Re: Memory Allocation

[Tom Duerbusch]

It really depends on how easy/hard is it, in your shop to change the size of 
the LPAR.
 
When it is easy, then I make the memory allocation fairly small, say 2 GB and 
grow it when it is proved that I need more memory.  However, I wouldn't ever 
try Websphere with anything under 4 GB.  This is under the theory that the 
first application takes all the memory it can and will leave nothing for the 
next application.
 
I have a 4 GB LPAR for z/VM 5.2 and Linux machines.
What I currently have running:
 
I have 18 Linux images running
 
1  production Oracle
1  test Oracle
1  development Oracle
1  DB2 Connect production
1  DB2 Connect test
4  Samba servers   (SLES 8)
2  NFS servers
2  FTP servers (that have the iso images of Linux and Oracle mounted for 
installation purposes)
3  VSE Virtual Tape servers
2  Samba Servers (SLES 11) that will be migrating to once testing is completed  
(these will also take over the VSE LANRES disk hosting function we are 
currently using)
 
None of these are heavy hitters.
Production Oracle has 200 users defined and 40 currently logged on.
DB2 Connect production has 22 clients at this time.
 
The VM system only pages when an image is bought up.  
All the linux images use vdisk for swap space.
 
The production DB2 Connect image:
Defined as 250 MB virtual.
Defined on a single 3390-9.
Has been running for 162 days;
linux69:~ # swapon -s
FilenameTypeSizeUsedPriority
/dev/dasdb1 partition   162468  271250
/dev/dasdc1 partition   81228   45224   60
Most of the swap space used was from me going into Yast.
 
And as VMSTAT shows, we are not actively swapping:
 
linux69:~ # vmstat 10 10
procs ---memory-- ---swap-- -io -system-- -cpu--
 r  b   swpd   free   buff  cache   si   sobibo   in   cs us sy id wa st
 1  0  47936   7680  22460 12444000 21510  4  1 94  0  1
 0  0  47936   7560  22468 12443200 014 2631  165  4  1 70  0 25
 0  0  47936   7560  22476 12442400 0 3 3130  179  4  1 80  0 15
 0  0  47936   8092  22432 12446800 018  351  213  4  1 95  0  0
 0  0  47936   8032  22448 12445200 014  328  203  3  1 95  0  0
 0  0  47936   8032  22456 1200 012 5486  188  0  1 99  0  0
 0  0  47936   8032  22472 12442800 026 6104  189  4  1 89  0  6
 0  0  47936   8032  22496 12440400 014 6267  189  4  1 88  0  7
 0  0  47936   7972  22520 12438000 014 4529  194  4  1 86  0  9
 0  0  47936   7972  22536 12436400 012  996  202  4  1 95  0  1

So, your initial plans for a z/VM plus a single DB2 Connect Server could be run 
in a LPAR of 512 MB.
However, since SUSE (with OSA connections) seems to want about 750 MB for 
installation, that would cause a lot of paging when you do a Linux install.  On 
a MP3000 with 1 GB and 9 VSE systems the 750 install requirement did cause a 
lot of paging but everything still ran fairly well.  
 
My current processor is a z/890 with a ficon attached DS6800.  So, yes, I do 
trade the very cheap ficon I/Os for the more expensive memory.  If we were 
driving the DS6800 harder, then I would add more real memory to the mix to 
drive down the I/O rate (paging/swaping reduction along with more memory for 
caching).
 
So the question goes back toWhat else are you planning (or playing) for in 
the future and how often can you reconfigure the LPAR to add more memory?
 
Tom Duerbusch
THD Consulting
 

>>> "Dazzo, Matt"  12/2/2010 8:45 AM >>>
We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb total 
memory. I am trying to decided what to allocate to zvm/linux as a starting 
point. Our initial thoughts and a recommendation from our VAR was 6gb. The 
first application will be DB2 Conn Server and not sure what's after that. I'd 
like to find out how much memory other shops have allocated and what 
applications they support. Is our initial 6gb a decent starting point? I know 
the IBM standard answer 'it depends' but I am looking for some clarity and 
guide lines.

Thanks
Matt

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
---

Re: Memory Allocation

[Mark Post]

>>> On 12/2/2010 at 09:45 AM, "Dazzo, Matt"  wrote: 
> We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb total 
> memory. I am trying to decided what to allocate to zvm/linux as a starting 
> point. Our initial thoughts and a recommendation from our VAR was 6gb. The 
> first application will be DB2 Conn Server and not sure what's after that. I'd 
> like to find out how much memory other shops have allocated and what 
> applications they support. Is our initial 6gb a decent starting point? I know 
> the IBM standard answer 'it depends' but I am looking for some clarity and 
> guide lines.

Given that you don't have a significant existing workload, and that you're 
trying to add more real storage to your other LPAR, then 6GB for z/VM is a 
decent starting point.  Reserve 1.5-2GB of that for expanded storage, since 
z/VM still uses that quite a bit as part of its overall virtual storage 
management strategy.  Make sure that you get a good z/VM and Linux performance 
monitor.  Even though 6GB is a decent starting point, when you start putting 
workload into that LPAR, you need to be able to measure the results and stay 
ahead of the curve.  One good sized application guest with bad application 
developer choices behind it can chew up a lot of real memory pretty quickly.


Mark Post

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Memory Allocation

[van Sleeuwen, Berry]

Oops, I'm sorry. I was thinking about the guest.

We have multiple VM's, each has 12G, 10G central and 2G Xstore. And they
also have about 30G pagedisks. We have about 40 guests on two LPARs, in
various configurations. One VM has about 25% page in use, with a lot of
very small mostly idle guests, this VM has a 1.8:1 virtual to real
ratio. We used to run in a 2.5G LPAR with a dozen guests. That proved to
be a bit too small when we got more guests, especially when we got a
very large and very active oracle guest.

