Re: High interrupt rate

2009-06-03 Thread Mark Post
>>> On 6/3/2009 at  8:18 AM, "David K. Kelly"  
>>> wrote: 

> 
> HI Marcy,
> 
> When you say SuSE's web site did you mean suse.org.  I tried Novell's
> site but under the link below I couldn't  find any of the kernel patches.

I'm pretty sure she didn't mean suse.org.  :)  More likely nu.novell.com, which 
is where YOU/rug/zmd go for updates.

> Do you have a better link?  Also when I tried rug I received the following:
> ERROR: 'kernel-default' is not available.

Do you have an update source configured?  "rug sl"  If yes, then what level is 
your kernel currently?  "uname -r".


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


Re: High interrupt rate

2009-06-03 Thread David K. Kelly


HI Marcy,

When you say SuSE's web site did you mean suse.org.  I tried Novell's
site but under the link below I couldn't  find any of the kernel patches.
Do you have a better link?  Also when I tried rug I received the following:
ERROR: 'kernel-default' is not available.

http://download.novell.com/index.jsp?product_id=&search=Search&families=2658&version=19925&date_range=&date_start=03+Jun+2009&date_end=03+Jun+2009&keywords=kernel&sort_by=&results_per_page=&x=36&y=7

(Embedded image moved to file: pic30932.jpg)


Thank you

David Kelly
The Supreme Court of Virginia





   
 Marcy Cortes  
  To
 Sent by: Linux on LINUX-390@VM.MARIST.EDU 
 390 Port   cc
           Subject
   Re: High interrupt rate 
   
 05/29/2009 03:21  
 PM
   
   
 Please respond to 
 Linux on 390 Port 
   
   
   




You can either grab the rpm off of SuSE's website or run "rug install
kernel-default" .


Marcy

"This message may contain confidential and/or privileged information. If
you are not the addressee or authorized to receive this for the addressee,
you must not use, copy, disclose, or take any action based on this message
or any information herein. If you have received this message in error,
please advise the sender immediately by reply e-mail and delete this
message. Thank you for your cooperation."


-Original Message-
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of David
K. Kelly
Sent: Friday, May 29, 2009 12:16 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] High interrupt rate

Hi Rob

Is it possible to just down load the kernel patch for this and not all the
extra packages
that you get when you request an update?

Thank you

David Kelly
The Supreme Court of Virginia





 Rob van der Heij
   To
 Sent by: Linux on LINUX-390@VM.MARIST.EDU
 390 Port   cc
       Subject
   Re: High interrupt rate

 05/28/2009 01:48
 PM


 Please respond to
 Linux on 390 Port
 






On Thu, May 28, 2009 at 5:12 PM, Deric Abel  wrote:

> We are running sles10 sp2 not redhat, so I am not sure if this bug was in
any of the red hat releases.  The kernel that comes with sp2 was
2.6.16.60-0.21, and the most recent kernel for sp2 is
2.6.16.60-0.37  (remember that the number after the "-" is the rpm
build/release version, so that will be different between distros).  I am
not sure want the underlying problem was, but from this mailing list it
seems there was a bug in the sles10sp2 kernel and it's fixed with the
newest.  So unless someone else would like to explain the bug, that's all
the information I have.

The cause is that the application asks for a wakeup in 1 ms or so
(this is silly design since it is polling). The kernel only provides
10 ms granularity and rounded incorrectly to 0 ms, thus making the
silly thing pretty bad. It also shows very high CP overhead when this
happens.
The fix is that the kernel rounds it to 10 ms to limit the impact.

Martin just told me the fix is in: 2.6.18-124.el5 für RedHat RHEL5 und
2.6.16.60-0.34 für SuSE SLES10

--
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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 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-39

Re: High interrupt rate

2009-06-01 Thread Deric Abel
David,   I just grabbed the kernel rpm from novell's site (since my guests 
don't have outside access) and applied the RPMs.  That has solved my problem 
across the board, where an interrupt rate of 5000/sec was normal they are all 
now at 105 or less.


Deric

--
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: High interrupt rate

2009-05-29 Thread Rob van der Heij
On Fri, May 29, 2009 at 9:15 PM, David K. Kelly
 wrote:
> Hi Rob
>
> Is it possible to just down load the kernel patch for this and not all the
> extra packages
> that you get when you request an update?

