Knutson, Sam
2004-05-07 16:42:25 UTC
Hi all,
I need to extend the retention of our RMF Monitor III data. The RMF Users
Guide's says
You can define up to 100 data sets for use with the data gatherer. At a
minimum, you should define at least two data sets, because the gatherer
deletes all data in a data set before writing to it, so a single data set
would be emptied immediately after it was filled. RMF can keep about 1100
sets of samples in one data set. Based on practical experience, we recommend
you to define six VSAM data sets, each with 50 cylinders disk space, this is
sufficient for 1100 set-of-samples for most installations. On small and
medium systems, this will allow for about two days of data.
Note: If you need to change the amount of space later to meet your
installation's needs, we recommend adding more data sets, but not making the
individual data sets larger. Increasing the size of the data sets may cause
RMF to run out of index entries and be unable to fill the additional space.
So for the benefit of others let me share our mistake
I looked at the current data sets we have allocated at 5100 cylinders
----Begin/End----
System --Date-- --Time-- -DDNAME- ---------------Data Set
Name----------------
ASYS 05/05/04 19.46.00 SYS00103 DSSV.RMFMON3.ASYS.DS001A
05/06/04 14.16.00
ASYS 05/06/04 14.16.00 SYS00101 DSSV.RMFMON3.ASYS.DS001C
05/07/04 08.46.00
ASYS 05/07/04 08.46.00 SYS00104 DSSV.RMFMON3.ASYS.DS001B
10.46.00 * * * Currently active * * *
ASYS 05/07/04 10.34.00
10.46.00 * * * In-storage buffer * * *
Hmm..... How many samples are in those?
REPRO INFILE(INPUT) OUTFILE(OUTPUT)
IDC3302I ACTION ERROR ON DSSV.RMFMON3.ASYS.DS099A
IDC3351I ** VSAM I/O RETURN CODE IS 28 - RPLFDBWD = X'2908001C'
IDC31467I MAXIMUM ERROR LIMIT REACHED.
IDC0005I NUMBER OF RECORDS PROCESSED WAS 1100
IDC3003I FUNCTION TERMINATED. CONDITION CODE IS 12
Yes.... 1100. 1100 is what you get in 50 cylinders and 1100 is what you get
in 5100 cylinders. Doh!
So the moral of the story is always read the manual.
The opportunity is go look at your own and if they are much more than 50
cylinders break them up into smaller data sets and save space, keep more
data,
or both.
I am going to try actually running the gatherer with 100 50 CYL datasets and
see if it works OK and if it chews up excessive CPU or storage in RMFGAT.
Anyone got any experience trying to maximize on-line RMF III history you
care to share?
Best Regards,
Sam Knutson, GEICO
Performance and Availability Management
mailto:***@geico.com
(office) 301.986.3574
The less effort, the faster and more powerful you will be. Bruce Lee
===================
This email/fax message is for the sole use of the intended recipient(s) and
may contain confidential and privileged information. Any unauthorized
review, use, disclosure or distribution of this email/fax is prohibited. If
you are not the intended recipient, please destroy all paper and electronic
copies of the original message.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
I need to extend the retention of our RMF Monitor III data. The RMF Users
Guide's says
You can define up to 100 data sets for use with the data gatherer. At a
minimum, you should define at least two data sets, because the gatherer
deletes all data in a data set before writing to it, so a single data set
would be emptied immediately after it was filled. RMF can keep about 1100
sets of samples in one data set. Based on practical experience, we recommend
you to define six VSAM data sets, each with 50 cylinders disk space, this is
sufficient for 1100 set-of-samples for most installations. On small and
medium systems, this will allow for about two days of data.
Note: If you need to change the amount of space later to meet your
installation's needs, we recommend adding more data sets, but not making the
individual data sets larger. Increasing the size of the data sets may cause
RMF to run out of index entries and be unable to fill the additional space.
So for the benefit of others let me share our mistake
I looked at the current data sets we have allocated at 5100 cylinders
----Begin/End----
System --Date-- --Time-- -DDNAME- ---------------Data Set
Name----------------
ASYS 05/05/04 19.46.00 SYS00103 DSSV.RMFMON3.ASYS.DS001A
05/06/04 14.16.00
ASYS 05/06/04 14.16.00 SYS00101 DSSV.RMFMON3.ASYS.DS001C
05/07/04 08.46.00
ASYS 05/07/04 08.46.00 SYS00104 DSSV.RMFMON3.ASYS.DS001B
10.46.00 * * * Currently active * * *
ASYS 05/07/04 10.34.00
10.46.00 * * * In-storage buffer * * *
Hmm..... How many samples are in those?
REPRO INFILE(INPUT) OUTFILE(OUTPUT)
IDC3302I ACTION ERROR ON DSSV.RMFMON3.ASYS.DS099A
IDC3351I ** VSAM I/O RETURN CODE IS 28 - RPLFDBWD = X'2908001C'
IDC31467I MAXIMUM ERROR LIMIT REACHED.
IDC0005I NUMBER OF RECORDS PROCESSED WAS 1100
IDC3003I FUNCTION TERMINATED. CONDITION CODE IS 12
Yes.... 1100. 1100 is what you get in 50 cylinders and 1100 is what you get
in 5100 cylinders. Doh!
So the moral of the story is always read the manual.
The opportunity is go look at your own and if they are much more than 50
cylinders break them up into smaller data sets and save space, keep more
data,
or both.
I am going to try actually running the gatherer with 100 50 CYL datasets and
see if it works OK and if it chews up excessive CPU or storage in RMFGAT.
Anyone got any experience trying to maximize on-line RMF III history you
care to share?
Best Regards,
Sam Knutson, GEICO
Performance and Availability Management
mailto:***@geico.com
(office) 301.986.3574
The less effort, the faster and more powerful you will be. Bruce Lee
===================
This email/fax message is for the sole use of the intended recipient(s) and
may contain confidential and privileged information. Any unauthorized
review, use, disclosure or distribution of this email/fax is prohibited. If
you are not the intended recipient, please destroy all paper and electronic
copies of the original message.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html