We have various applications running, a few large oracle, DB2, some CCL,
web and samba. It is a nice mix of very active guests and mostly idle
guests.

A DB2connect only requires a small amount, so the big questions is what
you expect to get next. Both in sizes and how active they will be. If
your guests are very active you could require more storage than if they
are mostly idle. A database can be very active, a web or samba could
very well be mostly idle.

Regards, Berry.




-Original Message-
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
Dazzo, Matt
Sent: donderdag 2 december 2010 15:46
To: LINUX-390@VM.MARIST.EDU
Subject: Memory Allocation

We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb
total memory. I am trying to decided what to allocate to zvm/linux as a
starting point. Our initial thoughts and a recommendation from our VAR
was 6gb. The first application will be DB2 Conn Server and not sure
what's after that. I'd like to find out how much memory other shops have
allocated and what applications they support. Is our initial 6gb a
decent starting point? I know the IBM standard answer 'it depends' but I
am looking for some clarity and guide lines.

Thanks
Matt

--
For LINUX-390 subscribe / signoff / archive access instructions, send
email to lists...@vm.marist.edu with the message: INFO LINUX-390 or
visit http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/
ÿþDit bericht is vertrouwelijk en kan 
geheime informatie bevatten enkel

bestemd voor de geadresseerde. Indien 
dit bericht niet voor u is bestemd,

verzoeken wij u dit onmiddellijk aan 
ons te melden en het bericht te

vernietigen.

Aangezien de integriteit van het 
bericht niet veilig gesteld is middels

verzending via internet, kan Atos 
Origin niet aansprakelijk worden 
gehouden

voor de inhoud daarvan.

Hoewel wij ons inspannen een virusvrij 
netwerk te hanteren, geven

wij geen enkele garantie dat dit 
bericht virusvrij is, noch aanvaarden 
wij

enige aansprakelijkheid voor de 
mogelijke aanwezigheid van een virus in 
dit

bericht.

 

Op al onze rechtsverhoudingen, 
aanbiedingen en overeenkomsten 
waaronder

Atos Origin goederen en/of diensten 
levert zijn met uitsluiting van alle

andere voorwaarden de 
Leveringsvoorwaarden van Atos Origin 
van toepassing.

Deze worden u op aanvraag direct 
kosteloos toegezonden.

 

This e-mail and the documents attached 
are confidential and intended solely

for the addressee; it may also be 
privileged. If you receive this e-mail

in error, please notify the sender 
immediately and destroy it.

As its integrity cannot be secured on 
the Internet,

Re: Memory Allocation

[van Sleeuwen, Berry]

Hi Matt,

You should allocate what is needed. So measure what your application
needs. A rule of thumb is hard to give since, indeed, it depends. Our
DB2connect is sized at 300M. And from measuring the guest we can see
that we could size it down to even 250M. We still don't size down since
it is still in test. After our stresstest we can decide if 300M is a
good value or that is needs to be adjusted, either more or less storage.
That would depend on the number of concurrent connections, to name one.
Obviously when you are running a full database 300M can be too small.
But then again, measure what your application needs. Our webserver is
sized at 200M, we have guests running as small as 130M.

Ask how they came up with 6G. Was it the recommendation from the books?
Was it due to the prior observation that with 6G the linux machine
didn't use swap? And did they try the same with 5G or 4G? Do they know
how much the application is actually using?

What are your intentions beyond this guest? Do you want to add more
guests? Then try to start with small machines. IMHO it is very hard to
size down afterwards since nobody wants to size down, they only want to
size up. Also, don't mix different workloads into one machine. In VM it
is very easy to add a new server instead of consolidating into one
guest. And your performance will benefit since different workloads
usually require different settings, both for the guests and the
application.

Regards, berry.

-Original Message-
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
Dazzo, Matt
Sent: donderdag 2 december 2010 15:46
To: LINUX-390@VM.MARIST.EDU
Subject: Memory Allocation

We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb
total memory. I am trying to decided what to allocate to zvm/linux as a
starting point. Our initial thoughts and a recommendation from our VAR
was 6gb. The first application will be DB2 Conn Server and not sure
what's after that. I'd like to find out how much memory other shops have
allocated and what applications they support. Is our initial 6gb a
decent starting point? I know the IBM standard answer 'it depends' but I
am looking for some clarity and guide lines.

Thanks
Matt

--
For LINUX-390 subscribe / signoff / archive access instructions, send
email to lists...@vm.marist.edu with the message: INFO LINUX-390 or
visit http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/
ÿþDit bericht is vertrouwelijk en kan 
geheime informatie bevatten enkel

bestemd voor de geadresseerde. Indien 
dit bericht niet voor u is bestemd,

verzoeken wij u dit onmiddellijk aan 
ons te melden en het bericht te

vernietigen.

Aangezien de integriteit van het 
bericht niet veilig gesteld is middels

verzending via internet, kan Atos 
Origin niet aansprakelijk worden 
gehouden

voor de inhoud daarvan.

Hoewel wij ons inspannen een virusvrij 
netwerk te hanteren, geven

wij geen enkele garantie dat dit 
bericht virusvrij is, noch aanvaarden 
wij

enige aansprakelijkheid voor de 
mogelijke aanwezigheid van een virus in 
dit

bericht.

 

Op al onze rechtsverhoudingen, 
aanbiedingen en overeenkomsten 
waaronder

Atos Origin goederen en/of diensten 
levert zijn met uitsluiting van alle

andere voorwaarden de 
Leveringsvoorwaarden van Atos Origin 
van toepassing.

