Re: CPU time/instruction table

[Edward Jaffe]

Gerhard Postpischil wrote:

Edward Jaffe wrote:
The first several pages would have to explain the pipeline design and 
the delays that can occur. After that, specific instruction timings 
would have to assume the pipeline is flowing freely.


Management doesn't care about individual instruction timings, but the 
overall throughput and response time.


An instruction timings book is not intended for use by management.

--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/

--
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

Re: z/OS 1.10 and Data Areas

[Linda Mooney]

Hi Peter,

My $0.02.  So many manuals, so little time, so much we need to know.  Book and 
PDF have different strengths.  Book is stellar at search, with the search 
across collections being, for me, a critically useful feature.  Research and 
digging info out of the docs would be so much more painful without it.  There 
is no way I would want to face a page full of PDFs and have to search through 
all of them on any specific subject.  When I am ready to print or ready to 
copy/paste into a local how-to or procedure, etc. I use PDF, for those are its 
strengths.  I need and use both formats for different parts of the same 
purpose.  

Thanks,

Linda Mooney
-- Original message -- 
From: Peter Relson <[EMAIL PROTECTED]> 

> As Bob Shannon noted, the initial z/OS 1.10 bookshelf did not have the data 
> areas books. IBM developers (and perhaps others) complained and apparently 
> now they do. There is no indication that this will necessarily continue in 
> subsequent releases. 
> 
> I would still appreciate your comments regarding "PDF-only" even if it 
> applies only to potential subsequent releases. 
> 
> Peter Relson 
> z/OS Core Technology Design 
> -- 
> 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

Re: CPU time/instruction table

[Ted MacNEIL]

>A similar set of benchmarks could and should be used by IBM for processors, to 
>give management an idea of what they're getting.

They exist.
It's called LSPR (suspect) and its inadequate aid zPCR.
-
Too busy driving to stop for gas!

--
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

Re: CPU time/instruction table

[Gerhard Postpischil]

Edward Jaffe wrote:
The first several pages would have to explain the pipeline design and 
the delays that can occur. After that, specific instruction timings 
would have to assume the pipeline is flowing freely.


Management doesn't care about individual instruction timings, 
but the overall throughput and response time. The service 
bureaus I worked at had a standard job stream, with matching 
initiator setup, that was used to determine new processor 
performance and rates for CPU, I/O, etc. These days I'd add some 
transaction processing to that. A similar set of benchmarks 
could and should be used by IBM for processors, to give 
management an idea of what they're getting. I realize there are 
different ways of configuring and capping multiple processors, 
but several reasonable combinations should be adequate to 
determine "management throughput" 


Gerhard Postpischil
Bradford, VT

--
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

Re: CPU time/instruction table

[Edward Jaffe]

Alan Altmark wrote:
On Mon, 1 Dec 2008 14:24:11 -0500, Shmuel Metz (Seymour J.) 
<[EMAIL PROTECTED]> wrote:
  

Were IBM to document it, the timing manual for any of the current
processors would be immense.



Nah.  Page 1:  "It depends".  Page 2 would be intentionally left blank.

With instruction pipelining, too much depends on what has happened *before* 
the instruction in question.  From a raw technology perspective, changes in 
cache design and memory structure can affect an instruction as well (for good 
or ill).
  


The first several pages would have to explain the pipeline design and 
the delays that can occur. After that, specific instruction timings 
would have to assume the pipeline is flowing freely.


--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/

--
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

Manual formats

[Gibney, Dave]

   As related many times by many posters, each has it's better
qualities. PDF is good for printing, fair for straight reading, lousy
for jumping around in. Bookmanager for searching and skipping around in,
lousy for printing. 
   I don't use the HTML much, but if that's all I can reach, I'll use it
in a pinch :)

 
> I would still appreciate your comments regarding "PDF-only" even if it
> applies only to potential subsequent releases.

--
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

Re: CPU time/instruction table

[Alan Altmark]

On Mon, 1 Dec 2008 14:24:11 -0500, Shmuel Metz (Seymour J.) 
<[EMAIL PROTECTED]> wrote:
>Were IBM to document it, the timing manual for any of the current
>processors would be immense.

Nah.  Page 1:  "It depends".  Page 2 would be intentionally left blank.

With instruction pipelining, too much depends on what has happened *before* 
the instruction in question.  From a raw technology perspective, changes in 
cache design and memory structure can affect an instruction as well (for good 
or ill).

As was said, write your programs for readability and maintainability.  Follow 
traditional rules about operand alignment.  Do try to use RI-format 
instructions like LOAD HALFWORD IMMEDIATE to avoid storage reference or 
address generation.

Alan Altmark
z/VM Development
IBM

--
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

Re: z/OS 1.10 and Data Areas

[Rick Fochtman]

---
As Bob Shannon noted, the initial z/OS 1.10 bookshelf did not have the 
data areas books. IBM developers (and perhaps others) complained and 
apparently now they do. There is no indication that this will 
necessarily continue in subsequent releases.


I would still appreciate your comments regarding "PDF-only" even if it 
applies only to potential subsequent releases.

--

Peter, I much prefer the BookManager form for all manuals. The search 
engine is far better, even if it does often provide "hits" that aren't 
really relevant to what I need.


--
Rick
--
Remember that if you’re not the lead dog, the view never changes.

--
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

Re: Merging systems into a sysplex.

[Arthur Gutowski]

On Wed, 26 Nov 2008 12:54:30 -0600, Mark Zelden 
<[EMAIL PROTECTED]> wrote:

>You don't have to share DAE at all do you?  [...]
>
>So can't you set up DAE as not shared and have each system use
>its own  (I realize you lose the sysplex wide benefit of DAE)?
>
>IOW, don't include SHARE in your ADYSETxx parms.  It was a long time
>ago so I can't look back, but I thought that is what I did.

You're spot on.  4 months seems like forever anymore.  For whatever reason, 
we retained DAE sharing on two of our sysplexes that merged into the 
aggregate, so we're getting the occasional nastygram on opposing "sides" (the 
aggregate spans buildings).  DAE recording and suppression seems to be 
working fine.  We left the former monplexes non-shared for now.  Merging to a 
single DAE will be tricky given our mirroring and DR setup.

