Received: from localhost (unknown [200.46.204.191]) by postgresql.org (Postfix) with ESMTP id 95BA69F9862 for ; Thu, 20 Sep 2007 14:28:44 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.191]) (amavisd-maia, port 10024) with ESMTP id 20050-08 for ; Thu, 20 Sep 2007 14:28:40 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from web30009.mail.mud.yahoo.com (web30009.mail.mud.yahoo.com [209.191.69.26]) by postgresql.org (Postfix) with SMTP id DB8639F9B90 for ; Thu, 20 Sep 2007 14:28:39 -0300 (ADT) Received: (qmail 77591 invoked by uid 60001); 20 Sep 2007 17:28:38 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com.ar; h=X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:MIME-Version:Content-Type:Message-ID; b=Su2I+Qg1cEBaRecuNL7nMgVVUwTTP15poss2UZBeoF90XFboXAYUzn2WN89jLmKUq+3nJkEa8yEG61l0F8OShpYdHm6NYuvR8PM8hImwI8TAA0sNjODpz1xJG1Fo+IigYOU9jKRGzEaTYDBnBcaAwsVUTWzgpyKdV3hAEtOP3Vg=; X-YMail-OSG: 2mc5DqwVM1lVZWvl6dDoTGqXqSfKvzKVC03hj.Sofy6LyhTZCOHnGN8.RXWrHrVUMNZECrNiPDAInPH8CMR_CeNhEyuRhbfAMzjLytaW2BZdPfiZ.OOQIqhWFjeTfA-- Received: from [68.142.201.143] by web30009.mail.mud.yahoo.com via HTTP; Thu, 20 Sep 2007 10:28:37 PDT X-Mailer: YahooMailRC/651.50 YahooMailWebService/0.7.134 Date: Thu, 20 Sep 2007 10:28:37 -0700 (PDT) From: brauagustin-susc@yahoo.com.ar Subject: Re: Low CPU Usage To: Jean-David Beyer , pgsql-performance@postgresql.org MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="0-1016616190-1190309317=:67716" Message-ID: <383851.67716.qm@web30009.mail.mud.yahoo.com> X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200709/332 X-Sequence-Number: 26877 --0-1016616190-1190309317=:67716 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable My new server postgresql.conf is equal to the old one. I'm doubting this is= a hardware issue.=0AGoogling with my hard HP Proliant ML 110 G3 I saw that= IHC7 controller has some problems, but looking and testing with hdparm it = looks ok.=0Ahdparm -tT /dev/sda=E7=0ATiming cached reads: 1722 MB in 2.00 s= econds =3D 860.38 MB/sec=0ATiming buffered disks reads: 164 MB in 3.01 seco= nds =3D 54.53 MB/sec=0A=0ADoing hdparm -I /dev/sda=0ADMA has * in udma5=0AW= hich other test can I do to find if this is a hardware, kernel o postgres i= ssue?=0A=0ARegards=0AAgustin=0A=0A----- Mensaje original ----=0ADe: Jean-Da= vid Beyer =0APara: pgsql-performance@postgresql.org= =0AEnviado: jueves 20 de septiembre de 2007, 9:31:36=0AAsunto: Re: [PERFORM= ] Low CPU Usage=0A=0Abrauagustin-susc@yahoo.com.ar wrote:=0A> Hola Beto.=0A= > I have no idea where to look for that configuration or settings.=0A=0AIn = postgreSQL, the main settings are in .../pgsql/data/postgresql.conf=0A=0A> = Yesterday I red about some drivers problems with SATA disk working=0A> togh= eter with IDE devices with DMA.=0A> =0A> Mi server server is a Pentium VI 3= .3 with hyper threading (enabled in=0A> BIOS), HP Proliant ML 110.=0A> =0A>= Then I entered to the BIOS and saw in IDE Configuration:=0A> ATA/IDE Con= figuration [Enhanced]=0A> Configure SATA as = [IDE] =3D> it has RAID=0A> option too=0A> =0A> I have any = idea how to continue!!! I don't know if this a SATA problem,=0A> a configur= ation problem or what else. I have installed several servers=0A> beggining = with postgres 6.4 and I've neved had this kind of problems=0A> (always with= IDE disks). I think this is a problem with SATA disk i/o,=0A> but I don't = see how to measure that (I have already set postgresql.conf).