Deze worden u op aanvraag direct 
kosteloos toegezonden.

 

This e-mail and the documents a

Re: Memory Allocation

[Dazzo, Matt]

 I should have clarified that the 16gb is for 3 lpars not just zvm. We would 
like to increase our production lpar but we are trying to get an idea of a 
starting point for zvm/linux. So that's where the 6gb came from, if we used 6gb 
for zvm we could roughly add an additional 2gb to the production lpar.  tks

-Original Message-
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of Scott 
Rohling
Sent: Thursday, December 02, 2010 10:47 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: Memory Allocation

I would ask why you don't allocate it all?   Say 14G of cstor and 2G of
xstor?Are you planning for a 2nd z/VM LPAR?

6GB would likely be very sufficient for z/VM and a single Linux DB2 Conn
Server guest.   I'll venture that much if it helps :-)   And probably
without even paging..   What it mostly depends on is the virtual memory size
of your guests.   As long as you have sufficient paging space - you can
overcommit your memory.   2 or 3:1 is likely as far as you want to go unless
you're willing to suffer performance degradation.   That is - if you have 3
Linux guests using 4G of memory each -- you would be overcommited  at a 2:1
(V:R) ratio.   It also means that since you are overcommitted by 6GB -
you'll want 12GB of paging space  (6 3390-3).   You want to have twice as
much paging space because you want paging to stay at 50% or less.The
paging requirements are based on 'worst case', though -- meaning that all 3
guests are using ALL of their memory and are all actively paging.

Scott Rohling

On Thu, Dec 2, 2010 at 7:45 AM, Dazzo, Matt  wrote:

> We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb total
> memory. I am trying to decided what to allocate to zvm/linux as a starting
> point. Our initial thoughts and a recommendation from our VAR was 6gb. The
> first application will be DB2 Conn Server and not sure what's after that.
> I'd like to find out how much memory other shops have allocated and what
> applications they support. Is our initial 6gb a decent starting point? I
> know the IBM standard answer 'it depends' but I am looking for some clarity
> and guide lines.
>
> Thanks
> Matt
>
> --
> For LINUX-390 subscribe / signoff / archive access instructions,
> send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or
> visit
> http://www.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Memory Allocation

[Scott Rohling]

I would ask why you don't allocate it all?   Say 14G of cstor and 2G of
xstor?Are you planning for a 2nd z/VM LPAR?

6GB would likely be very sufficient for z/VM and a single Linux DB2 Conn
Server guest.   I'll venture that much if it helps :-)   And probably
without even paging..   What it mostly depends on is the virtual memory size
of your guests.   As long as you have sufficient paging space - you can
overcommit your memory.   2 or 3:1 is likely as far as you want to go unless
you're willing to suffer performance degradation.   That is - if you have 3
Linux guests using 4G of memory each -- you would be overcommited  at a 2:1
(V:R) ratio.   It also means that since you are overcommitted by 6GB -
you'll want 12GB of paging space  (6 3390-3).   You want to have twice as
much paging space because you want paging to stay at 50% or less.The
paging requirements are based on 'worst case', though -- meaning that all 3
guests are using ALL of their memory and are all actively paging.

Scott Rohling

On Thu, Dec 2, 2010 at 7:45 AM, Dazzo, Matt  wrote:

> We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb total
> memory. I am trying to decided what to allocate to zvm/linux as a starting
> point. Our initial thoughts and a recommendation from our VAR was 6gb. The
> first application will be DB2 Conn Server and not sure what's after that.
> I'd like to find out how much memory other shops have allocated and what
> applications they support. Is our initial 6gb a decent starting point? I
> know the IBM standard answer 'it depends' but I am looking for some clarity
> and guide lines.
>
> Thanks
> Matt
>
> --
> For LINUX-390 subscribe / signoff / archive access instructions,
> send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or
> visit
> http://www.marist.edu/htbin/wlvindex?LINUX-390
> --
> For more information on Linux on System z, visit
> http://wiki.linuxvm.org/
>

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Memory Allocation

[Dazzo, Matt]

We are entering the world of zvm/linux with a z10bc-2098 n04 and 16gb total 
memory. I am trying to decided what to allocate to zvm/linux as a starting 
point. Our initial thoughts and a recommendation from our VAR was 6gb. The 
first application will be DB2 Conn Server and not sure what's after that. I'd 
like to find out how much memory other shops have allocated and what 
applications they support. Is our initial 6gb a decent starting point? I know 
the IBM standard answer 'it depends' but I am looking for some clarity and 
guide lines.

Thanks
Matt

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
--
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Re: Linux and buffers and memory allocation

[James Tison]

I used to fight this one all the time too: I was convinced that the harm
induced by paging-related VM I/O waits was killing machine performance.
It was, before I discovered VDISK units for paging. I was just about
always in Q3 waiting I/O response from one of my (then-) real paging
minidisks, even under conditions of very light memory & load stress.
I changed my paging units to VDISK, and now whenever I find myself in Q3,
it's for a better reason (waiting on TCP/IP, or even real disk I/O).

Someone posted a link to a PDF document produced by Red Hat with sysctl
tweaks for RHEL 3, which (despite not being even 50% relevant for me)
I'll admit was a very valuable discussion that shed some more light
on Linux paging & file caching for me, at least. But under VM, it doesn't
matter as much to me as it would if I were running LPAR. Here, I'd trust
the counsel of my VM guys before I'd start tweaking sysctls. Under LPAR,
I'd have no choice other than to tweak or suffer.

