Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Mainframe Based Calendar Engine

30 views
Skip to first unread message

David Alcock

unread,
Feb 19, 2000, 3:00:00 AM2/19/00
to
"Longnecker, Dennis" wrote:
> Anyone aware of a Calendaring Engine available on the OS/390 that is API
> callable?

If I were writing new web applications that did calendaring, I'd go
with Perl. I'm sure that you could find a lot of code on the web that
does this.

But if you want callable routines, LE/370 has a rich set of date
functions.

My assembler date routine, DA$DATE, has a function to return a month
array. It can be seen here:
http://users.ticnet.com/davea/mvs/freeware/da$wdate.htm

Longnecker, Dennis

unread,
Feb 19, 2000, 3:00:00 AM2/19/00
to
Dave -

Thanks for the information. We are looking for something more robust,
something along the lines of features available in Outlook where we can
check schedules of Judges, Courtrooms, Clerks, Police officers, etc. and
schedule them in. More than just date manipulations. We need it enterprise
wide, 100,000's of users.

Dennis

McKown, John

unread,
Feb 20, 2000, 3:00:00 AM2/20/00
to
Try looking at http://calendar.greathill.com . It is a Web base calendaring
system. It does not have S/390 port, but they say to call. It's written in
C++, at least according to the Web site

Also perhaps http://www.webevent.com - they say that they are a Perl CGI. If
so, it should work on OS/390 Unix.

I got pointers to these from http://freshmeat.net and used the search on
calendar

Jim Falgout

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
I would definitely think about Lotus Domino for S/390 with the amount of
users you are talking about. Built-in calendaring, E-mail, discussion
groups, etc. Jim

"Longnecker, Dennis" <Dennis.L...@COURTS.WA.GOV> wrote in message
news:9101A9448BE6D011AEC5...@oacsrv3.courts.wa.gov...
> We are developing new WEB applications to interface with our Green Screen
> applications. One of the needs is to create a calendaring function to
> schedule court cases. We are looking for someone who has developed a
> Calendaring Engine that runs on the OS/390 (either side is fine - OS/390
or
> USS).


>
> Anyone aware of a Calendaring Engine available on the OS/390 that is API

> callable? Vendors can e-mail me privately if they wish.
>
> Dennis Longnecker
> Office of the Administrator for the Courts
> (360) 705-5269
> dennis.l...@courts.wa.gov

Zulkifly Samad

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
Hi all,

Mainframe downsizing???

It is a new assignment for me. I have to study wheather our current
application running on OS/390 can be ported to new platform (unix, NT, or
OS/400). With the current size we have, I don't think that our application
can be handled by other than OS/390. I have to prove to the management, and
I have to write a paper about this topic.

Can anyone share any idea with me. Or help me to get any write-up or report
regarding this matter. Or any experience regarding this matter.

Thanks in advance.

McKown, John

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
With respect to Windows2000 (NT 5.0). Read this -
http://www.zdnet.com/sr/stories/column/0,4712,2439997,00.html
The author says, "No way for production!"

I can't help much with other platforms. My main concern would be sustainable
I/O rates. The S/390 can move data better than the best of the Intel/PC
systems.

Bill Becker

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
You need to be a little more specific as to what the application
is attempting to do? Is it a batch thing? Online? How many users?

Zulkifly Samad

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
We are running OS/390 Ver 1.2 with CICS/ESA Ver 4.1 and DB2 Ver 3.1.
CPU is CMOS 9672 - R24, with 512 Mbyte memory.

We have both online and batch application.

During online (day) - peak CPU utilization is 70%
During batch (night) - peak CPU utilization is 85%

Currently we have about 7,000 users - LU0 (majority) and LU2

Jim McAlpine

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
I would agree with John - I/O throughput is probably going to be the main
determining factor.

I would also say the potential downsized platforms will not have anywhere
near the same RAS as the S/390 CMOS machine.

A good place to start might be
http://www.s390.ibm.com/marketing/gf225122_intro.html

This document describes the differences between the S/390 platform and other
so called "mainframe class" platforms.

