开云体育

Scott Johnson wrote:

Recently, when saving a member to a PDS I've been getting
this message:

SCOTT ,ISPFTSO ,294,DA,SYS00083,WRITE ,OVERFLOW
INCOMP,0000000A000A02,BSAM

That's an MVS message, yes? Does anything also appear on the Hercules console when this error message? If yes, then it's a bona fide I/O error.

Second question: is it reproducible? Is there a sequence of actions that can trigger this error on demand? If yes, then I would suggest enabling CCW tracing on device 294, so we can the CCW chain that MVS is issuing. That would provide a clue as to what's going on.

May we see your Hercules log when this error occurs? Seeing your config file might also help.

Thanks.

--
"Fish" (David B. Trout)
Software Development Laboratories

mail: fish@...

Recently, when saving a member to a PDS I've been getting this message

SCOTT? ?,ISPFTSO ,294,DA,SYS00083,WRITE ,OVERFLOW INCOMP,0000000A000A02,BSAM

returning to REVEDIT, displays "I/O error", and another save operation works fine.? The 294 DASD is a 3380 volume that I've created with dasdinit, and seems to work just fine otherwise. This is the only place I've seen an error in regards to this volume.

Any ideas?? Yes, I'm backing up the entire dasd set 2 or thee times a day to not lose any work.

Thanks in advance,

Scott

John,

You may need to do a CLPA or at least an MLPA ...

Mark

On 20/01/2024 6:40 am, John James wrote:

Does anyone have any idea what a system FFF abend is?

I'd say it means that SVC 255 was not installed on this system.

Cheers,
Greg

Hello Mark, good to know that there's t least one kindred spirit out there.

So I followed the advice in the old Volker Bandke post that was contributed a few days ago, and re-ran the DBDGEN using macros from MVT instaed of MVS (DCB and 3 IHB macros). Then I re-tried the TESTLOAD COBOL program and specified a VSAM cluster with AMP instead of an ISAM file for the TESTDB DD statement.

Didn't get very far, only as far as the first DL/I ISRT call...

Does anyone have any idea what a system FFF abend is? I found only one brief mention anywhere online, and it talked about "contention". I'll keep looking for other information online, but all ideas accepted in the meantime!

Thanks, JJ

I am also interested in MVSDDT because of some debugging I needed to do the day before yesterday, but it seems it requires communication with a Java task and run a gui next to the 3270 so I lapsed into stacking TSO TEST commands in a Rexx (and temporarily moving stuff to z/OS 2.5 (the employer's) because I never did that in bRexx yet and I had a presentation in the morning).

Meanwhile I have located documentation of MVSDDT and I will give it a shot when time allows, because it looks brilliant and just what the doctor ordered. Will keep you abreast of developments.

best regards,

搁别苍é.

On Wed, 17 Jan 2024, Mike Schwab wrote:

The quantum computers are suited toward simulations of reality, more along
the CDC/Cray massive parallel?processing.

They can do far more than this, even in traditional
fields.

Like linear search, in O(sqrt(n)) instead of O(n), just
to speak about data.

Or integer prime factorization in polynomial time instead
of exponential time: RSA as we know it would be gone
in seconds if we get access to quantum registers
of thousands of entangled qbits with a decent depth.

It is true they not displace classical computing
in all fields, but would any way makes the ICT
field completely different.

Simulation of reality is another topic, quantum
computing is not a simulation, quantum computing
is the "real" thing ;-)

Peppe.

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia Phone:+39.075.585-5518
-----------------------------------------------------------------------------

On Wed, 17 Jan 2024, Dave Wade wrote:

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Giuseppe
Vitillaro
Sent: Wednesday, January 17, 2024 9:09 AM
To: [email protected]
Subject: Re: [H390-MVS] IBM METALC curiosity.

On Wed, 17 Jan 2024, Greg Price wrote:

On 17/01/2024 6:38 am, Giuseppe Vitillaro wrote:

The word "puntglio" is when you insist on a point just because you
are not likely to admit you were wrong, even against your own
interest.