If you don't trust Linux to make page/fs cache decisions, try letting
VM do it through VDISK. I'd try this before I start risking the stability
of my system with radical sysctl tweaks. When I did it, I went all the
way:
I now have 1.1 GB page space on VDISK and zero on 3390-x, which backs 768
MB main memory. Current average load pressure says I'd be doing very
little
paging at about 1 GB main; but every now and again I get a runaway process

that changes the phrase "memory stress" to "paging frenzy", and the
machine
still recovers from frenzy-induced runqueue loads as high as 34.0 as long
as
the memory pig is found and slain before too much time goes by. My load
(interactive developers) sounds very different from yours; but if paging-
related waits are your problem, try VDISK. It might help you out, if your
VM
guys will let you use it. I love VM :-)

Peace,
--Jim--
James S. Tison
Senior Software Engineer
TPF Laboratory / Architecture
IBM Corporation
"If dogs don't go to heaven, then, when I die, I want to go where they
do."
-- Will Rogers

Linux on 390 Port <[EMAIL PROTECTED]> wrote on 04/26/2004 09:30:31:

> Current environment has 2 GB central, 0 expanded for VM (again, because
the
> decision was made after our last maint window or we would have moved
some
> memory to expanded). The VM guest in question  actually did NOT have 1.3
GB
> real, as I was told, but in fact 768 MB. The VM guy (who has about 2
weeks
> worth of experience with VM now) had the 1.3 GB on a diff guest.
> The Linux Guest being used for WebSphere also has a 843 MB physical disk
> swap volume. It is the Linux guest that is writing pages to the swap
> volume, not VM.  Our VM is not paging hardly at all, except a smidgin at
> guest IPL. Then it gives it back.

> Anyway, I'd like to see the amount of memory being used for buffers
> reduced, since this system doesn't have much except the Java code and
the
> WebSphere configuration/deployment stuff being accessed from local disk.
> All of the data for the applications is sitting on DB2 being accessed by
> the DB2 Connect Linux Client.
>


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Linux and buffers and memory allocation

[James Melin]

Alas, no. No expanded. Management approved the trial of VM etc 3 days AFTER
our IPL in which we could have created some expanded stor for that LPAR.

At the end of may on our next 2-month maintenance window, we are going to
be able to move VM to the z-900 under 2 IFL's (trial hardware provided by
IBM for our POC) and depending, put 2.5-3 GB of memory in the VM LPAR. Now,
I was thinking that we should allocate (assuming we get the 3 gb) 2,5 GB
real and 500 meg expanded.  If it's just the 2.5, we would want to go with
2 GB real and 500 meg expanded, from what I've been able to gather.

As to the numbers being confusing, I think I am mixing my terminology. Let
me clarify.

Current environment has 2 GB central, 0 expanded for VM (again, because the
decision was made after our last maint window or we would have moved some
memory to expanded). The VM guest in question  actually did NOT have 1.3 GB
real, as I was told, but in fact 768 MB. The VM guy (who has about 2 weeks
worth of experience with VM now) had the 1.3 GB on a diff guest.
The Linux Guest being used for WebSphere also has a 843 MB physical disk
swap volume. It is the Linux guest that is writing pages to the swap
volume, not VM.  Our VM is not paging hardly at all, except a smidgin at
guest IPL. Then it gives it back.

the current show of /proc/meminfo indicates no swapping going on (I IPL'ed
the guest Friday before I left) but that's because no testing is going on.
Previously, I've seen SwapCached go to 13300.
When its doing that, free memory is dropping to around 47000 or so.

Anyway, I'd like to see the amount of memory being used for buffers
reduced, since this system doesn't have much except the Java code and the
WebSphere configuration/deployment stuff being accessed from local disk.
All of the data for the applications is sitting on DB2 being accessed by
the DB2 Connect Linux Client.

itasca:/proc/sys # cat /proc/meminfo
total:used:free:  shared: buffers:  cached:
Mem:  792133632 789549056  25845760 50958336 33734656
Swap: 8846295040 884629504
MemTotal:   773568 kB
MemFree:  2524 kB
MemShared:   0 kB
Buffers: 49764 kB
Cached:  32944 kB
SwapCached:  0 kB
Active:  26420 kB
Inactive:56344 kB
HighTotal:   0 kB
HighFree:0 kB
LowTotal:   773568 kB
LowFree:  2524 kB
SwapTotal:  863896 kB
SwapFree:   863896 kB
BigFree: 0 kB
itasca:/proc/sys #






 Rob van der Heij
 <[EMAIL PROTECTED]>
 Sent by: Linux on  To
 390 Port  [EMAIL PROTECTED]
 <[EMAIL PROTECTED]  cc
 IST.EDU>
   Subject
   Re: Linux and buffers and memory
 04/23/2004 04:17  allocation
 PM


 Please respond to
 Linux on 390 Port
 <[EMAIL PROTECTED]
 IST.EDU>






James Melin wrote:

>I have one guest with 1.3 GB real (for Webspher ND) and it tends to page
>about 90-113 KB when doing a large application deploy but still indicating
>150-200 MB free and 500 MB of buffers. I've heard the less is more
>approach, but has anyone got Websphere Network Deployment running on
Linux?
>
>
Your numbers are confusing. You mean the virtual machine size is 1.3 GB,
or is it even bigger and do you find 1.3 GB on average resident? If your
VM system is unable to keep enough of the virtual machine in, then you
should make it smaller. When your VM system is larger than 2 GB and you
have more of these Linux virtual machines, then contention under the 2
GB bar may cause the paging. The 500 MB excess memory in WebSphere makes
that likely.  Does VM have expanded storage for paging?  That should
help you overcome some of the contention under the bar. What performance
monitor do you have?

How about mini disk cache. Contrasting what others may recommend, I'd be
interested to see what happens when you enable MDC in storage (at least
for  the disks that WebSphere has) after you reduce the virtual machine
with 500 MB. And disable MDC in expanded by setting it to max 0M.