-----Original Message-----
From: McKown, John [mailto:JMc...@INSURDATA.COM]
Sent: 21 February 2000 16:50
To: IBM-...@BAMA.UA.EDU

Subject: Re: Mainframe Downsizing


With respect to Windows2000 (NT 5.0). Read this -
http://www.zdnet.com/sr/stories/column/0,4712,2439997,00.html
The author says, "No way for production!"

I can't help much with other platforms. My main concern would be sustainable
I/O rates. The S/390 can move data better than the best of the Intel/PC
systems.

Craig Otway

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
A must read are some of the papers on
http://www.s390.ibm.com/marketing/position.html. One of the best comparisons
I have seen on this is topic is
http://www.s390.ibm.com/marketing/gf225122.html.

Chuck

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
what transaction arrival rate and IO rates? They are probably more germane
than CPU usage.
Zulkifly Samad <zu...@PC.JARING.MY> wrote in message
news:006901bf7c8f$4f71f860$0b0a0a8c@zuldomain...

> We are running OS/390 Ver 1.2 with CICS/ESA Ver 4.1 and DB2 Ver 3.1.
> CPU is CMOS 9672 - R24, with 512 Mbyte memory.
>
> We have both online and batch application.
>
> During online (day) - peak CPU utilization is 70%
> During batch (night) - peak CPU utilization is 85%
>
> Currently we have about 7,000 users - LU0 (majority) and LU2
>
>
>
> ----- Original Message -----
> From: Bill Becker <Bill....@HDS.COM>
> Newsgroups: bit.listserv.ibm-main
> To: <IBM-...@BAMA.UA.EDU>
> Sent: Tuesday, February 22, 2000 12:45 AM
> Subject: Re: Mainframe Downsizing
>
>
> > You need to be a little more specific as to what the application
> > is attempting to do? Is it a batch thing? Online? How many users?
> >
> >
> > -----Original Message-----
> > From: Zulkifly Samad [mailto:zu...@PC.JARING.MY]
> > Sent: Monday, February 21, 2000 08:34 am
> > To: IBM-...@BAMA.UA.EDU

Marcos A. Morelatto

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
Two years ago we received the same mission (port COBOL applications to UNIX)
to support a new plant of one of one client, due to a manager "magazine
driven" approach. After spend many times the planned time and costs we
finally end the project. It works... but the response time of applications
are very poor (3 to 30 seconds) causing negative impact in the client
business, primary due to the dasd performance (not all I/Os can be done in
the database memory). The same application running in a "legacy" mainframe
costs only a quarter than the UNIX resulting cost.

Note: recently we added a ESS (shark) machine to main plant configuration of
the same client, causing the medium response times to be mesured in
sub-seconds.

In resume, don't forget to ask your "downsize gurus" about the capacity
planning for the I/O part of the downsize. If they say "we will use memory
to reduce the I/Os", put it in a paper and ask him to sign the paper. The
final costs of storage (dasd and backup tapes) and the database software are
much more expensive than that numbers that are in the managers mind. Another
mistake is the administrative tools available for the UNIX box, when
compared with S/390 "legacy" tools.
Ask the gurus about "Workload Balance", "Tape Management System",
"Performance Monitoring", "Batch Scheduling Control", "High Quality Printed
Output", etc... For all that you will have additional and hidden big costs.
Ops... there is the NT option also... :)

Focus your attention in the "business" side... The choice of the correct
platform for your applications has direct impact on the live and money of
your enterprise. Ops... there is the NT option also... :)

Regards,

Marcos A. Morelatto
debis humaita - IT Services Latin America

Longnecker, Dennis

unread,
Feb 21, 2000, 3:00:00 AM2/21/00
to
Are current plans are to utilize JAVA to access the calendars.