Regards,
Art Gutowski
Ford Motor Company

--
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

My Understanding of Concurrent TCP/IP MVS program

[Joe Reichman]

Hi.

 I want to thank all those who helped me understand this concept

from my understanding the main pourpose in the Select API is to inform the main 
Task

That there are one or more I/O requests pending


At that point the main task should take this number which is typicall returned  
in the
retocde field load into a BCT register (I guess my native tounge is Assembler)

and loop processing all the request(s)

 The First function in the loop is to do a Accept the output of this which is 
socket
address sturcture will make it clear which Client is ready to either read or 
Write

 Then the main Task does a GiveSocket

 Post the appropriate Subtask


 Seems like one of diffrence between IBM implementation of SOCKETS and Lets say 
Windows is
that in Windows the Select api is an array of sockets

 While IBM it is a bitstring

 In Addtiion Windows Does doesn't have the Give/take/Socket api

--
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

Definition of a Shannon (was Big Sorts)

[john gilmore]

6.022[1415] x 10^23 in this putative definition of a Shannon is Avogadro's 
number, a count of the number of atoms in 12 grams of isotopically pure 
carbon-12.   Claude Shannon certainly deserves to be memorialized in some such 
way; but an analogous count of the number of atoms in 28 grams of isotopically 
pure silicon-28 would have more, if still only tenuous, relevance to what we do.
 
John Gilmore Ashland, MA 01721-1817 USA
_
Color coding for safety: Windows Live Hotmail alerts you to suspicious email.
http://windowslive.com/Explore/Hotmail?ocid=TXT_TAGLM_WL_hotmail_acq_safety_112008
 
--
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

Dec 3 Live Virtual Class - Big Blue Goes Green: Update on Project Big Green

[Pamela Christina in snow-free Endicott NY]

Here's a pdf flyer about the lvc/webcast in case you want it...
http://www.vm.ibm.com/education/lvc/lvcf1203.pdf

And you can find links on the VM events calendar
http://www.vm.ibm.com/events/

and on the education/lvc page...
http://www.vm.ibm.com/education/lvc/

Regards,
Pam C

--
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

Dec 3 Live Virtual Class - Big Blue Goes Green: Update on Project Big Green

[Pamela Christina in snow-free Endicott NY]

Cross-posted to IBMVM, IBMMAIN, and Linux390 listservs for
those who who are interested in these complimentary web casts
about System z.

There is no charge to participate in this technical education session

What's new and different about this LVC ?
 1 - This is the first System z LVC that will be delivered using the
 Adobe Connect Pro which does not require a client code download.
 You just need a web browser and the Adobe Flashù Player runtime
 software.  You register and create a userid and password that you
 remember to attend future webcasts.
 If you recall the past LVCs required you to do a client
 download, and there was not a choice for Linux workstations.
 We hope this new vehicle works well for you.  If not, no doubt
 you'll let us know :-)

 2 - You have a choice of times (so that you can ask questions
 during a live broadcast).  Replay will also be available.

Topic:Big Blue Goes Green: Update on Project Big Green

Speaker:  Bill Reeder, IBM System z Sales and Strategy; Americas Group

Date: Wednesday, December 3

Times:11:00 AM ET or 4:00 PM ET

Duration:  75 Minutes

Abstract:
On August 1; 2007 IBM announced that its Project Big Green would spur a
global shift to Linux on the mainframe. Further; IBM claimed to
double computing capacity with no increase in consumption or impact
by 2010. And; as if that were not ambitious enough; IBM also announced
that it would consolidate 3; 900 of its own distributed servers to
33 mainframe computers in 3 years.

Listen to how IBM's efforts will save 80% on energy consumption; while
realizing significant savings in labor and software costs.Bill will
highlight our journey to date as well as the solutions that we
have used or developed to realize success thus far; with more to come.

How to attend:
Register for the time of the webcast that you plan to attend:

December 3, 2008
11:00 AM Eastern U.S. (New York)  / 10:00 AM CT / 8:00AM PT / 4:00 PM London/ 
5:00 PM Central Europe (Paris)

http://ibmstg.na3.acrobat.com/systemzt1203e/event/registration.html

OR

December 3, 2008
4:00 PM Eastern U.S. (New York) /  3:00 PM CT / 1:00 PM PT/  9:00 PM London 
/10:00 PM  Central Europe (Paris)

http://ibmstg.na3.acrobat.com/systemzt1204e/event/registration.html

Reminder:
This is the first Linux on System z webcast using Adobe. Adobe Connect Pro
does not require any client code to be downloaded.
You just need a web browser and the Adobe Flashù Player runtime software.
You will only need to register one time for Adobe webcasts - where you will
create a userid and password.

Please remember your userid and password to register and attend upcoming
Adobe webcasts.


Confirmation:  After completing your pre-registration for this webcast, you
will receive an e-mail invitation from Darla Erdmann that
includes the URL to attend the webcast.   You will need to use the userid
and password you created to access the live session.


Test Adobe Connection:  Please test your connection to ensure you are ready
to attend this event by clicking on this URL:
https://admin.adobe.acrobat.com/common/help/en/support/meeting_test.htm.

Audio Delivery for Webcast:.  After connecting to the webcast on December
3, you will be prompted for a callback telephone number,
ie. country code + area/city code + phone number, so that Adobe Connect can
"call" you for the audio portion of the webcast.
   (Audio for this webcast is being provided via phone only).

Replay:
This session will be recorded and a replay is planned to
be available on December 5, 2008.

Webcast Help:  The Adobe Connect webcast Helpdesk can be contacted by
calling +1-877-239-6378  OR  +1-719-944-1681.
If you've never used Connect Pro, get a quick overview:
http://www.adobe.com/go/connectpro_overview



Replays: As a reminder, the 2008 z/VM, Linux on System z and z/VSE sessions
are available for replay from the z/VM website at:
http://www.vm.ibm.com/education/lvc

The Project Big Green presentation handout is planned to be posted
on the z/VM website on Tuesday Dec 2nd.

