If you are seeing high numbers (read > 50 in small systems and 100 in large)
in program page faulting, in *BASE it's entirely likely the activity level
is too low, and the memory is also constrained. You can of course us the
performance tools to verify but generally I start to increase the activity
level/memory about 10% at a time until the numbers settle down over time.
An activity level that's really too high is not going to hurt *BASE like it
might other pools. I tend to set them fairly high to avoid unpleasant
outcomes, like the one Rob has experienced.
This assumes you have pushed all user work out of *BASE (meaning SQL based
communications and all other processes that tend to land there) the
exception being of course subsystem monitor jobs. Those should always be
there.
Those of you with hosted environments should be using the ability to push
the network servers into a shared pool as well. I usually use *SHAREPOOL 50
since almost no one has that many shared pools.
Watch the job transitions. Those are also a good indicator of health, or
the lack thereof in the shared memory pools.
--
Jim Oberholtzer
Agile Technology Architects
-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of Rob
Berendt
Sent: Friday, November 1, 2019 6:54 AM
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxxxxxxxx>
Subject: RE: QBASACTLVL recommendations.
I too wondered why current can be greater than max. This is true for me on
other pools, like spool. I wonder if current is not necessarily "active"?
From a high level perspective all I know is that when the system will not
let any new jobs start in any subsystem, in any pool, I need to increase
QBASACTLVL and they fire off again.
Then again, what version of the OS are you running?
Rob Berendt
--
IBM Certified System Administrator - IBM i 6.1 Group Dekko Dept 1600 Mail
to: 2505 Dekko Drive
Garrett, IN 46738
Ship to: Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com
-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of Don
Brown via MIDRANGE-L
Sent: Thursday, October 31, 2019 5:07 PM
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxxxxxxxx>
Cc: Don Brown <DBrown@xxxxxxxxxx>
Subject: RE: QBASACTLVL recommendations.
CAUTION: This email originated from outside of the organization. Do not
click links or open attachments unless you recognize the sender and know the
content is safe.
I tried that on one system and got the following results;
Pool=2
Name=*BASE
MaxThreads=738
CurThreads=1133
%=1.535
It has 64GB Memory
So you are suggesting the MaxThreads should be increased ? Using what
formula ?
Don Brown
From: "Rob Berendt" <rob@xxxxxxxxx>
To: "Midrange Systems Technical Discussion"
<midrange-l@xxxxxxxxxxxxxxxxxx>
Date: 31/10/2019 11:52 PM
Subject: RE: QBASACTLVL recommendations.
Sent by: "MIDRANGE-L" <midrange-l-bounces@xxxxxxxxxxxxxxxxxx>
I may add this to Halcyon also:
-- Are we approaching maximum number of threads?
With T1 as (
select
system_pool_id,
pool_name,
maximum_active_threads,
current_threads,
dec(
dec(current_threads, 15, 3) / dec(maximum_active_threads, 15, 3)
, 15, 3) as Thread_Percent
FROM gdihq.QSYS2.MEMORY_POOL_INFO
where system_pool_id = 2)
select *
from T1
where Thread_Percent > .40
;
The low value of .40 will be raised, after testing...
Rob Berendt
--
IBM Certified System Administrator - IBM i 6.1
Group Dekko
Dept 1600
Mail to: 2505 Dekko Drive
Garrett, IN 46738
Ship to: Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com
-----Original Message-----
From: MIDRANGE-L <midrange-l-bounces@xxxxxxxxxxxxxxxxxx> On Behalf Of Rob
Berendt
Sent: Thursday, October 31, 2019 9:31 AM
To: Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxxxxxxxx>
Subject: QBASACTLVL recommendations.
What are the recommendations for QBASACTLVL? As you can see below it's
currently at 6050. We had to raise it from 5050 on an emergency basis.
IDK if 7.4 is really going wild with threading but QPFRADJ is not
adjusting this at all and it's chopping us off at the knees at the most
inopportune times. Yes I have a case opened with this.
WRKSYSSTS
% CPU used . . . . . . . : 13.2
Elapsed time . . . . . . : 00:05:03
Jobs in system . . . . . : 75263
% perm addresses . . . . : 33.636
% temp addresses . . . . : .267
System ASP . . . . . . . : 7829 G
% system ASP used . . . : 76.2636
Total aux stg . . . . . : 7829 G
Current temporary used . : 243896 M
Peak temporary used . . : 260207 M
Sys Pool Reserved Max ------DB------- ----Non-DB-----
Pool Size M Size M Act Faults Pages Faults Pages
1 7263.3 3777.8 +++++ .0 .0 .1 .2
2 31944.9 21.7 6050 .7 111.4 29.2 100.0
3 1331.0 1.2 30 .0 .0 .6 11.6
4 21581.0 .0 558 10.2 3734.3 67.0 113.9
5 70983.5 4.1 2863 8.0 2420.0 224.8 356.9
DSPHDWRSC *PRC
Resource Type-model
CEC01 9009-42A
PN02 6B5C
PN03 6B5D
MP17 5C28
MP18 5C28
MP19 5C28
MP20 5C28
MP21 5C28
MP22 5C28
MP23 5C28
MP24 5C28
MP25 5C28
MP26 5C28
MP27 5C28
MP28 5C28
MP29 5C28
MP30 5C28
MP31 5C28
MP32 5C28
PV02 562F
SP02 2DEA
BCC02
MS17 324F
MS18 324F
MS19 324F
MS20 324F
MS21 324F
MS22 324F
MS23 324F
MS24 324F
MS25 324F
MS26 324F
MS27 324F
MS28 324F
MS29 324F
MS30 324F
MS31 324F
MS32 324F
MS33 324F
MS34 324F
MS35 324F
MS21 324F
MS22 324F
MS23 324F
MS24 324F
MS25 324F
MS26 324F
MS27 324F
MS28 324F
MS29 324F
MS30 324F
MS31 324F
MS32 324F
MS33 324F
MS34 324F
MS35 324F
Rob Berendt
--
IBM Certified System Administrator - IBM i 6.1
Group Dekko
Dept 1600
Mail to: 2505 Dekko Drive
Garrett, IN 46738
Ship to: Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit:
https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at
https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxx for any subscription related
questions.
Help support midrange.com by shopping at amazon.com with our affiliate
link:
https://amazon.midrange.com
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit:
https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at
https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxx for any subscription related
questions.
Help support midrange.com by shopping at amazon.com with our affiliate
link:
https://amazon.midrange.com
______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
______________________________________________________________________
______________________________________________________________________
This email has been scanned for computer viruses. Although MSD has taken
reasonable precautions to ensure no viruses are present in this email, MSD
cannot accept responsibility for any loss or damage arising from the use of
this email or attachments.
______________________________________________________________________
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit:
https://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxxxxxxxx
Before posting, please take a moment to review the archives
at
https://archive.midrange.com/midrange-l.
Please contact support@xxxxxxxxxxxx for any subscription related questions.
Help support midrange.com by shopping at amazon.com with our affiliate link:
https://amazon.midrange.com
midrange.com
