Hi...
Thanks for the reply. After the longer data analysis from the SMF 119
record we have some anxieties as for what we will get. More precisely we
(how you recalled) earlier gathered information from RTM. where we had
time interval every 5 minutes and we would like so that it is in this
case. (We changed terminals on connected by TCP/IP and we can't get
needed data from RTM). We have "anxieties" that we will get only data
from the end of the session rather than ranges 5 minute's. Therefore my
next question: whether someone have experience with this SMF 119 and are
we right as for the received information?.. and perhaps somebody has
some job to "working out" of this record.?
.......I apologize for my written English
--
Best regards,
Rafal Hanzel
Systems Programmer, R&D of Computer System Department
Z.E.T.O Katowice Sp. z o.o.
ul. Owocowa 1
40-158 Katowice, Poland
Phone: +48 32 3589 246
e-mail: [EMAIL PROTECTED]
Chris Mason pisze:
Rafal
If you have a look in the CS IP Configuration Reference manual, you will see
there are some appendices. If you look in Appendix C, you will see that it has
the title "SMF type 119 records". I can tell from the command you used that
you appear to be interested in the "TN3270E Telnet server SNA session
termination record" - or, if you didn't know that before, you do now!
If you examine "Table 137. TN3270E Telnet server SNA session termination
record self-defining section" (the highest level of the SMF record since it
contains the SMF header), you will find starting at offset 60(x’3C’), the
following 3 fields, "Offset to TN3270 server session time bucket performance
data section", "Length of TN3270 server session time bucket performance data
section" and "Number of TN3270 server session time bucket performance data
sections".
If you look at "Table 141. TN3270E Telnet server time bucket performance
section" I believe you will have found exactly that for which you are looking.
I hope you understand how these "buckets" work. If not please post again. It
looks like something plagiarised from the SNA response time measurement
(RTM) function which emerged all of 25 years ago in places such as the 3174
and NetView Session Monitor, then called Network Logical Data Manager and
which I used to teach in some detail. It's good to see TN3270 catching up
with SNA 3270 - even if all of a quarter century later!
I even now recall in the mid-80's having students create beautiful
presentations of response time ranges in GDDM graphs from the SMF record
analysis of RTM data by the SMF-processing product which IBM offered in
those days by the name of SLR - sadly I needed Google to help me recall what
SLR meant: "Service Level Reporter". Unless SLR still exists, you will need to
find today's equivalent product in order to process the SMF 119 records - and
maybe today's equivalent of GDDM in order to present the analysis! I hope
others can offer recommendations - or, since the topic of SMF analysis comes
up in the list from time to time, check the archives.
Incidentally I found this in the CS IP Configuration Guide:
<quote>
An external mapping (EZASMF77 macro) is available for customers to parse
the SMF type 119 records that TCP/IP generates.
</quote>
Scanning online versions of the manuals with the right search words can be a
rewarding experience.
Chris Mason
On Wed, 2 Jul 2008 12:19:52 +0200, Rafal Hanzel
<[EMAIL PROTECTED]> wrote:
Hi all ...
I have a question about SMF record type 119 ... is there the same
information as after I issued /D TCPIP,,T,CONN,CONN=xxxx
MONGRP1 --------
0090 PERIOD: 300 MULT: 60
0090 /W AVG LOC AVG SUM R/T SSQ R/T ST DEV
0090 ======= ======= ======== ============ =======
0090 SNA: 26 26 7985 500243
31
0090 IP: 143 143 43975 7365709
59
0090 TOTAL: 169 169 51960 10307156 70
0090 COUNT: 279 307
0090 BUCKET1 BUCKET2 BUCKET3 BUCKET4 BUCKET5
0090 1000 2000 5000
10000 NO LMT
0090 307 0 0
0 0
BUCKET is most important for me ....
If answer is YES .... maybe someone have any samples to read this
information from SMF records...
or at least some pointers....
thank you in advance
--
Best regards,
Rafal Hanzel
Systems Programmer, R&D of Computer System Department
Z.E.T.O Katowice Sp. z o.o.
ul. Owocowa 1
40-158 Katowice, Poland
Phone: +48 32 3589 246
e-mail: [EMAIL PROTECTED]
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