Please direct any questions to Julie Liesenfelt at [EMAIL PROTECTED]
Please direct any questions to Julie Liesenfelt at [EMAIL PROTECTED]



Regards,
Pam C

--
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

Re: TCP/IP

[Patrick O'Keefe]

On Mon, 1 Dec 2008 17:23:36 -0600, Paul Gilmartin <[EMAIL PROTECTED]> wrote:

>...
>I'd call fork() standard and spawn() idiosyncratic.
>...

Actually, the OP didn't even mention a Unix environment.  He just asked
about TCBs.   Fork() and spawn() might not be involved at all.

Pat O'Keefe

--
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

Re: SYSLOGD and Assembler

[John McKown]

I am looking at the source code to the Linux "logger" command. Note this is
not the same as the z/OS UNIX "logger" command at all!

>From what I can see, the command simply sends a formatted message to UDP
port 514 (as the default). This consists of the "priority", which is a
number enclosed in <> and defined in syslog.h. That is followed immediately
by a string which is the "tag". That is followed by a space, then the
"userid" (getlogin/RACF id) of the process writing the record. Then a colon
followed by the PID (process id) of the process writing the record. Lastly a
space followed by the free-form message.

This assumes that you have set your SYSLOGD up to listen on udp port 514. I
think that there is a switch that you need to specify in order to tell
SYSLOGD to do this. Otherwise, you need to use the AF_UNIX interface to talk
(default) to /dev/log.

The above is "as best as I can tell" from looking at the Linux source. I
cannot be sure the z/OS UNIX works the same way, but I would really hope
that it does.

--
John

--
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

Re: z/OS 1.10 and Data Areas

[Peter Relson]

As Bob Shannon noted, the initial z/OS 1.10 bookshelf did not have the data
areas books. IBM developers (and perhaps others) complained and apparently
now they do. There is no indication that this will necessarily continue in
subsequent releases.

I would still appreciate your comments regarding "PDF-only" even if it
applies only to potential subsequent releases.

Peter Relson
z/OS Core Technology Design
--
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

Re: Big sorts

[Shane Ginnane]

And I was intrigued to learn "mole" is (now) a SI base unit. When I were a 
lad I only recall its usage in relation to Chemistry; a mole of atoms or 
molecules for example.
How things change ...

Shaene ...

Jack wrote on 02/12/2008 02:25:27 AM:

> I had never previously seen the term "Shannon", defined to mean 1 mole 
of 
> bits (6.02 x 10^23 bits).

--
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

Re: TCP/IP

[Paul Gilmartin]

On Mon, 1 Dec 2008 15:16:59 -0600, Patrick O'Keefe wrote:
>
>I know nothing about this vis-a-vis fork() and spawn(), but
>givesocket() / takesocket() are (I assume) standard socket calls
>
A Google search for GIVESOCKET returns a preponderance
of IBM-centric hits.  So "standard" depends on the
parameters of your universe of discourse.

I'd call fork() standard and spawn() idiosyncratic.

-- gil

--
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

Re: IBM DFRMM Reporting

[Mike Wood]

Jim, rmm has multiple ways to egenrate information for you, depending what 
exactly you are after; The simplest is simply to use the ISPF dialog and its 
search capabilities and view the results interactively.

Then there is Rexx; if you want something programmed and need to query 
results and issue more commands - go that way

There are built-in reports; movement and inventory reports using supplied 
utilities.

Then there are a set of Rexx based reports - easily customized, that provide 
similar reports to what many are used to.

For customized reporting look to the rmm Report Generator - an ISPF based 
dialog that shows you record layouts and lets you pick the fields to be 
selected,columns to be sorted and the report layout. That generates you the 
batch JCL to use ICETOOL or SYNCTOOL.  It is easy.

For more statistics and so on, look to IBM Omegamon XE for Storage and its 
ability to pull information from rmm and produce charts etc.

Mike  Wood   RMM Development

--
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

Re: CPU time/instruction table

[Edward Jaffe]

Shmuel Metz (Seymour J.) wrote:

On the current machines, there are two classes of instructions.



AFAIK, there are three: hardwired, microcoded and millicoded.
  


No more microcode since millicode took over circa 9672-G4?

--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/

--
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

Re: is there anyp rogram can read all user catalog from master catalog

[Mike Wood]

On Fri, 28 Nov 2008 04:21:04 -0500, Gerhard Postpischil 
<[EMAIL PROTECTED]> wrote:

>Tommy Tsui wrote:
>> Actually I want to seach all volumes entry catalog for tape volume only
>
>I'm not sure there is a simple way of doing this.
>

No, it really is simple; dont orientate to a specific catalog, specify the 
CSIDTYP correctly for the records you want, 'W', and specify a filter for the 
entries you want - all start with 'V', then specify the volser or * for all

CSI will go through all the connected volume catalogs and return you all 
matching volume entries.

Mike Wood   RMM Development

--
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

Re: SYSLOGD and Assembler

[Big Iron]

There is a C API (openlog, syslog, closelog, I think) which is described in
the C Run-time library publication; it should be possible to invoke that
from LE-enabled assembler code with appropriately constructed
arguments. Apparently, you can also write to a socket to send info to
syslogd. Most of the information about syslogd on z/OS now lives in the
z/OS Communication Server documentation.

Bill

On Mon, 1 Dec 2008 15:03:01 -0600, Roland Schiradin <[EMAIL PROTECTED]> wrote:

>I received the following suggestion for SHOWzOS but I have no idea how it
>can be done. Can I use the SYSLOGD stuff from Assembler? Is it just a simple
>TCP/IP connection or an API?
>
>I want to thank you for all the work you've done on ShowMVS/ShowZos.  It is
>one of my "go to" utilities, and I appreciate everyone's efforts to make it
>better.  I wanted to offer an idea for a possible suggestion.  I believe that
>most of us have syslogd running on various systems, and by default, listening
>for UDP on port 514.  Would you consider an option, similar to the TCP one, to
>send in syslog format?
>
>Roland
>

--
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

Re: TCP/IP

[Ron Wells]

anyone ever recv following msg after startup of NJE/IP link ??