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-...@BAMA.UA.EDU]On

Behalf Of Preston L. Bannister
Sent: Saturday, February 19, 2000 7:39 PM
To: IBM-...@BAMA.UA.EDU

Subject: Re: Mainframe Based Calendar Engine


Well, if we're going to pick a framework for doing web-applications,
I'd have to pick Java as a first choice. Might want to start with the
JetSpeed framework:

http://java.apache.org/jetspeed/index.html

If you want to maximize the impact on you resume, you could incorporate
XML by making use of Cocoon:

http://xml.apache.org

This even makes some sense. You need to beware the computational overhead
of using XML and Java in combination. An older/slower IBM mainframe may not
be able to turn around requests in reasonable periods of time.

From: Frank Clarke
>
> On 19 Feb 2000 08:04:53 -0800, da...@TICNET.COM (David Alcock) wrote:


>
> >"Longnecker, Dennis" wrote:
> >> Anyone aware of a Calendaring Engine available on the OS/390 that is
API
> >> callable?
> >

> >If I were writing new web applications that did calendaring, I'd go
> >with Perl. I'm sure that you could find a lot of code on the web that
> >does this.
>

> If I were thinking along the lines of PERL, I would look first at
> REXX, ORexx, Regina, and NetREXX. Why limit yourself?

James A. Williams

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
Last year there was a good paper at www.xephon.com on this very
subject.

Nolly Unvala

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
I managed a 10-man project team to port all the DB2/IMS-DB-DC/Cobol
backoffice applications to an NT/Solaris platform for an investment bank in
1998 (Money Market instruments, Foreign Exchange & commercial loans). The
application was approx. 750K lines of Cobol code, about 200K JCL code,
database size at conversion - about 4G. Most of the application (95%) used
DB2 but there was some legacy IMS-DB code which was converted to DB2 during
the port, online IMS-DC green screens were ported to Powerbuilder front-ends
and the IBM Cobol2 code was converted via a custom Rexx script to Microfocus
(NetExpress) Cobol. We designed and implemented a custom middleware
component to map the Powerbuilder GUI streams to IMS-DC (3270) datastreams
so that Cobol code was minimally impacted. Batch JCL was replaced by Rexx
scripts and job scheduling was handled by AutoSys. The entire project was
completed in 9 months. Transaction (ie. online) performance on the new
platform was marginally better than on the 390, but the nightly batch run
dropped from 5hrs to 3hrs (an unexpected surprise since I'd calculated it to
perform about the same). I attribute the latter to a badly-tuned DB2 system
on the 390, for this application's workflow.

So yes, downsizing can be successful. However, this was a unique
circumstance in that the requirement to port was mandated by external
business reasons (ie. it *HAD* to be done), which resulted in a highly
motivated project team.

I'd be glad to answer any further questions you may have (without disclosing
proprietary information).

Nolly
no...@spamcop.net

"Zulkifly Samad" <zu...@PC.JARING.MY> wrote in message

news:005901bf7c89$823776e0$0b0a0a8c@zuldomain...

Gilbert Saint-flour

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
In <Dcms4.1093$7_2.1...@typhoon2.gnilink.net>, on 22 Feb 2000 at 02:08,
"Nolly Unvala" <no...@spamcop.net> said:

>I managed a 10-man project team to port all the DB2/IMS-DB-DC/Cobol
>backoffice applications to an NT/Solaris platform for an investment bank

>I'd be glad to answer any further questions you may have

How many users does the ported application serve? The initial poster
indicated they have 7000 users.

Gilbert Saint-flour
http://members.home.net/gsf/


Nolly Unvala

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
This was a low transaction volume app (approx. 5000/day - 90% of the
activity between 2pm & 6pm) and was designed for 50 concurrent users;
however though connections were held I was never able to detect more than
about 15-20 concurrently active users (via monitoring the UDB connections).
I missed the part about 7000 users in the original msg - that's a whole
different ball game. Also, the project was a "throw-away" in that the
ported app is to be replaced by the end of 2001. It was a fun, exciting &
challenging nine months though, and we all derived considerable satisfaction
in proving that this particular 20-year old legacy, but "mission-critical"
mainframe app could be ported fairly easily to a modern platform.

"Gilbert Saint-flour" <g...@ibm.net> wrote in message
news:38b286d9$1$tfs$mr2ice@news...