=0A=0AAre you = sure you are really having a problem with insufficient CPU time=0Abeing dev= oted to your program(s)? When I run postgreSQL and do the initial=0Apopulat= ing of my database, which takes several hours due to the nature of=0Athe in= put data, it runs just 25% to 50% of one CPU, even though I have two=0A3.06= GHz hyperthreaded Xeon processors and six 10,000 rpm Ultra/320 SCSI=0Ahard= drives on two SCSI controllers. If I look at the results of the Linux=0Ato= p command, and iostat and vmstat, I see that I am in io-wait state 100% of= =0Athe time. The transfer rate to the hard drives averages about 2=0AMegaby= tes/second even though I have seen 90 Megabytes/second at times (when=0Adoi= ng a database restore). So the IO system can be quite fast when it is not= =0Awaiting (for seeks, no doubt). If the postgreSQL processes wanted more C= PU=0Atime, they could have it as the machine does not do much else most of = the=0Atime. Actually, it runs a four BOINC processes, but they run at nice = level=0A19, so they run only if no other process wants processing time. Whe= n I do a=0Adatabase backup, it will run more than 100% of a CPU (remember I= have two or=0Afour processors, depending on how you count them) for extend= ed periods, so=0Athe OS is certainly capable of supplying CPU power when I = need it. And=0ApostgreSQL runs multiple processes at once, so in theory, th= ey could gert=0A400% of a processor if they needed it. They do not seem to = need to do this=0Afor me.=0A> =0A> Regards=0A> Agustin=0A> =0A> =0A> ----- = Mensaje original ----=0A> De: Norberto Meijome =0A> P= ara: brauagustin-susc@yahoo.com.ar=0A> CC: pgsql-performance@postgresql.org= =0A> Enviado: jueves 20 de septiembre de 2007, 7:53:05=0A> Asunto: Re: [PER= FORM] Low CPU Usage=0A> =0A> On Wed, 19 Sep 2007 12:13:33 -0700 (PDT)=0A> b= rauagustin-susc@yahoo.com.ar wrote:=0A> =0A>> max_stack_depth =3D 7MB #in t= he old server is 8MB but if I set in here=0A> give me the ulimit error=0A> = =0A> Hola Agustin :)=0A> otro argentino en el extranjero x aca ;)=0A> =0A> = anyway, back to English ;)=0A> =0A> a long shot but...=0A> =0A> check if yo= u have any limits set on the host for CPU usage... you may be=0A> limited t= o x number of secs / % by the OS scheduler. When you query your=0A> CPU,=0A= > it will say u are only using 5% or so...=0A> =0A=0A=0A-- =0A .~. Jean-D= avid Beyer Registered Linux User 85642.=0A /V\ PGP-Key: 9A2FC99A= Registered Machine 241939.=0A /( )\ Shrewsbury, New Jersey ht= tp://counter.li.org=0A ^^-^^ 08:15:01 up 6 days, 42 min, 1 user, load avera= ge: 4.24, 4.25, 4.14=0A=0A---------------------------(end of broadcast)----= -----------------------=0ATIP 3: Have you checked our extensive FAQ?=0A=0A = http://www.postgresql.org/docs/faq=0A=0A=0A=0A=0A=0A=0A=0A = Segu=ED de cerca a la Selecci=F3n Argentina de Rugby en el Mundial de Fr= ancia 2007.=0Ahttp://ar.sports.yahoo.com/mundialderugby --0-1016616190-1190309317=:67716 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
My new server postgresql.conf is equal to the old o= ne. I'm doubting this is a hardware issue.
Googling with my hard HP Prol= iant ML 110 G3 I saw that IHC7 controller has some problems, but looking an= d testing with hdparm it looks ok.
hdparm -tT /dev/sda=E7
Timing cach= ed reads: 1722 MB in 2.00 seconds =3D 860.38 MB/sec
Timing buffered disk= s reads: 164 MB in 3.01 seconds =3D 54.53 MB/sec