Rob

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or
visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Re: Linux and buffers and memory allocation

[Rob van der Heij]

James Melin wrote:

I have one guest with 1.3 GB real (for Webspher ND) and it tends to page
about 90-113 KB when doing a large application deploy but still indicating
150-200 MB free and 500 MB of buffers. I've heard the less is more
approach, but has anyone got Websphere Network Deployment running on Linux?

Your numbers are confusing. You mean the virtual machine size is 1.3 GB,
or is it even bigger and do you find 1.3 GB on average resident? If your
VM system is unable to keep enough of the virtual machine in, then you
should make it smaller. When your VM system is larger than 2 GB and you
have more of these Linux virtual machines, then contention under the 2
GB bar may cause the paging. The 500 MB excess memory in WebSphere makes
that likely.  Does VM have expanded storage for paging?  That should
help you overcome some of the contention under the bar. What performance
monitor do you have?
How about mini disk cache. Contrasting what others may recommend, I'd be
interested to see what happens when you enable MDC in storage (at least
for  the disks that WebSphere has) after you reduce the virtual machine
with 500 MB. And disable MDC in expanded by setting it to max 0M.
Rob

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Linux and buffers and memory allocation

[James Melin]

I know that linux is buffer happy. Exceptionally so. Just got the while VM
thing going, and that went smoothly enough. Now I'm trying to tune memory
so that linux paging is minimized but VM paging does not increase as a
result.

I have one guest with 1.3 GB real (for Webspher ND) and it tends to page
about 90-113 KB when doing a large application deploy but still indicating
150-200 MB free and 500 MB of buffers. I've heard the less is more
approach, but has anyone got Websphere Network Deployment running on Linux?

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

RE : Kernel memory allocation

[Carsten Otte]

Hi Gerard!

The question is not that easy to answer. Let be give some background
first:
Linux holds all user space related data in the page cahce.  Basically you
have
two different types of pages in the cache:

- anonymous pages
These are simple. When your program does malloc(), anonymous memory will
be added
to the page cache and refered by the page table of the user process. When
under memory
pressure, anonymous pages are put on the swap device.

- non anonymous pages
These correspond with a file on disk. Application binaries and libraries
are loaded into the
page cache and refered by the page table(s) of the user process(es) that
run the application.
In addition, when a process accesses a file for reading, the file contents
are loaded into the
page cache and from there copied to the given user buffer. On writing, the
corresponding page
cache pages are marked "dirty", and they will be written back to disk
(then become clean again).
When under memory pressure, non-anonymous pages can be discarded when they
are clean
because their content can be loaded from a file anyway. *These pages will
never go to the swap
device*.

When you start up your system, your page cache will be empty (reported as
free memory). Now
your system starts doing some of the above activities which fill the page
cache. The page cache
will fill up quite quickly because only anonymous pages get discarded when
the corresponding
programs have ended running. Non-anonymous pages will remain in the page
cache.
Later on, your page cache will get full (only a little rest of free memory
is reported). Now the kernel
will always discard the least recently used pages from the cache. In case
non-anonymous (&clean)
pages are least recently used, they will just be discarded. In case of
anonymous memory they will
go to the swap device.

The result of above strategy is that your page cache will grow full soon
after system startup and
will always be about full all the time from then on. Therefore, the free
memory reported has no
indication on whether the system is under memory pressure or not.

In order to figure out if your system is under memory pressure, you need
to keep an eye on the
swap rates of your system. In case the page cache is too small to cover
the working set of your
applications, the kernel will have to move frequently accessed pages to
the swap device. The next
access to the page will force the kernel to fetch the page back from the
swap device. This behavior
is called "trashing". The kernel does do trashing when your swap device
data rates are resonably
high all the time. Therefore, the swap rate is the real indication to see
if Linux is under memory
pressure.

For your concrete problem, the next thing you need to do is watch the swap
rates for a while and
look if you see them growing higher after a while and if this corresponds
with the slowdown you
see. If this is the case, the machine needs more memory.
In case you do not see the swap rate increase significantly but you still
see the slowdown, another
idea is to look at VM to see if it is memory constrained. If this is the
case, you need to _decrease_
the size of this and other virtual machines in order to fix the problem.

This is a balancing issue.

with kind regards
Carsten Otte
--
I saw screens of green, red messages too, then came blue, shubidu
And i think to myself, what a wonderful world

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

RE : Kernel memory allocation

[Barton Robinson]

There is a way to do this.  In one of the
IBM Downloads there is some code that will respond to storage
management requests. Here's the abstract.
You could look for use of this function in appropriate products
in the future

+Collaborative memory management
+CONFIG_CMM
+  Select this option, if you want to enable the kernel interface
+  to reduce the memory size of the system. This is accomplished
+  by allocating pages of memory and put them "on hold". This only
+  makes sense for a system running under VM where the unused pages
+  will be reused by VM for other guest systems. The interface
+  allows an external monitor to balance memory of many systems.
+  Everybody who wants to run Linux under VM should select this
+  option.