Gilbert Saint-flour

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
In <17xs4.898$lw1.1...@typhoon1.gnilink.net>, on 22 Feb 2000 at 14:33,
"Nolly Unvala" <no...@spamcop.net> said:

>This was a low transaction volume app (approx. 5000/day - 90% of the
>activity between 2pm & 6pm) and was designed for 50 concurrent users;
>however though connections were held I was never able to detect more than
>about 15-20 concurrently active users (via monitoring the UDB
>connections). I missed the part about 7000 users in the original msg -
>that's a whole different ball game.

The 7000-user figure was in his 2nd message.

>Also, the project was a "throw-away" in that the ported app is to be
>replaced by the end of 2001.

Maybe I'm way off-track here, but was a P/390 considered as an alternative
to the NT/Solaris platform you mentioned? For less than $100k, you could
have bought a fully-equipped P/390 with IMS, DB2 and all and my guess is
that your 9-month project ended up costing more than that.

>It was a fun, exciting & challenging nine months though, and we all
>derived considerable satisfaction in proving that this particular
>20-year old legacy, but "mission-critical" mainframe
>app could be ported fairly easily to a modern platform.

"modern platform" ? In what respect are NT and Solaris more "modern" than
OS/390 and DB2? Because of their sketchy approach to security,
reliability and availability?

Gilbert Saint-flour
25 Billingsley Drive
Livingston, NJ 07039
(973) 992-9318
http://members.home.net/gsf/


Nolly Unvala

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
Responding to
:"Gilbert Saint-flour" <g...@ibm.net> wrote in message
news:38b2a93e$2$tfs$mr2ice@news...

In <17xs4.898$lw1.1...@typhoon1.gnilink.net>, on 22 Feb 2000 at 14:33,

> was a P/390 considered as an alternative

Nope! I presented it as an option in fall '97 but the mandate was Solaris
and/or NT and it came from a rabidly anti-IBM cabal of decision-makers.
The only concession made to me was UDB.

> For less than $100k, you could have bought a fully-equipped P/390 with
> IMS, DB2 and all and my guess is that your 9-month project ended up
> costing more than that.

How about well over $1M (most of it, of course, manpower costs)? I
cannot defend the technical rationality of the decision - I was hired to
implement it, if possible.

> "modern platform" ?

Sorry, wrong choice of word. I'm with you on your sentiments.

Gilbert Saint-flour

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
In <wjys4.1172$7_2.1...@typhoon2.gnilink.net>, on 22 Feb 2000 at 15:55,
"Nolly Unvala" <no...@spamcop.net> said:

>> was a P/390 considered as an alternative
>Nope! I presented it as an option in fall '97 but the mandate was
>Solaris and/or NT and it came from a rabidly anti-IBM cabal of
>decision-makers. The only concession made to me was UDB.
>> For less than $100k, you could have bought a fully-equipped P/390 with
>> IMS, DB2 and all and my guess is that your 9-month project ended up
>> costing more than that.
>How about well over $1M (most of it, of course, manpower costs)?
>I cannot defend the technical rationality of the decision - I was hired
>to implement it, if possible.

Well, I guess that's yet another battle won by the PHBs, or was it more
the "airline-magazine" managers?

Gilbert Saint-flour
http://members.home.net/gsf/


Mullins, Craig

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
You do not mention whether the applications you are looking at downsizing
use databases (DB2 or other) or flat files. One of your concerns will be the
scalability of the DBMS chosen and its cross-platform compatibility. If you
are using, say DB2 on OS/390 for example, what is being targeted on other
platforms - Unix with DB2 (or with Oracle), NT with DB2 (or Oracle or SQL
Server). DB2 is not 100 per cent the same from platform to platform. The SQL
is mostly the same, but the underlying technology and components will differ
(causing your database administration staff to learn new technology - - not
just the OS but also the DBMS).