IAZ0543I NETSRV6 TCP/IP connection with IP Addr: 10.239.53.121 Port: 175 
 Initiated 
IAZ0543I NETSRV6 TCP/IP connection with IP Addr: 10.239.53.121 Port: 175 
 ended due to subtask goaway post 
<<

Re: Processor model in HCD

[Ted MacNEIL]

>Since the processor model doesn't change (it is still 2094), do we need to 
>define our processors as S38 in HCD? Does this definition have any effect on 
>CPU's hw definition or performance?  Has anyone experienced the 
>same situation? 

IIRC, any processor 'definitions' are just comments.
You get your performance/capacity based on the processor size/number of CP's 
and the LPAR weight.

-
Too busy driving to stop for gas!

--
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

Re: TCP/IP

[David Logan]

If it's TCBs in the same address space, you don't have to go through any
official mechanism to transfer a socket. Simply having a subtask pick the
socket value up out of storage somewhere is fine.


David Logan
Manager of Product Development, Pitney Bowes Business Insight
http://centrus.com
W: (720) 564-3056
C: (303) 818-8222

-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On Behalf
Of Patrick O'Keefe
Sent: Monday, December 01, 2008 14:17
To: IBM-MAIN@bama.ua.edu
Subject: Re: TCP/IP

On Sun, 30 Nov 2008 22:45:37 -0600, Paul Gilmartin 
<[EMAIL PROTECTED]> wrote:

>On Sun, 30 Nov 2008 19:34:41 -0600, John McKown wrote:
>>
>>Quite often, in the UNIX world, what is done is that the main process
>>(task) will do all of this. It will then fork() or spawn() another process
>>and give ownership of the socket created above to that process. 
>>...
>>
>Isn't the statement "give ownership of the socket" superfluous? I
>think a socket is just a variety of descriptor which fork()
>automatically replicated in the child process space.  What may be
>important is that the parent close() its instance of that descriptor
>in order that the client can detect the close() by the child.
>...

I know nothing about this vis-a-vis fork() and spawn(), but
givesocket() / takesocket() are (I assume) standard socket calls
that pass ownership of a connection.  In z/OS that passing can be
between address spaces.   Websphere passes connections from 
Server to Servant (or whatever they are called) address spaces.
I assume it can be done between TCBs in the same address space.

I certainly could be wrong, but I think the connection stays active
during this process.

Pat O'Keefe

--
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

Re: Startio Question

[Shmuel Metz (Seymour J.)]

In <[EMAIL PROTECTED]>, on
11/22/2008
   at 01:01 AM, Lindy Mayfield <[EMAIL PROTECTED]> said:

>Well, this was one of the first IBM related things that popped up in
>Google: http://www.freepatentsonline.com/6453277.html

It looks like a defensive patent; I would expect that the defendant in an
infringement suit would allege prior art and that it was obvious to a
practitioner.
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
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

Re: sending snmp traps from z/OS

[Patrick O'Keefe]

On Mon, 1 Dec 2008 09:36:45 -0500, Michael Schmutzok 
<[EMAIL PROTECTED]> wrote:

>...   A subagent can use DPI to generate a trap.  
>
>Isn't that what the OP was asking how to do? Yes, this requires
>the standard MVS SNMP agent to be up and running.

My hesitation was over the word "subagent".   I wasn't sure that
DPI could be used without a lot of extra subagent bagage, but I 
think you said it can.

>
>> However, (if I understand the term manager) it sounds like this 
>> really should be a SNMP manager function.  ...
>
>> z/OS Communications Server also provides an extension of the 
>> SNMP Manager API, the SNMP Notification API ... to send 
>> notifications to SNMP agents and/or SNMP Notification
>> Receivers. Available notifications include Informs and both 
>> Version 1 and Version 2 Traps.  
>
>> I think this API is really the way to go if you have 1.9 or later. 
>
>The OP wanted to send traps, not receive them.

The doc I quoted says "send".  (It talks about receiving elsewhere.)

Pat O'Keefe
 

--
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

Re: TCP/IP

[Patrick O'Keefe]

On Sun, 30 Nov 2008 22:45:37 -0600, Paul Gilmartin 
<[EMAIL PROTECTED]> wrote:

>On Sun, 30 Nov 2008 19:34:41 -0600, John McKown wrote:
>>
>>Quite often, in the UNIX world, what is done is that the main process
>>(task) will do all of this. It will then fork() or spawn() another process
>>and give ownership of the socket created above to that process. 
>>...
>>
>Isn't the statement "give ownership of the socket" superfluous? I
>think a socket is just a variety of descriptor which fork()
>automatically replicated in the child process space.  What may be
>important is that the parent close() its instance of that descriptor
>in order that the client can detect the close() by the child.
>...

I know nothing about this vis-a-vis fork() and spawn(), but
givesocket() / takesocket() are (I assume) standard socket calls
that pass ownership of a connection.  In z/OS that passing can be
between address spaces.   Websphere passes connections from 
Server to Servant (or whatever they are called) address spaces.
I assume it can be done between TCBs in the same address space.

I certainly could be wrong, but I think the connection stays active
during this process.

Pat O'Keefe

--
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

Processor model in HCD

[Mursel Tasgin]

Hi,
We plan to upgrade our 2094 S28 machine to S38 (via adding book on the same 
box). We try to figure out what changes will occur in our definitions, 
parameters etc. Other than some software licences(ie: # of CPUs), the only  
place we have coded "S28" is HCD's "Processors" section.
Since the processor model doesn't change (it is still 2094), do we need to 
define our processors as S38 in HCD? Does this definition have any effect on 
CPU's hw definition or performance?  Has anyone experienced the same situation? 

Thanks and regards.
Mursel Tasgin
Akbank




--
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

SYSLOGD and Assembler

[Roland Schiradin]

I received the following suggestion for SHOWzOS but I have no idea how it 
can be done. Can I use the SYSLOGD stuff from Assembler? Is it just a simple 
TCP/IP connection or an API?