>Date: Tue, 23 Mar 2004 19:07:38 +0100
>Sender: Linux on 390 Port <[EMAIL PROTECTED]>
>From: Monteleone <[EMAIL PROTECTED]>
>
>Hello Mark,
>
>The first thing is because my application runs normally when I
>have enough free memory.  The free memory decreases along the day
>before giving me a memory constraint, linux gets few pages from
>cache to serve the request before swapping.  It results a slow
>down processing.
>
>I don't understand why should i lose "memory page cache" when VM
>use MDC.
>
>
>Thanks for your help.
>
>Gerard MONTELEONE
>  Ingenieur Systeme & Reseau
>SI.TE.C Z.I du vazzio 20090 AJACCIO
>( +33495236809 =C8 +33687727032
>  www.sitec.fr
>=20
>
>-Message d'origine-
>De=A0: Linux on 390 Port [mailto:[EMAIL PROTECTED] De la part de
>Post, Mark K
>Envoy=E9=A0: mardi 23 mars 2004 17:33
>=C0=A0: [EMAIL PROTECTED]
>Objet=A0: Re: Kernel memory allocation
>
>I don't believe there is a way to do that.  I would have to think
>that an application trying to allocate virtual storage would
>force Linux to give up some of the buffers/file cache pages.
>What "memory problem" are you seeing?
>
>
>Mark Post
>
>-Original Message-
>From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
>Monteleone
>Sent: Tuesday, March 23, 2004 10:13 AM
>To: [EMAIL PROTECTED]
>Subject: Kernel memory allocation
>
>
>Hello,
>
>Is there a way to limit the number of "page cached" on a SLES8
>system without limiting the size of the guest ?  I want to
>reserve a large number of free memory pages for the application.
>Have a look to the allocation please:
>
>In the morning,
>
>   procs  memory  swap  io
>system cpu
> r  b  w   swpd   free   buff  cache   si  sobibo   incs
>us  sy  id
> 0  0  1  0 407880  17296 191360   0   0 8 4058   2
>0  98
> 1  0  0  0 410132  17320 191760   0   0 0220   305   8
>2  90
> 1  0  1  0 409560  17352 192440   0   0 6170   319   6
>2  92
> 0  0  1  0 406008  17360 192940   0   0 3170   297   7
>2  91
> 1  0  0  0 408220  17372 193504   0   0 5170   248   7
>1  92
> 0  0  1  0 407056  17384 193900   0   0 0150   311   9
>2  89
> 0  0  1  0 401012  17400 194512   0   0 0190   452  12
>2  86
> 0  0  1  0 402420  17412 194976   0   0 0170   308  12
>2  86
> 1  0  0  0 399020  17420 194548   0   0 3210   395  11
>2  86
> 0  0  0  0 396060  17424 195000   0   0 0170   403  10
>2  88
> 1  0  0  0 404200  17432 195360   0   0 0190   312   8
>2  90
> 1  0  0  0 397712  17444 195828   0   0 0130   394  10
>2  88
> 0  0  0  0 400108  17456 196248   0   0 0210   370   7
>2  91
> 3  0  1  0 400468  17532 197124   0   0 8190   332  10
>2  89
> 1  0  0  0 383604  17804 202784   0   090230   426  14
>2  84
> 0  0  1  0 388692  17836 204224   0   017230   450   9
>2  89
> 0  0  1  0 383816  17900 205768   0   017210   440  12
>3  86
> 0  0  0  0 389004  17920 206412   0   0 4230   366   8
>2  90
> 0  0  0  0 383012  17928 206900   0   0 0190   424  10
>2  88
>..
>...
>in the evening,
> 2  0  0  0 123940  23888 447436   0   0 0360   464  38
>4  58
> 0  0  0  0 130812  23920 448940   0   016320   503  41
>4  55
> 2  0  1  0 129868  23936 449484   0   0 0200   443  14
>3  84
> 6  0  0  0 108252  23956 450172   0   0 0270   524  15
>3  82
> 3  0  1  0  89744  23972 450800   0   0 0220   525  13
>3  84
> 2  0  1  0  82496  24000 451636   0   0 0240   555  34
>4  62
> 

RE : Kernel memory allocation

[Monteleone]

Hello Mark,

The first thing is because my application runs normally when I have
enough free memory. The free memory decreases along the day before
giving me a memory constraint, linux gets few pages from cache to serve
the request before swapping. It results a slow down processing.

I don't understand why should i lose "memory page cache" when VM use
MDC.


Thanks for your help.

Gerard MONTELEONE
  Ingenieur Systeme & Reseau
SI.TE.C Z.I du vazzio 20090 AJACCIO
( +33495236809 È +33687727032
  www.sitec.fr
 

-Message d'origine-
De : Linux on 390 Port [mailto:[EMAIL PROTECTED] De la part de
Post, Mark K
Envoyé : mardi 23 mars 2004 17:33
À : [EMAIL PROTECTED]
Objet : Re: Kernel memory allocation

I don't believe there is a way to do that.  I would have to think that
an
application trying to allocate virtual storage would force Linux to give
up
some of the buffers/file cache pages.  What "memory problem" are you
seeing?


Mark Post

-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Monteleone
Sent: Tuesday, March 23, 2004 10:13 AM
To: [EMAIL PROTECTED]
Subject: Kernel memory allocation