You are tackling a very large job. I would suggest that you document all of
the components and parts of the applications in question. Then work with
Gartner Group, if you have a contract with them, to determine what platform
(or platforms) are appropriate for the applications and your service level
requirements. I suggest Gartner Group because I know they have done some
research into the area of DBMS scalability and the suitability of platforms
based on the type of application, number of users, and transactions per
second (although I have not seen any recent research published from Gartner
I know there was quite a bit done two or three years ago).

Good luck!
Craig S. Mullins
Director, DB2 Technology Planning
BMC Software
http://www.bmc.com
mailto:Craig_...@BMC.com


-----Original Message-----
From: Zulkifly Samad [mailto:zu...@PC.JARING.MY]
Sent: Monday, February 21, 2000 11:16 AM
Subject: Re: Mainframe Downsizing

We are running OS/390 Ver 1.2 with CICS/ESA Ver 4.1 and DB2 Ver 3.1.
CPU is CMOS 9672 - R24, with 512 Mbyte memory.

We have both online and batch application.

During online (day) - peak CPU utilization is 70%
During batch (night) - peak CPU utilization is 85%

Currently we have about 7,000 users - LU0 (majority) and LU2

----- Original Message -----
From: Bill Becker <Bill....@HDS.COM>
Newsgroups: bit.listserv.ibm-main
To: <IBM-...@BAMA.UA.EDU>

Sent: Tuesday, February 22, 2000 12:45 AM
Subject: Re: Mainframe Downsizing


> You need to be a little more specific as to what the application
> is attempting to do? Is it a batch thing? Online? How many users?
>
>
> -----Original Message-----
> From: Zulkifly Samad [mailto:zu...@PC.JARING.MY]
> Sent: Monday, February 21, 2000 08:34 am
> To: IBM-...@BAMA.UA.EDU
> Subject: Mainframe Downsizing
>
>

Gerard S.

unread,
Feb 22, 2000, 3:00:00 AM2/22/00
to
Gilbert Saint-flour wrote in message <38b2a93e$2$tfs$mr2ice@news>...

|In <17xs4.898$lw1.1...@typhoon1.gnilink.net>, on 22 Feb 2000 at 14:33,
|"Nolly Unvala" <no...@spamcop.net> said:
|
|>This was a low transaction volume app (approx. 5000/day - 90% of the
|>activity between 2pm & 6pm) and was designed for 50 concurrent users;
|>however though connections were held I was never able to detect more than
|>about 15-20 concurrently active users (via monitoring the UDB
|>connections). I missed the part about 7000 users in the original msg -
|>that's a whole different ball game.
|
|The 7000-user figure was in his 2nd message.
|
|>Also, the project was a "throw-away" in that the ported app is to be
|>replaced by the end of 2001.
|
|Maybe I'm way off-track here, but was a P/390 considered as an alternative
|to the NT/Solaris platform you mentioned? For less than $100k, you could

|have bought a fully-equipped P/390 with IMS, DB2 and all and my guess is
|that your 9-month project ended up costing more than that.
|
|>It was a fun, exciting & challenging nine months though, and we all
|>derived considerable satisfaction in proving that this particular
|>20-year old legacy, but "mission-critical" mainframe
|>app could be ported fairly easily to a modern platform.
|
|"modern platform" ? In what respect are NT and Solaris more "modern" than
|OS/390 and DB2? Because of their sketchy approach to security,
|reliability and availability?
|
| Gilbert Saint-flour
| 25 Billingsley Drive
| Livingston, NJ 07039
| (973) 992-9318
| http://members.home.net/gsf/
______________________________________________________________

When I first read about 7,000 users, I really thought that was a bit too much
for
MVS/TSO, although I assumed TSO users, not IMS or CICS users.

Aside from all that, just how many concurrent logged-on TSO users can MVS
handle (whatever flavor of MVS it's called, say OS/390, etal)? Does anyone
have some numbers what their shop has for a peak? Any MVS' over 10,000
concurrent logged on users?

Gerard S.

Beate Kawelke

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Hi everybody,

as stated above, I'd like to find out if DFHSM is currently active. I need to do
this from an HLASM program, so one option might be testing for the enqueue
"SYSZARC (major) ACTIVE (minor)". Problem is that this is not documented and
thus not sure to work in all releases...

