnings |
Previous Message | Heikki Linnakangas | 2012-05-11 14:11:36 | Re: BUG #6629: Creating a gist index fails with "too many LWLocks taken" |