An easy way to see if this is indeed your problem: set the hz_timer to
1 and see whether it goes away. If you have a buggy application that
is polling, you will not drop from queue anyway. So if it fixes the
issue you could leave it until the new kernel is there.

Rob
--
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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: High interrupt rate

2009-05-29 Thread Marcy Cortes
You can either grab the rpm off of SuSE's website or run "rug install 
kernel-default" .


Marcy

"This message may contain confidential and/or privileged information. If you 
are not the addressee or authorized to receive this for the addressee, you must 
not use, copy, disclose, or take any action based on this message or any 
information herein. If you have received this message in error, please advise 
the sender immediately by reply e-mail and delete this message. Thank you for 
your cooperation."


-Original Message-
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of David K. 
Kelly
Sent: Friday, May 29, 2009 12:16 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] High interrupt rate

Hi Rob

Is it possible to just down load the kernel patch for this and not all the
extra packages
that you get when you request an update?

Thank you

David Kelly
The Supreme Court of Virginia





 Rob van der Heij
   To
 Sent by: Linux on LINUX-390@VM.MARIST.EDU
 390 Port   cc
   Subject
           Re: High interrupt rate

 05/28/2009 01:48
 PM


 Please respond to
 Linux on 390 Port
 






On Thu, May 28, 2009 at 5:12 PM, Deric Abel  wrote:

> We are running sles10 sp2 not redhat, so I am not sure if this bug was in
any of the red hat releases.  The kernel that comes with sp2 was
2.6.16.60-0.21, and the most recent kernel for sp2 is
2.6.16.60-0.37  (remember that the number after the "-" is the rpm
build/release version, so that will be different between distros).  I am
not sure want the underlying problem was, but from this mailing list it
seems there was a bug in the sles10sp2 kernel and it's fixed with the
newest.  So unless someone else would like to explain the bug, that's all
the information I have.

The cause is that the application asks for a wakeup in 1 ms or so
(this is silly design since it is polling). The kernel only provides
10 ms granularity and rounded incorrectly to 0 ms, thus making the
silly thing pretty bad. It also shows very high CP overhead when this
happens.
The fix is that the kernel rounds it to 10 ms to limit the impact.

Martin just told me the fix is in: 2.6.18-124.el5 für RedHat RHEL5 und
2.6.16.60-0.34 für SuSE SLES10

--
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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 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 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: High interrupt rate

2009-05-29 Thread David K. Kelly
Hi Rob

Is it possible to just down load the kernel patch for this and not all the
extra packages
that you get when you request an update?

Thank you

David Kelly
The Supreme Court of Virginia




   
 Rob van der Heij  
   To
 Sent by: Linux on LINUX-390@VM.MARIST.EDU 
 390 Port   cc
   Subject
   Re: High interrupt rate 
   
 05/28/2009 01:48  
 PM
   
   
 Please respond to 
 Linux on 390 Port 
   
   
   




On Thu, May 28, 2009 at 5:12 PM, Deric Abel  wrote:

> We are running sles10 sp2 not redhat, so I am not sure if this bug was in
any of the red hat releases.  The kernel that comes with sp2 was
2.6.16.60-0.21, and the most recent kernel for sp2 is
2.6.16.60-0.37  (remember that the number after the "-" is the rpm
build/release version, so that will be different between distros).  I am
not sure want the underlying problem was, but from this mailing list it
seems there was a bug in the sles10sp2 kernel and it's fixed with the
newest.  So unless someone else would like to explain the bug, that's all
the information I have.

The cause is that the application asks for a wakeup in 1 ms or so
(this is silly design since it is polling). The kernel only provides
10 ms granularity and rounded incorrectly to 0 ms, thus making the
silly thing pretty bad. It also shows very high CP overhead when this
happens.
The fix is that the kernel rounds it to 10 ms to limit the impact.