Anybody knows an "official" way to do this ?

Thanks in advance,
Beate

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to list...@bama.ua.edu with the message: GET IBM-MAIN INFO

Rob Scott

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
The SYSZARC enqueue test seems pretty good to me.

You could also try the following :

CVT --> CVTHSM --> QCT

Offset 8 into the QCT gives the HSM ASCB address.

This method is probably not supported either by has worked for
me for the last 8 years at least.

* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!


Thomas Ramseier

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Beate,

You might use the List_Status Service IFAEDLIS described in "MVS
Programming: Product Registration (GC28-1729)" to obtain the status of HSM.
See Gilbert Saint-Flours SHOWMVS-Cmd for a coding sample.
Cheers
Tom

Beate Kawelke

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Hi Thomas,

thanks for the answer - but if I RTFM correctly, I will only get the
information if HSM is enabled on the system. What I'd need to know is if a
subsequent HBACKDS will complete successfully, i.e. if the HSM address space is
currently running and able to react on this service request...

But maybe it will be enough to just check for HSM's enablement - the dataset
will be backed up in the next cycle, anyway. Hmmmm - will have to think this
over...

Thanks,
Beate

PS: Is it possible that an installation enables HSM without ever starting it ?
If yes, I might get into trouble with the above strategy...

Bob Richards

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Beate,

Just finding out if DFSMShsm is active is not enough. What about if
functions are HELD?

Rexx could be used with HSEND Query commands, parsing the output of the
query and then appropriate action taken.

=====
Bob Richards, OS/390 Consultant Internet: richa...@yahoo.com
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

Beate Kawelke

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Hi Bob,

is there an API for this ? The request needs to be done within an STC address space, with no REXX
environment present. We've already checked the HSM macro manual but didn't find an API / macro to
do the HSEND...

Thanks,
Beate

Bob Richards wrote:

> Beate,
>
> Just finding out if DFSMShsm is active is not enough. What about if
> functions are HELD?
>
> Rexx could be used with HSEND Query commands, parsing the output of the
> query and then appropriate action taken.

----------------------------------------------------------------------

john....@gerling.de

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Take a look at the HSM QCT control block (chained off the CVT +X'3DC').

MQCTFACT is set to 1 if HSM is active. If you want to see what functions are
active, that is harder. You have to chain to the MCVT which is in the HSM
private address space.

For more info. see DFSMS/MVS V1R4 DFSMShsm Diagnosis Reference.

John

john....@gerling.de

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Bob, If I remember correctly, HSM HSEND output is not 'catchable' from TSO.

I remember trying this out (quite a while ago), and I believe I had to parse
the messages via an MPF exit (after issuing the equivalent MVS MODIFY
command) (unsolicited messages???).

John

> Bob Richards wrote:
>
> > Beate,
> >
> > Just finding out if DFSMShsm is active is not enough. What about if
> > functions are HELD?
> >
> > Rexx could be used with HSEND Query commands, parsing the
> output of the
> > query and then appropriate action taken.

----------------------------------------------------------------------

Kloth Ralf

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Hello,

is there anyone who knows and please can tell me, where I
can find the IBM character set codepage definitions documented ?
Especially I need the layout of codepages 037 (American
EBCDIC) and 273 (German EBCDIC).

Thank you in advance.

Ralf

Bob Richards

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
John,

It is if the syntax of the command allows for ODS(your.dsn).

I have trapped the output that way and then allocated the ODS and used
EXECIO to read it.