Doing hdparm -I /de= v/sda
DMA has * in udma5
Which other test can I do to find if this is= a hardware, kernel o postgres issue?

Regards
Agustin

----- Mensaje original ----
De: Jean-David Beyer <jeandavid8@verizon.net>
Para: pgsql-performance@postgresql.orgEnviado: jueves 20 de septiembre de 2007, 9:31:36
Asunto: Re: [PERFORM= ] Low CPU Usage

brauagustin-susc@yahoo.com.ar wrote:
> Ho= la Beto.
> I have no idea where to look for that configuration or set= tings.

In postgreSQL, the main settings are in .../pgsql/data/postgr= esql.conf

> Yesterday I red about some drivers problems with SATA= disk working
> togheter with IDE devices with DMA.
>
> = Mi server server is a Pentium VI 3.3 with hyper threading (enabled in
&g= t; BIOS), HP Proliant ML 110.
>
> Then I entered to the BIOS a= nd saw in IDE Configuration:
>   ATA/IDE Configuration = ;            &n= bsp;      [Enhanced]
>   = ;        Configure SATA as            =        [IDE]  =3D> it has RAID> option too
>
> I have any idea how to continue!!! I don= 't know if this a SATA problem,
> a configuration problem or what els= e. I have installed several servers
> beggining with postgres 6.4 and= I've neved had this kind of problems
> (always with IDE disks). I th= ink this is a problem with SATA disk i/o,
> but I don't see how to me= asure that (I have already set postgresql.conf).

Are you sure you ar= e really having a problem with insufficient CPU time
being devoted to yo= ur program(s)? When I run postgreSQL and do the initial
populating of my= database, which takes several hours due to the nature of
the input data= , it runs just 25% to 50% of one CPU, even though I have two
3.06 GHz hy= perthreaded Xeon processors and six 10,000 rpm Ultra/320 SCSI
hard drives on two SCSI controllers. If I look at the results of the Linux
t= op command, and iostat and vmstat, I see that I am in io-wait state 100% of=
the time. The transfer rate to the hard drives averages about 2
Mega= bytes/second even though I have seen 90 Megabytes/second at times (when
= doing a database restore). So the IO system can be quite fast when it is no= t
waiting (for seeks, no doubt). If the postgreSQL processes wanted more= CPU
time, they could have it as the machine does not do much else most = of the
time. Actually, it runs a four BOINC processes, but they run at n= ice level
19, so they run only if no other process wants processing time= . When I do a
database backup, it will run more than 100% of a CPU (reme= mber I have two or
four processors, depending on how you count them) for= extended periods, so
the OS is certainly capable of supplying CPU power= when I need it. And
postgreSQL runs multiple processes at once, so in theory, they could gert
400% of a processor if they needed it. Th= ey do not seem to need to do this
for me.
>
> Regards
&g= t; Agustin
>
>
> ----- Mensaje original ----
> De= : Norberto Meijome <beto@octantis.com.au>
> Para: brauagustin-s= usc@yahoo.com.ar
> CC: pgsql-performance@postgresql.org
> Envia= do: jueves 20 de septiembre de 2007, 7:53:05
> Asunto: Re: [PERFORM] = Low CPU Usage
>
> On Wed, 19 Sep 2007 12:13:33 -0700 (PDT)
= > brauagustin-susc@yahoo.com.ar wrote:
>
>> max_stack_de= pth =3D 7MB #in the old server is 8MB but if I set in here
> give me = the ulimit error
>
> Hola Agustin :)
> otro argentino en= el extranjero x aca ;)
>
> anyway, back to English ;)
>=
> a long shot but...
>
> check if you have any limits = set on the host for CPU usage... you may be
> limited to x number of secs / % by the OS scheduler. When you query your
> CPU,> it will say u are only using 5% or so...
>


--
=   .~.  Jean-David Beyer     &n= bsp;    Registered Linux User 85642.
  /V\=   PGP-Key: 9A2FC99A       &nbs= p; Registered Machine   241939.
/( )\ Shrewsbury, New Jersey&= nbsp;   = http://counter.li.org
^^-^^ 08:15:01 up 6 days, 42 min, 1 user, loa= d average: 4.24, 4.25, 4.14

---------------------------(end of broad= cast)---------------------------
TIP 3: Have you checked our extensive F= AQ?

          &nbs= p;    http://www.postgresql.org/docs= /faq


=0A=0A=0A=0A
=
Los referentes m=E1s importantes en = compra/venta de autos se juntaron:
Demotores y Yahoo!.=0AAhora comprar o= vender tu auto es m=E1s f=E1cil.
Visit=E1 http://ar.autos.yahoo.com/<= /font> --0-1016616190-1190309317=:67716--