DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com;
	h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding;
	b=MVSyXAvwPdVRoJZvVsv51uwkHM9D/0Fu8mtGo0FffbdLmPXY75ruj17q0XElZm96I9R6fe5l25okeGC0HfpPiq2Jn+lQo/xX1N1DYWZmQcC9xB+VbordVkY6EUgyyhosxmIpV0uKaXWUjlZcpN28afZYt4spGfdPDXGIwz1Xb/Q=;
Message-ID: <816542.35556.qm@web39704.mail.mud.yahoo.com>
Date: Mon, 15 Feb 2010 12:20:50 -0800 (PST)
From: Bob Lunney <bob_lunney@yahoo.com>
Subject: Re: AIX - Out of Memory
To: Kenneth Marshall <ktm@rice.edu>
Cc: pgsql-admin@postgresql.org
In-Reply-To: <20100215161823.GB16501@it.is.rice.edu>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

=0A=0A--- On Mon, 2/15/10, Kenneth Marshall <ktm@rice.edu> wrote:=0A=0A> Fr=
om: Kenneth Marshall <ktm@rice.edu>=0A> Subject: Re: [ADMIN] AIX - Out of M=
emory=0A> To: "Tom Lane" <tgl@sss.pgh.pa.us>=0A> Cc: "Thorne, Francis" <tho=
rnef@cromwell.co.uk>, pgsql-admin@postgresql.org=0A> Date: Monday, February=
 15, 2010, 11:18 AM=0A> On Mon, Feb 15, 2010 at 10:57:06AM=0A> -0500, Tom L=
ane wrote:=0A> > "Thorne, Francis" <thornef@cromwell.co.uk>=0A> writes:=0A>=
 > > Looking for some help with regards to an 'Out of=0A> Memory' issue I h=
ave=0A> > > with our Postgresql install on AIX.=A0 When=0A> running large u=
pdates or=0A> > > select queries we get an out of memory error=0A> returned=
 and details=0A> > > entered in the log file like below.=A0 This is=0A> a 6=
4-bit install and I have=0A> > > set the ulimit for the postgres user to=0A=
> unlimited.=A0 =0A> > =0A> > The bloat seems to be here:=0A> > =0A> > >=A0=
 =A0=A0=A0AfterTriggerEvents:=0A> 131063808 total in 26 blocks; 576 free (7=
=0A> > > chunks); 131063232 used=0A> > =0A> > but it's hard to believe you'=
d be getting "out of=0A> memory" after only=0A> > 130MB in a 64-bit build.=
=A0 Are you *sure* the=0A> postgres executable is=0A> > 64-bit?=A0 Are you =
*sure* the postmaster has been=0A> launched with=0A> > nonrestrictive ulimi=
t?=A0 On lots of setups that=0A> takes modifying the=0A> > PG startup scrip=
t, not just fooling with some user's=0A> .profile.=0A> > =0A> > > This is a=
 64-bit install (8.3) on AIX 5.3=0A> > =0A> > 8.3.what?=0A> > =0A> > =A0=A0=
=A0 =A0=A0=A0=0A> =A0=A0=A0 regards, tom lane=0A> =0A> I no longer have an =
AIX box, but I had similar problems=0A> with other=0A> applications that ne=
eded large amounts of memory. Some OS=0A> specific=0A> steps needed to be t=
aken to allow normal users to allocate=0A> large=0A> blocks of memory. The =
information needed was in their=0A> on-line docs=0A> as I recall, but I do =
not remember the details. The=0A> executables may=0A> need to be built with=
 specific options/flags to work.=0A> =0A> Regards,=0A> Ken=0A> =0A=0AKen,=
=0A=0AI recently saw a similar issue.  It is two-fold:=0A=0A1.  I used "su =
-" to become the postgres user, and inherited the previous account's memory=
 limits,=0A2.  AfterTriggerEvents queues are caused by foreign key constrai=
nts, one per row.  If you're loading data, dropping or disabling that const=
raint makes a world of difference.  Just be sure to check afterwards if the=
 RI has been violated prior to recreating the FK constraint.=0A=0ABob=0A=0A=
=0A