I want to thank you for all the work you've done on ShowMVS/ShowZos.  It is 
one of my "go to" utilities, and I appreciate everyone's efforts to make it 
better.  I wanted to offer an idea for a possible suggestion.  I believe that 
most of us have syslogd running on various systems, and by default, listening 
for UDP on port 514.  Would you consider an option, similar to the TCP one, to 
send in syslog format?

Roland

--
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

Re: SMPE EOS Question

[Thompson, Steve]

Thanks. I had searched and scanned in all the places I could think of
and just couldn't find this.

Regards,
Steve Thompson

-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Jerry Fuchs
Sent: Monday, December 01, 2008 1:49 PM
To: IBM-MAIN@bama.ua.edu
Subject: Re: SMPE EOS Question

It is not in SMPE

See:

www-306.ibm.com/software/support/lifecycle/




--
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

Re: Simple IEFBR14 proc

[Shmuel Metz (Seymour J.)]

In <[EMAIL PROTECTED]>, on
11/21/2008
   at 02:37 PM, Howard Brazee <[EMAIL PROTECTED]> said:

>And the proc looks like this:

> 

>//TEBP   EXEC PGM=IEFBR14,

>// USER1=NULLFILE   

Your keyword defaults must go on the PROC statement, not on the EXEC
statement.
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
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

Re: CPU time/instruction table

[Shmuel Metz (Seymour J.)]

In <[EMAIL PROTECTED]>, on 12/01/2008
   at 05:39 PM, Miklos Szigetvari <[EMAIL PROTECTED]>
said:

>Long long time ago there has been this CPU time/instruction tables.

That was when processors were simpler and the timing formulae shorter,
involving fewer variables.

>Something like this for a z9 ?

No, nor would I expect it for anything but a RISC.
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
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

Re: Servicelink view format - was: Re: IBMLink planned outage

[Shmuel Metz (Seymour J.)]

In <[EMAIL PROTECTED]>, on 11/17/2008
   at 06:28 PM, Bobbie Justice <[EMAIL PROTECTED]> said:

>ditto, me neither. I also submitted a feedback on this issue. I'm
>wondering  who exactly requested that change? .

I'm wondering what they actually requested.
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
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

Re: CPU time/instruction table

[Shmuel Metz (Seymour J.)]

In <[EMAIL PROTECTED]>, on 12/01/2008
   at 11:20 AM, John McKown <[EMAIL PROTECTED]> said:

>I know that some of the people on the ASSEMBLER list have done
>instruction timings for some specific instructions. But they are going to
>vary based on the machine. So, short of coding up each routine and timing
>it yourself, I don't know of any way to do this. IBM simply does not
>document this any more.

Were IBM to document it, the timing manual for any of the current
processors would be immense.

>On the current machines, there are two classes of instructions.

AFAIK, there are three: hardwired, microcoded and millicoded.
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
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

Re: SMPE EOS Question

[Jerry Fuchs]

It is not in SMPE

See:

www-306.ibm.com/software/support/lifecycle/





"Thompson, Steve" <[EMAIL PROTECTED]> 
Sent by: IBM Mainframe Discussion List 
12/01/2008 02:45 PM
Please respond to
IBM Mainframe Discussion List 


To
IBM-MAIN@bama.ua.edu
cc

Subject
SMPE EOS Question






Does anyone know where SMP/E lists the End of Service/Life for its
releases?

Regards,
Steve Thompson

--
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

SMPE EOS Question

[Thompson, Steve]

Does anyone know where SMP/E lists the End of Service/Life for its
releases?

Regards,
Steve Thompson

--
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

Re: CPU time/instruction table

[(IBM Mainframe Discussion List)]

 
 
In a message dated 12/1/2008 11:53:31 A.M. Central Standard Time,  
[EMAIL PROTECTED] writes:

Thank  you very much, I will try to messure


There is a wealth of info on this topic already on the Archives,  q.v.  
Discussions range from why it is no longer meaningful to some serious  
benchmark 
results.
 
Bill  Fairchild
Rocket Software
**Finally, one site has it all: your friends, your email, your 
favorite sites. Try the NEW AOL.com. 
(http://www.aol.com/?optin=new-dp&icid=aolcom40vanity&ncid=emlcntaolcom0006)

--
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

Re: Big sorts

[(IBM Mainframe Discussion List)]

 
 
In a message dated 12/1/2008 10:26:13 A.M. Central Standard Time,  
[EMAIL PROTECTED] writes:
>I had never previously seen the term "Shannon", defined to mean 1 mole of  
bits (6.02 x 10^23 bits).
 
Since the title of the linked article includes the word "Information", I  
would assume that "Shannon" refers to the late Claude Elwood Shannon, who did  
seminal research and published papers on Information Theory.  
_http://en.wikipedia.org/wiki/Claude_Shannon_ 
(http://en.wikipedia.org/wiki/Claude_Shannon) 


 
Bill  Fairchild
Rocket Software
**Finally, one site has it all: your friends, your email, your 
favorite sites. Try the NEW AOL.com. 
(http://www.aol.com/?optin=new-dp&icid=aolcom40vanity&ncid=emlcntaolcom0006)

--
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

Re: is there anyp rogram can read all user catalog from master catalog

[Peter Ten Eyck]

If you can get a copy, the IBM program MCNVTCAT will read through 
IDCAMS listcat output. This program is very handy to make mass 
changes to a catalog.

--
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

Re: IBM DFRMM Reporting

[John Kelly]

and of course there is always Mr. Sort's (Frank Yeager) various ICETOOL 
reports including RMM. It maps the RMM Inventory Management Activity File 
Record and then does a dozen reports. 


Jack Kelly
202-502-2390 (Office)

--
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

Re: IBM DFRMM Reporting

[John McKown]

>From what little I know (we don't use DFRMM), the way to write reports is to
use REXX. There are a number of REXX functions which retrieve DFRMM
information into REXX variables. You then go from there.