--- john....@GERLING.DE wrote:
> Bob, If I remember correctly, HSM HSEND output is not 'catchable'
> from TSO.
>
> I remember trying this out (quite a while ago), and I believe I had
> to parse
> the messages via an MPF exit (after issuing the equivalent MVS MODIFY
> command) (unsolicited messages???).
>
> John
>
> > Bob Richards wrote:
> >
> > > Beate,
> > >
> > > Just finding out if DFSMShsm is active is not enough. What about
> if
> > > functions are HELD?
> > >
> > > Rexx could be used with HSEND Query commands, parsing the
> > output of the
> > > query and then appropriate action taken.
>
>

----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to list...@bama.ua.edu with the message: GET IBM-MAIN
> INFO
>

=====


Bob Richards, OS/390 Consultant Internet: richa...@yahoo.com
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com

----------------------------------------------------------------------

Kloth Ralf

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Hello,

a few minutes ago, I asked the list

>is there anyone who knows and please can tell me, where I
>can find the IBM character set codepage definitions documented ?
>Especially I need the layout of codepages 037 (American
>EBCDIC) and 273 (German EBCDIC).

and got the following kind and fast reply from Mr. Knut Feiert,
which was exactly what I needed :

>These tables are in:
>3174 Establishment Controller,
>Character Set Reference
>GA27-3831

This manual is on the IBM Hardware Collection CD, SK2T-5843.

Thank you !

Ralf

Bruce Black

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Note for FDRABR users: MQCTFACT is also set if ABR is installed and enabled for
auto-recall. ABR has HSM compatibility for recalls, so any program that tests
that flag and issues HSM recall requests will also work with ABR.

Here is another flag you can test for HSM or ABR recall:
CVTDFA (CVT+4C0) points to the DFA (Data Facilities Area) which has flags for
DFP functions and facilities. It is mapped by macro IHADFA. Flag DFARECAL
indicates that HSM-type recalls are available.

john....@GERLING.DE wrote:
>
> Take a look at the HSM QCT control block (chained off the CVT +X'3DC').
>
> MQCTFACT is set to 1 if HSM is active. If you want to see what functions are
> active, that is harder. You have to chain to the MCVT which is in the HSM
> private address space.
>
> For more info. see DFSMS/MVS V1R4 DFSMShsm Diagnosis Reference.
>
> John
>

> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to list...@bama.ua.edu with the message: GET IBM-MAIN INFO

--
Bruce A. Black
Senior Software Developer for
FDR, CPK, ABR, SOS, UPSTREAM, FATS/FATAR
Innovation Data Processing
Little Falls, NJ 07424
973-890-7300
personal: bbl...@fdrinnovation.com
sales info: sa...@fdrinnovation.com
tech support: sup...@fdrinnovation.com

Roland Schiradin

unread,
Mar 10, 2000, 3:00:00 AM3/10/00
to
Beate,

have you seen the Instance at EDAAENUMINSTANCES?


Roland

Thierry FALISSARD

unread,
Mar 11, 2000, 3:00:00 AM3/11/00
to
For EBCDIC-Ascii translation, I use the EDCUxxxx tables
that are in SCEERUN (sources are in SCEEGXLT).
This is doc'ed in the C/C++ Programmer's Guide.
I would appreciate that IBM provide "official" translation tables
usable everywhere...

http://os390-mvs.hypermart.net
Is there life after MVS ?

Kloth Ralf wrote


>Hello,
>
>a few minutes ago, I asked the list
>
> >is there anyone who knows and please can tell me, where I
> >can find the IBM character set codepage definitions documented ?
> >Especially I need the layout of codepages 037 (American
> >EBCDIC) and 273 (German EBCDIC).
>
>and got the following kind and fast reply from Mr. Knut Feiert,
>which was exactly what I needed :
>
> >These tables are in:
> >3174 Establishment Controller,
> >Character Set Reference
> >GA27-3831
>
>This manual is on the IBM Hardware Collection CD, SK2T-5843.
>
>Thank you !
>
> Ralf
>

Beate Kawelke

unread,
Mar 15, 2000, 3:00:00 AM3/15/00
to
Hi Roland,

no, I haven't - where can I find information on that ?

Oh, and while I'm at it - thanks to y'all for your input, it really helped
us tremendously. We have now implemented the "check the product
enablement" solution and will see where this takes us in our data centers.
If we get into trouble we will have to follow the control blocks but I'd
rather try the other way first...

Beate


Roland Schiradin schrieb:

> Beate,
>
> have you seen the Instance at EDAAENUMINSTANCES?
>
> Roland
>

0 new messages