Hello,

Is there a way to limit the number of  "page cached" on a SLES8 system
without limiting the size of the guest ? I want to reserve a large
number of
free memory pages for the application. Have a look to the allocation
please:
In the morning,
   procs  memory  swap  io
system cpu
 r  b  w   swpd   free   buff  cache   si  sobibo   incs
us  sy  id
 0  0  1  0 407880  17296 191360   0   0 8 4058   2
0  98
 1  0  0  0 410132  17320 191760   0   0 0220   305   8
2  90
 1  0  1  0 409560  17352 192440   0   0 6170   319   6
2  92
 0  0  1  0 406008  17360 192940   0   0 3170   297   7
2  91
 1  0  0  0 408220  17372 193504   0   0 5170   248   7
1  92
 0  0  1  0 407056  17384 193900   0   0 0150   311   9
2  89
 0  0  1  0 401012  17400 194512   0   0 0190   452  12
2  86
 0  0  1  0 402420  17412 194976   0   0 0170   308  12
2  86
 1  0  0  0 399020  17420 194548   0   0 3210   395  11
2  86
 0  0  0  0 396060  17424 195000   0   0 0170   403  10
2  88
 1  0  0  0 404200  17432 195360   0   0 0190   312   8
2  90
 1  0  0  0 397712  17444 195828   0   0 0130   394  10
2  88
 0  0  0  0 400108  17456 196248   0   0 0210   370   7
2  91
 3  0  1  0 400468  17532 197124   0   0 8190   332  10
2  89
 1  0  0  0 383604  17804 202784   0   090230   426  14
2  84
 0  0  1  0 388692  17836 204224   0   017230   450   9
2  89
 0  0  1  0 383816  17900 205768   0   017210   440  12
3  86
 0  0  0  0 389004  17920 206412   0   0 4230   366   8
2  90
 0  0  0  0 383012  17928 206900   0   0 0190   424  10
2  88
..
...
in the evening,
 2  0  0  0 123940  23888 447436   0   0 0360   464  38
4  58
 0  0  0  0 130812  23920 448940   0   016320   503  41
4  55
 2  0  1  0 129868  23936 449484   0   0 0200   443  14
3  84
 6  0  0  0 108252  23956 450172   0   0 0270   524  15
3  82
 3  0  1  0  89744  23972 450800   0   0 0220   525  13
3  84
 2  0  1  0  82496  24000 451636   0   0 0240   555  34
4  62
 1  0  1  0  84892  24012 451432   0   0 0310   634  18
4  78
 4  0  0  0  80232  24028 452252   0   0 0220   668  16
3  81
 1  0  2  0  82200  24040 453028   0   0 0260   574  20
3  76
 0  0  1  0  81536  24052 453856   0   0 0240   621  17
3  80
 0  0  1  0  83396  24060 454448   0   0 0240   512  14
3  83
 0  0  1  0  84584  24076 454864   0   0 0200   385  11
2  87
 1  0  1  0  63404  24088 455428   0   0 0180   683  12
2  85
 1  0  0  0  50252  24108 456164   0   0 0260   590  17
3  79
 0  0  1  0  46016  24128 456820   0   0 0220   479  12
3  86
 2  0  1  0  47052  24144 456688   0   0 0270   621  16
3  81
 0  0  0  0  44228  24160 457324   0   0 0260   501  11
2  86
 3  0  1  0  47480  24172 458020   0   0 0220   506  11
3  87
 1  0  1  0  44656  24188 458660   0   0 0220   495  14
3  83
 0  0  0  0  40288  24204 459416   0   0 0240   536  15
3  82
 1  0  0  0  40036  24216 460192   0   0 0260   574  15
3  82

all free memory is kept by cache.
When memory problem occurs, only few pages are freed. What parameter
should
I set to limit the number of page cache ?

Thanks in advance.


   Gerard MONTELEONE
  Ingenieur Sys

Re: Kernel memory allocation

[Post, Mark K]

I don't believe there is a way to do that.  I would have to think that an
application trying to allocate virtual storage would force Linux to give up
some of the buffers/file cache pages.  What "memory problem" are you seeing?


Mark Post

-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Monteleone
Sent: Tuesday, March 23, 2004 10:13 AM
To: [EMAIL PROTECTED]
Subject: Kernel memory allocation


Hello,

Is there a way to limit the number of  "page cached" on a SLES8 system
without limiting the size of the guest ? I want to reserve a large number of
free memory pages for the application. Have a look to the allocation please:
In the morning,
   procs  memory  swap  io
system cpu
 r  b  w   swpd   free   buff  cache   si  sobibo   incs
us  sy  id
 0  0  1  0 407880  17296 191360   0   0 8 4058   2
0  98
 1  0  0  0 410132  17320 191760   0   0 0220   305   8
2  90
 1  0  1  0 409560  17352 192440   0   0 6170   319   6
2  92
 0  0  1  0 406008  17360 192940   0   0 3170   297   7
2  91
 1  0  0  0 408220  17372 193504   0   0 5170   248   7
1  92
 0  0  1  0 407056  17384 193900   0   0 0150   311   9
2  89
 0  0  1  0 401012  17400 194512   0   0 0190   452  12
2  86
 0  0  1  0 402420  17412 194976   0   0 0170   308  12
2  86
 1  0  0  0 399020  17420 194548   0   0 3210   395  11
2  86
 0  0  0  0 396060  17424 195000   0   0 0170   403  10
2  88
 1  0  0  0 404200  17432 195360   0   0 0190   312   8
2  90
 1  0  0  0 397712  17444 195828   0   0 0130   394  10
2  88
 0  0  0  0 400108  17456 196248   0   0 0210   370   7
2  91
 3  0  1  0 400468  17532 197124   0   0 8190   332  10
2  89
 1  0  0  0 383604  17804 202784   0   090230   426  14
2  84
 0  0  1  0 388692  17836 204224   0   017230   450   9
2  89
 0  0  1  0 383816  17900 205768   0   017210   440  12
3  86
 0  0  0  0 389004  17920 206412   0   0 4230   366   8
2  90
 0  0  0  0 383012  17928 206900   0   0 0190   424  10
2  88
..
...
in the evening,
 2  0  0  0 123940  23888 447436   0   0 0360   464  38
4  58
 0  0  0  0 130812  23920 448940   0   016320   503  41
4  55
 2  0  1  0 129868  23936 449484   0   0 0200   443  14
3  84
 6  0  0  0 108252  23956 450172   0   0 0270   524  15
3  82
 3  0  1  0  89744  23972 450800   0   0 0220   525  13
3  84
 2  0  1  0  82496  24000 451636   0   0 0240   555  34
4  62
 1  0  1  0  84892  24012 451432   0   0 0310   634  18
4  78
 4  0  0  0  80232  24028 452252   0   0 0220   668  16
3  81
 1  0  2  0  82200  24040 453028   0   0 0260   574  20
3  76
 0  0  1  0  81536  24052 453856   0   0 0240   621  17
3  80
 0  0  1  0  83396  24060 454448   0   0 0240   512  14
3  83
 0  0  1  0  84584  24076 454864   0   0 0200   385  11
2  87
 1  0  1  0  63404  24088 455428   0   0 0180   683  12
2  85
 1  0  0  0  50252  24108 456164   0   0 0260   590  17
3  79
 0  0  1  0  46016  24128 456820   0   0 0220   479  12
3  86
 2  0  1  0  47052  24144 456688   0   0 0270   621  16
3  81
 0  0  0  0  44228  24160 457324   0   0 0260   501  11
2  86
 3  0  1  0  47480  24172 458020   0   0 0220   506  11
3  87
 1  0  1  0  44656  24188 458660   0   0 0220   495  14
3  83
 0  0  0  0  40288  24204 459416   0   0 0240   536  15
3  82
 1  0  0  0  40036  24216 460192   0   0 0260   574  15
3  82