--
John

--
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

Re: CPU time/instruction table

[Miklos Szigetvari]

Hi

Thank you very much, I will try to messure

John McKown wrote:


On Mon, 1 Dec 2008 17:59:32 +0100, Miklos Szigetvari
<[EMAIL PROTECTED]> wrote:

 


Hi

I just want to get a feeling about the  speed , if I  change an
unconditional  modulo  i.e DL (divide logical)  instruction
to a  logic before call the DL .

   



I know that some of the people on the ASSEMBLER list have done instruction
timings for some specific instructions. But they are going to vary based on
the machine. So, short of coding up each routine and timing it yourself, I
don't know of any way to do this. IBM simply does not document this any more.

On the current machines, there are two classes of instructions. The "simple"
instructions (like SR and LA and so forth) are "hard wired". The more
difficult instructions (such as MVCLE) are "millicoded". So, if you can
replace a millicoded instruction with a small number of "hard wired"
instructions, then you should be better off, speed wise. However, once
again, IBM does not document which instructions are "hard coded" and which
are millicode. I have heard that IBM has, at times, "fixed" a "broken" hard
coded instruction by replacing it with a millicoded instruction installed by
the CE. 


Wish I could be of more help.

The ASSEMBLER list is [EMAIL PROTECTED] You'd need to
subscribe by sending mail to mailto:[EMAIL PROTECTED] and
subscribing to ASSEMBLER-LIST.

--
John

--
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


 



--
Miklos Szigetvari

Development Team
ISIS Information Systems Gmbh 
tel: (+43) 2236 27551 570
Fax: (+43) 2236 21081 

E-mail: [EMAIL PROTECTED] 

Info: [EMAIL PROTECTED] 
Hotline: +43-2236-27551-111 

Visit our Website: http://www.isis-papyrus.com 
---

This e-mail is only intended for the recipient and not legally
binding. Unauthorised use, publication, reproduction or
disclosure of the content of this e-mail is not permitted.
This email has been checked for known viruses, but ISIS accepts
no responsibility for malicious or inappropriate content.
--- 


--
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

Re: CPU time/instruction table

[John McKown]

On Mon, 1 Dec 2008 17:59:32 +0100, Miklos Szigetvari
<[EMAIL PROTECTED]> wrote:

>Hi
>
>I just want to get a feeling about the  speed , if I  change an
>unconditional  modulo  i.e DL (divide logical)  instruction
>to a  logic before call the DL .
>

I know that some of the people on the ASSEMBLER list have done instruction
timings for some specific instructions. But they are going to vary based on
the machine. So, short of coding up each routine and timing it yourself, I
don't know of any way to do this. IBM simply does not document this any more.

On the current machines, there are two classes of instructions. The "simple"
instructions (like SR and LA and so forth) are "hard wired". The more
difficult instructions (such as MVCLE) are "millicoded". So, if you can
replace a millicoded instruction with a small number of "hard wired"
instructions, then you should be better off, speed wise. However, once
again, IBM does not document which instructions are "hard coded" and which
are millicode. I have heard that IBM has, at times, "fixed" a "broken" hard
coded instruction by replacing it with a millicoded instruction installed by
the CE. 

Wish I could be of more help.

The ASSEMBLER list is [EMAIL PROTECTED] You'd need to
subscribe by sending mail to mailto:[EMAIL PROTECTED] and
subscribing to ASSEMBLER-LIST.

--
John

--
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

IBM DFRMM Reporting

[Jim Marshall]

I was a long time user of the CA-1 Tape Management System (TMS) and 
remember the reporting capabilities (TMSGRW, etc) which gave me the chance 
to slice and dice the TMC info.  In the present position as "Coach" of the 
Sysprogs I have to rely on them to get me info I need. We run DFRMM and I 
am told there is no generalized tool for reporting RMM "stuff". 

Would appreciate any suggestions about what people use, techniques, or 
even Open Source Tools (aka, SHAREWARE) which could assist me. Yes, I 
could delve into the manuals and relive my techie days but I must leave it to 
the Contractors; besides there are many non-contractor issues to deal with 
besides getting ones hands dirty again. 

Comments appreciated.   jim 

--
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

Re: CPU time/instruction table

[Miklos Szigetvari]

Hi

I just want to get a feeling about the  speed , if I  change an 
unconditional  modulo  i.e DL (divide logical)  instruction

to a  logic before call the DL .

John McKown wrote:


On Mon, 1 Dec 2008 17:39:49 +0100, Miklos Szigetvari
<[EMAIL PROTECTED]> wrote:

 


Hi

Long long time ago there has been this CPU time/instruction tables.
Something like this for a z9 ?

--
Miklos Szigetvari
   



No. IBM not longer publishes such information as it is misleading. There are
simply too many things which are done which make the data non repeatable.
Even simple things such as branches. The branch prediction logic in modern
systems makes that non deterministic. Today, it is generally felt that it is
more important to write code that is reliable, understandable, and
maintainable than squeezing out the last microsecond of performance (e.g. SR
vs SLR vs XR to clear a register).

--
John

--
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


 



--
Miklos Szigetvari

Development Team
ISIS Information Systems Gmbh 
tel: (+43) 2236 27551 570
Fax: (+43) 2236 21081 

E-mail: [EMAIL PROTECTED] 

Info: [EMAIL PROTECTED] 
Hotline: +43-2236-27551-111 

Visit our Website: http://www.isis-papyrus.com 
---

This e-mail is only intended for the recipient and not legally
binding. Unauthorised use, publication, reproduction or
disclosure of the content of this e-mail is not permitted.
This email has been checked for known viruses, but ISIS accepts
no responsibility for malicious or inappropriate content.
--- 


--
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

Re: CPU time/instruction table

[John McKown]

On Mon, 1 Dec 2008 17:39:49 +0100, Miklos Szigetvari
<[EMAIL PROTECTED]> wrote:

>Hi
>
>Long long time ago there has been this CPU time/instruction tables.
>Something like this for a z9 ?
>
>--
>Miklos Szigetvari

No. IBM not longer publishes such information as it is misleading. There are
simply too many things which are done which make the data non repeatable.
Even simple things such as branches. The branch prediction logic in modern
systems makes that non deterministic. Today, it is generally felt that it is
more important to write code that is reliable, understandable, and
maintainable than squeezing out the last microsecond of performance (e.g. SR
vs SLR vs XR to clear a register).