Martin just told me the fix is in: 2.6.18-124.el5 für RedHat RHEL5 und
2.6.16.60-0.34 für SuSE SLES10

--
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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 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: High interrupt rate

2009-05-28 Thread Rob van der Heij
On Thu, May 28, 2009 at 5:12 PM, Deric Abel  wrote:

> We are running sles10 sp2 not redhat, so I am not sure if this bug was in any 
> of the red hat releases.  The kernel that comes with sp2 was 2.6.16.60-0.21, 
> and the most recent kernel for sp2 is 2.6.16.60-0.37  (remember that the 
> number after the "-" is the rpm build/release version, so that will be 
> different between distros).  I am not sure want the underlying problem was, 
> but from this mailing list it seems there was a bug in the sles10sp2 kernel 
> and it's fixed with the newest.  So unless someone else would like to explain 
> the bug, that's all the information I have.

The cause is that the application asks for a wakeup in 1 ms or so
(this is silly design since it is polling). The kernel only provides
10 ms granularity and rounded incorrectly to 0 ms, thus making the
silly thing pretty bad. It also shows very high CP overhead when this
happens.
The fix is that the kernel rounds it to 10 ms to limit the impact.

Martin just told me the fix is in: 2.6.18-124.el5 für RedHat RHEL5 und
2.6.16.60-0.34 für SuSE SLES10

-- 
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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: High interrupt rate

2009-05-28 Thread Deric Abel
We are running sles10 sp2 not redhat, so I am not sure if this bug was in any 
of the red hat releases.  The kernel that comes with sp2 was 2.6.16.60-0.21, 
and the most recent kernel for sp2 is 2.6.16.60-0.37  (remember that the number 
after the "-" is the rpm build/release version, so that will be different 
between distros).  I am not sure want the underlying problem was, but from this 
mailing list it seems there was a bug in the sles10sp2 kernel and it's fixed 
with the newest.  So unless someone else would like to explain the bug, that's 
all the information I have.

>>> Alan Ackerman  5/27/2009 2:30 PM >>>
I read through this thread. I see information about db2fmcd, which
you say did not cause your problem. I don't see anything in the thread
that tells me about the problem or fix to the kernel.

Can you point me to information about the kernel problem?

Just to confuse things, we have Red Hat, not SuSE.

>Thanks everyone for your help!  Based on everything I've read, I concluded
>to upgrade the kernel to 2.6.16.60-0.37.  That has solve our issue.  As
>for the db2fmcd process, I did not find that running nor in the /etc/inittab,
>so no changes were made there.

Alan Ackerman
Alan (dot) Ackerman (at) Bank of America (dot) com

--
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 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: High interrupt rate

2009-05-27 Thread Alan Ackerman
I read through this thread. I see information about db2fmcd, which
you say did not cause your problem. I don't see anything in the thread
that tells me about the problem or fix to the kernel.

Can you point me to information about the kernel problem?

Just to confuse things, we have Red Hat, not SuSE.

>Thanks everyone for your help!  Based on everything I've read, I concluded
>to upgrade the kernel to 2.6.16.60-0.37.  That has solve our issue.  As
>for the db2fmcd process, I did not find that running nor in the /etc/inittab,
>so no changes were made there.

Alan Ackerman
Alan (dot) Ackerman (at) Bank of America (dot) com

--
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: High interrupt rate

2009-05-26 Thread Rob van der Heij
On Tue, May 26, 2009 at 7:34 PM, Deric Abel  wrote:
> Thanks everyone for your help!  Based on everything I've read, I concluded to 
> upgrade the kernel to 2.6.16.60-0.37.  That has solve our issue.  As for the 
> db2fmcd process, I did not find that running nor in the /etc/inittab, so no 
> changes were made there.  Thanks again,

I've been unable to keep track of which Novel kernel has which fixes
from IBM. But if the new kernel fixes the problem, then it sure looks
like another db2 component has the same issue. So it's probably a
design fault rather than a bug ;-)
The kernel issue only shows the problem by actually doing the thing
that the application was asking for.

Rob
-- 
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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: High interrupt rate

2009-05-26 Thread Deric Abel
Thanks everyone for your help!  Based on everything I've read, I concluded to 
upgrade the kernel to 2.6.16.60-0.37.  That has solve our issue.  As for the 
db2fmcd process, I did not find that running nor in the /etc/inittab, so no 
changes were made there.  Thanks again,

Deric

--
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: High interrupt rate

2009-05-26 Thread Richard Gasiorowski
You can also issue the db2fmcu -d command.  This erases the fmc entries in
inittab.  Also ./sqllib/fm.whatever.reg  check the instances and make sure
everyone FM_ON and FM_ACTIVE are set to NO.