all free memory is kept by cache.
When memory problem occurs, only few pages are freed. What parameter should
I set to limit the number of page cache ?

Thanks in advance.


   Gerard MONTELEONE
  Ingenieur Systeme & Reseau
SI.TE.C Z.I du vazzio 20090 AJACCIO
* +33495236809 * +33687727032
  www.sitec.fr <http://www.sitec.fr/>


--
For LINUX-390 subscribe / signoff / archive access instructions, send email
to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Kernel memory allocation

[Monteleone]

Hello,

Is there a way to limit the number of  "page cached" on a SLES8 system
without limiting the size of the guest ?
I want to reserve a large number of free memory pages for the
application.
Have a look to the allocation please:
In the morning,
   procs  memory  swap  io
system cpu
 r  b  w   swpd   free   buff  cache   si  sobibo   incs
us  sy  id
 0  0  1  0 407880  17296 191360   0   0 8 4058   2
0  98
 1  0  0  0 410132  17320 191760   0   0 0220   305   8
2  90
 1  0  1  0 409560  17352 192440   0   0 6170   319   6
2  92
 0  0  1  0 406008  17360 192940   0   0 3170   297   7
2  91
 1  0  0  0 408220  17372 193504   0   0 5170   248   7
1  92
 0  0  1  0 407056  17384 193900   0   0 0150   311   9
2  89
 0  0  1  0 401012  17400 194512   0   0 0190   452  12
2  86
 0  0  1  0 402420  17412 194976   0   0 0170   308  12
2  86
 1  0  0  0 399020  17420 194548   0   0 3210   395  11
2  86
 0  0  0  0 396060  17424 195000   0   0 0170   403  10
2  88
 1  0  0  0 404200  17432 195360   0   0 0190   312   8
2  90
 1  0  0  0 397712  17444 195828   0   0 0130   394  10
2  88
 0  0  0  0 400108  17456 196248   0   0 0210   370   7
2  91
 3  0  1  0 400468  17532 197124   0   0 8190   332  10
2  89
 1  0  0  0 383604  17804 202784   0   090230   426  14
2  84
 0  0  1  0 388692  17836 204224   0   017230   450   9
2  89
 0  0  1  0 383816  17900 205768   0   017210   440  12
3  86
 0  0  0  0 389004  17920 206412   0   0 4230   366   8
2  90
 0  0  0  0 383012  17928 206900   0   0 0190   424  10
2  88
..
...
in the evening,
 2  0  0  0 123940  23888 447436   0   0 0360   464  38
4  58
 0  0  0  0 130812  23920 448940   0   016320   503  41
4  55
 2  0  1  0 129868  23936 449484   0   0 0200   443  14
3  84
 6  0  0  0 108252  23956 450172   0   0 0270   524  15
3  82
 3  0  1  0  89744  23972 450800   0   0 0220   525  13
3  84
 2  0  1  0  82496  24000 451636   0   0 0240   555  34
4  62
 1  0  1  0  84892  24012 451432   0   0 0310   634  18
4  78
 4  0  0  0  80232  24028 452252   0   0 0220   668  16
3  81
 1  0  2  0  82200  24040 453028   0   0 0260   574  20
3  76
 0  0  1  0  81536  24052 453856   0   0 0240   621  17
3  80
 0  0  1  0  83396  24060 454448   0   0 0240   512  14
3  83
 0  0  1  0  84584  24076 454864   0   0 0200   385  11
2  87
 1  0  1  0  63404  24088 455428   0   0 0180   683  12
2  85
 1  0  0  0  50252  24108 456164   0   0 0260   590  17
3  79
 0  0  1  0  46016  24128 456820   0   0 0220   479  12
3  86
 2  0  1  0  47052  24144 456688   0   0 0270   621  16
3  81
 0  0  0  0  44228  24160 457324   0   0 0260   501  11
2  86
 3  0  1  0  47480  24172 458020   0   0 0220   506  11
3  87
 1  0  1  0  44656  24188 458660   0   0 0220   495  14
3  83
 0  0  0  0  40288  24204 459416   0   0 0240   536  15
3  82
 1  0  0  0  40036  24216 460192   0   0 0260   574  15
3  82

all free memory is kept by cache.
When memory problem occurs, only few pages are freed. What parameter
should I set to limit the number of page cache ?

Thanks in advance.


   Gerard MONTELEONE
  Ingenieur Systeme & Reseau
SI.TE.C Z.I du vazzio 20090 AJACCIO
* +33495236809 * +33687727032
  www.sitec.fr 


--
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390