--
John

--
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

CPU time/instruction table

[Miklos Szigetvari]

Hi

Long long time ago there has been this CPU time/instruction tables.
Something like this for a z9 ?

--
Miklos Szigetvari

Development Team
ISIS Information Systems Gmbh 
tel: (+43) 2236 27551 570
Fax: (+43) 2236 21081 

E-mail: [EMAIL PROTECTED] 

Info: [EMAIL PROTECTED] 
Hotline: +43-2236-27551-111 

Visit our Website: http://www.isis-papyrus.com 
---

This e-mail is only intended for the recipient and not legally
binding. Unauthorised use, publication, reproduction or
disclosure of the content of this e-mail is not permitted.
This email has been checked for known viruses, but ISIS accepts
no responsibility for malicious or inappropriate content.
--- 


--
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

Big sorts

[Jack . Hamilton]

If you're interested in more details on the Google petabyte sort, 
discussed here recently (can't find the original subject line because 
Notes doesn't have a useful search feature), see:

 
http://scienceblogs.com/goodmath/2008/11/scale_how_large_quantities_of.php

Some of your objections to the article are addressed in the comments.

I had never previously seen the term "Shannon", defined to mean 1 mole of 
bits (6.02 x 10^23 bits).


--
Jack Hamilton
Management Information & Analysis - Analytic Information Services
Kaiser Foundation Health Plan, Inc.
1950 Franklin Street, Oakland, California 94612
+1 510 987-1556 (KP tieline 8-427-1556)

NOTE:  This email document and attachments are covered by CA Evidence Code 
§1157 and CA Health and Safety Code §1370.

NOTICE TO RECIPIENT:  If you are not the intended recipient of this 
e-mail, you are prohibited from sharing, copying, or otherwise using or 
disclosing its contents.  If you have received this e-mail in error, 
please notify the sender immediately by reply e-mail and permanently 
delete this e-mail and any attachments without reading, forwarding or 
saving them.  Thank you.

--
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

Re: VSAM delete issue

[Peter Ten Eyck]

I was able to get this VSAM delete with the following job.

 //TSIDCDEL JOB CLASS=S  
//S1   EXEC PGM=IDCAMS  
//FILE1DD DISP=SHR,VOL=SER=B90311,UNIT=3390,
//SYSPRINT DD SYSOUT=*  
//SYSINDD * 
 DELETE  CICS.TEST.CMT.BATSTAT3 -   
 FILE(FILE1) -  
 CATALOG(CATALOG.ZOS.V17CT1)
//

Thanks, Peter

--
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

Re: TCP/IP

[John McKown]

On Mon, 1 Dec 2008 15:17:39 +0900, Timothy Sipples
<[EMAIL PROTECTED]> wrote:

>Paul Gilmartin writes:
>>"in the UNIX world"???  Isn't z/OS Unix?
>
>Yes, that's a very good point:
>
>http://www.opengroup.org/openbrand/register/brand3470.htm
>
>I tend to prepend the word "distributed" if I'm referring to other
>(non-z/OS) UNIXes (UNIces?), and that serves pretty well. Or you could say
>"the non-z/OS UNIX world," or perhaps "the UNIX world (ex-z/OS)."
>
>- - - - -
>Timothy Sipples

OK, so maybe saying "in the UNIX world" was a bit off target. z/OS is UNIX
branded. I know that. But, how many people out there are using UNIX
facilities for __applications__? For example, which is more prevalent:
ATTACHing a subtask, or using pThreads? If you are "thinking UNIX", then you
use pThreads. If you are "thinking z/OS", then you use ATTACH. How many
programs are designed to take either z/OS datasets or UNIX files (ignoring
the QSAM compatibility interface)? How many out there use "cron" to schedule
repeating tasks? What about "at" for a "single shot"? That's kind of what I
was getting at. At least in mindset, z/OS and UNIX are still separated in
the z/OS community. At least as best as I can tell.

Personal opinion only, of course.

--
John

--
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

Re: searching all members of a pds using wildcards

[Shmuel Metz (Seymour J.)]

In <[EMAIL PROTECTED]>, on
11/30/2008
   at 07:07 PM, John McKown <[EMAIL PROTECTED]> said:

>>From my testing, the port of GNU grep will work on a PDS, but the standard
>IBM supplied grep will not.

"Standards are wonderful thing; everyone should have one of his very own."
(GM)

>Do you know if IBM grep
>will work with a PDS on later releases of z/OS? And, will IBM awk work,
>or will I need a port of GNU awk?

Don't know.
 
-- 
 Shmuel (Seymour J.) Metz, SysProg and JOAT
 ISO position; see  
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)

--
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

Re: sending snmp traps from z/OS

[Michael Schmutzok]

It can be done but it does require a lot of research and you get to become real 
familiar with the SNMP RFC's. We've used the DPI to create a sub-agent, 
register it with the agent (SNMP), send SNMP traps (also using TRAPFWD) and 
handle GET requests from other platforms. 

>>> Don Poitras <[EMAIL PROTECTED]> 11/26/2008 11:29 AM >>>
In article <[EMAIL PROTECTED]> you wrote:
> On Tue, 25 Nov 2008 09:23:10 -0500, Don Poitras 
> <[EMAIL PROTECTED]> wrote:

> >...
> >If you're willing to write some code, you can use DPI to issue the
> >traps. See:
> >
> >http://publibz.boulder.ibm.com/cgi- 
> bin/bookmgr_OS390/BOOKS/ezap4002/5.0
> >...

> I hadn't looked at DPI for several years so I had to refesh my
> memory.   I could have easily overlooked something, but as near 
> as I can tell, DPI is primarily for communication between agent and
> subagent.   A subagent can use DPI to generate a trap.  