'Where ever you go - There you are!! '
Richard (Gaz) Gasiorowski
System z - Linux Product Manager
Portfolio Platform Services
CSC
3170 Fairview Park Dr., Falls Church, VA 22042
845-889-8533|Work|845-392-7889 Cell|rgasi...@csc.com|www.csc.com




This is a PRIVATE message. If you are not the intended recipient, please
delete without copying and kindly advise us by e-mail of the mistake in
delivery.
NOTE: Regardless of content, this e-mail shall not operate to bind CSC to
any order or other contract unless pursuant to explicit written agreement
or government initiative expressly permitting the use of e-mail for such
purpose.



Rob van der Heij 
Sent by: Linux on 390 Port 
05/26/2009 11:22 AM
Please respond to
Linux on 390 Port 


To
LINUX-390@VM.MARIST.EDU
cc

Subject
Re: High interrupt rate






On Tue, May 26, 2009 at 5:10 PM, Deric Abel 
wrote:
> Hi,  I have an issue with one of my linux guests (sles10) where when we
enable HADR with db2 the HADR db guest interrupt rate jumps from a few
thousand to 300,000 per second.  This causes the guest to be using 20-30%
of one IFL, despite doing very little.  Has anyone seen this issue before,
or know of a good way to debug it?

Without performance data but you probably have db2fmcd running.
This has nothing to do with HADR but frequently interferes with it.
We're not sure whether it is a bug or design problem with db2fmcd that
it sometimes starts to poll. Combined with a kernel bug in SLES10 SP2,
it typically causes a lot of overhead. Since db2fmcd has no function
in this environment, the easy way out is to stop it (and remove the
entry in /etc/inittab )

Rob
--
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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 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: High interrupt rate

2009-05-26 Thread David Boyes
> Hi,  I have an issue with one of my linux guests (sles10) where when we
> enable HADR with db2 the HADR db guest interrupt rate jumps from a few

HADR?

> (BTW, I am brand new to this mailing list as well as new to the Z
> system, but not to linux)

Cool. Welcome to the party. 

-- db

--
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: High interrupt rate

2009-05-26 Thread David Boyes
> it typically causes a lot of overhead. Since db2fmcd has no function
> in this environment, the easy way out is to stop it (and remove the
> entry in /etc/inittab )

It would also be really useful if someone sat down with the people in DB/2 
development and explained that using /etc/inittab to manage services on Linux 
is Generally Considered Rude. Grr. What a PITA. 

--d b
 

--
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: High interrupt rate

2009-05-26 Thread Rob van der Heij
On Tue, May 26, 2009 at 5:10 PM, Deric Abel  wrote:
> Hi,  I have an issue with one of my linux guests (sles10) where when we 
> enable HADR with db2 the HADR db guest interrupt rate jumps from a few 
> thousand to 300,000 per second.  This causes the guest to be using 20-30% of 
> one IFL, despite doing very little.  Has anyone seen this issue before, or 
> know of a good way to debug it?

Without performance data but you probably have db2fmcd running.
This has nothing to do with HADR but frequently interferes with it.
We're not sure whether it is a bug or design problem with db2fmcd that
it sometimes starts to poll. Combined with a kernel bug in SLES10 SP2,
it typically causes a lot of overhead. Since db2fmcd has no function
in this environment, the easy way out is to stop it (and remove the
entry in /etc/inittab )

Rob
-- 
Rob van der Heij
Velocity Software
http://www.velocitysoftware.com/

--
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: High interrupt rate

2009-05-26 Thread Deric Abel
>>> David Boyes  5/26/2009 9:14 AM >>>
> Hi,  I have an issue with one of my linux guests (sles10) where when we
> enable HADR with db2 the HADR db guest interrupt rate jumps from a few

HADR?

High Availability Disaster Recovery
It's a DB2 process that replicates db's between servers (guests in our case)

Deric

> (BTW, I am brand new to this mailing list as well as new to the Z
> system, but not to linux)

Cool. Welcome to the party. 

Thanks, very happy to be here!

-- db

--
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 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