Isn't that what the OP was asking how to do? Yes, this requires
the standard MVS SNMP agent to be up and running.

> However, (if I understand the term manager) it sounds like this 
> really should be a SNMP manager function.  Sort of a manager to
> manager process (which may be outside of official SNMP use.)  I 
> see in a z/OS 1.9 manual that there is an SNMP Manager API 
> available in 1.9.  I'm stuck on 1.8 so I can't check this out, but I 
> saw a "What's new" doc on the web that says:

> z/OS Communications Server also provides an extension of the 
> SNMP Manager API, the SNMP Notification API, which
> leverages the functionality of the SNMP Manager API to send 
> notifications to SNMP agents and/or SNMP Notification
> Receivers. Available notifications include Informs and both 
> Version 1 and Version 2 Traps.  

> I think this API is really the way to go if you have 1.9 or later. 

The OP wanted to send traps, not receive them.

> I've never looked into "Informs", but I gather they are sort of
> "guaranteed delivery" traps.
> Pat O'Keefe

-- 
Don Poitras - SAS Development  -  SAS Institute Inc. - SAS Campus Drive
[EMAIL PROTECTED]   (919) 531-5637Cary, NC 27513

--
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

Re: is there anyp rogram can read all user catalog from master catalog

[Dick de Groot]

Look at CBTTAPE.ORG for


//***FILE 542 is from Alastair Gray and contains some handy tools * FILE 542



-- 
Met vriendelijke groeten/With kind regards

Dick de Groot






2008/11/29 Tommy Tsui <[EMAIL PROTECTED]>

> tha't great thanks all information
>
> On Fri, Nov 28, 2008 at 9:59 PM, Lizette Koehler
> <[EMAIL PROTECTED]> wrote:
> > Tommy,
> >
> > Have looked at the STORAGE Tools that are in the
> > ftp://ftp.software.ibm.com/storage/tapetool/
> > Directory?  There might be some examples there.  VOLLIST, IBMTOOLS etc...
> >
> > Lizette
> >
> >
> >>
> >> I am looking a program which can read all user catalog information
> >> from master cataloganyone have an idea?
> >> I am written a tape managment system which will find all tape
> >> information from user catalog.
> >>
> >>
> >> any help will be appreciated..
> >>
> >
> > --
> > 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
>



-- 
Met vriendelijke groeten/With kind regards

Dick de Groot

--
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

Re: VSAM delete issue

[Faber, Hermann]

I'm not sure about the success, did you try a repro mergecat of that VSAM data 
set to the usercatalog where the alias points to?
Hermann

-Ursprüngliche Nachricht-
Von: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] Im Auftrag von 
Peter Ten Eyck
Gesendet: Montag, 1. Dezember 2008 14:40
An: IBM-MAIN@bama.ua.edu
Betreff: Re: VSAM delete issue

To answer a few questions...

This is a VSAM dataset on disk.

B90311 is the volser of an SMS volume.

The 3.4 screen is the ISPF option P.3.4 data set list utility in our shop.

I am not sure how the =3.4 lookup works when the master catalog has
an alias and a dataset that would qualify to use the alias is cataloged
in the master. Apparently it lists the dataset cataloged in the master.

In my original post you will see that I have used the catalog parameter
on my IDCAMS DELETE.

I am still trying to figure out how to delete this dataset without
deleting my alias.

Any help work be great, thanks.

--
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

Re: VSAM delete issue

[Peter Ten Eyck]

To answer a few questions...

This is a VSAM dataset on disk.

B90311 is the volser of an SMS volume.

The 3.4 screen is the ISPF option P.3.4 data set list utility in our shop.

I am not sure how the =3.4 lookup works when the master catalog has 
an alias and a dataset that would qualify to use the alias is cataloged 
in the master. Apparently it lists the dataset cataloged in the master. 

In my original post you will see that I have used the catalog parameter 
on my IDCAMS DELETE.

I am still trying to figure out how to delete this dataset without 
deleting my alias.

Any help work be great, thanks.

--
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

Re: BINDER instructions in source code?

[Jan MOEYERSONS]

On Fri, 28 Nov 2008 13:30:09 -0600, Frank Swarbrick 
<[EMAIL PROTECTED]> wrote:

>The good news, however, is if I do a *dynamic* call to CBLTDLI then it
>*does* work.  (Well, once I defined CBLTDLI to CICS as an assembler
>program.)

Dynamic call is really a Good Thing (tm). ;-)


>I see that both DSNCLI and DSNHLI are in SDFHLOAD as aliases for DFHD2LI,
>so it seems like as long as I used the DYNAM compile option for by DB2
>subroutines this should also work.

I am not 100% sure that will work, but it is sure worth a try. 


>I'll have to fool around a bit with EZASOKET.  I wonder if I copy EZACICAL to
>SDFHRESL and then create an alias EZASOKET that points to it...  Seems like
>that might work.  Can't hurt to try.

Drawback is that you will have to copy and create an alias yourself. Next time 
some maintenance comes along, you will have to remember that somewhere 
there is a copy... that you will have to update manually then.

Cheers,

Jantje.

--
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

Re: TCP/IP

[Hunkeler Peter (KIUK 3)]

>I think that your terminology may be a bit off. A socket is a single
>communications path in TCPIP. Normally, a server will first BIND to a
>port, then LISTEN on that port. This is not a socket. When a request 
>comes in, the code needs to ACCEPT. At this point you have an actual 
>socket.

I is my understanding that both client and server code first needs to 
obtain a socket using socket(). Servers then bind() the socket to a 
specific port and finally listen() for incoming connection requests. 

Clients connect() to the desired address, wacking up the server form 
the listen() call. The server then accepts the connection via accept() 
which creates a *new* socket reperesenting this very connection. The 
initial server side socket will be used to listen() again.

A concurrent sercer will spawn() a new process to handle the new
client request. The main process will close the socket returned by
accept() and will go back to listen().

A single threaded server will handle client request, eventually
close the socket (returned by accept()) and will the go back to 
listen().

-- 
Peter Hunkeler
Credit Suisse

--
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