I'll suggest "bloody-minded"...

Cheers,
Greg

I believe that the English word "punctilious" is derived from this Italian
word...

Nice to know.

The english vocabulary is probably one of the largest
between the european languages ;-)

English speakers have access to latin and german
words. We, poor italians, must accomplish our
conversations with just the latin vocabulary.

Peppe.

Dave

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia Phone:+39.075.585-5518
-----------------------------------------------------------------------------

On Wed, 17 Jan 2024, Dave Wade wrote:

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Giuseppe
Vitillaro
Sent: Wednesday, January 17, 2024 9:27 AM
To: [email protected]
Subject: Re: [H390-MVS] IBM METALC curiosity.

On Tue, 16 Jan 2024, Charles Bailey wrote:

When I worked for IBM I saw how protective IBM was of the PL/S, PL/AS,
PL/X compiler. All of the manuals were marked IBM Confidential.
There were security audit police that would check a particular MVS
system where the compiler was installed to make sure it was installed
according to corporate instructions. I think it had to reside in its
own linklist load library and the compiler was marked with a RACF
attribute of "execute only", meaning that the OS could read the load
module from disk for the purpose of executing it but a user couldn't
make a copy of the load module. I don't remember for sure but it
might have been the case that anyone who wanted to use the compiler had

to

be a member of a particular RACF group.

I understad, really an hyper protective behaviour.

Just because MVS had been rewritten in this language?

I believe its used elsewhere.

By the way, from what I've seen, MVS3.8j sources are pure assembler, with
some PL/S source stored as comments, but it looks to my eyes as a pure
assembler OS.

This is the assembler output from the PL/S compiler. The comments you see
are the real PL/S source.

We have the souce of MVS3.8j on board, I've seen
them under TK4-.

All the PL/S code, as a comment, is there?

Peppe.

When MVS had been rewritten in PL/S?

Its always been PL/S. Well as far as I know, MVS has, but some earlier
products in the OS family must have been Assembler or one of the
predecessors to PL/S.
According to Wikipedia the TSO in MVT was one of the first products, but I
am a VM/CMS person and VM/CMS has always been pure assembler.

Peppe.

Dave
G4UGM

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia

Phone:+39.075.585-5518

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

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia Phone:+39.075.585-5518
-----------------------------------------------------------------------------

On Tue, 16 Jan 2024, Charles Bailey wrote:

When I worked for IBM I saw how protective IBM was of the PL/S, PL/AS, PL/X compiler. All of the manuals were marked IBM Confidential. There were security audit police that would check a particular MVS system where the compiler was installed to make sure it was installed according to corporate instructions. I think it had to reside in its own linklist load library and the compiler was marked with a RACF attribute of "execute only", meaning that the OS could read the load module from disk for the purpose of executing it but a user couldn't make a copy of the load module. I don't remember for sure but it might have been the case that anyone who wanted to use the compiler had to be a member of a particular RACF group.

I understad, really an hyper protective behaviour.

Just because MVS had been rewritten in this language?

By the way, from what I've seen, MVS3.8j sources
are pure assembler, with some PL/S source stored
as comments, but it looks to my eyes as a pure
assembler OS.

When MVS had been rewritten in PL/S?

Peppe.

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia Phone:+39.075.585-5518
-----------------------------------------------------------------------------

On Wed, 17 Jan 2024, Greg Price wrote:

On 17/01/2024 6:38 am, Giuseppe Vitillaro wrote:

The word "puntglio" is when you insist on a
point just because you are not likely to admit
you were wrong, even against your own interest.

I'll suggest "bloody-minded"...

By the way, going off-topic now, for what
I may understand of IBM in these days, the
corporation, "mom", may be on the border
of leaving even from the mainframe businness.

IBM defintely has the most advanced technology,
I'm aware about, for "real" quantum computing.

From the announces, a modular quantum chip,
133 qubits entangled, is almost on the line
(2025) of becoming a commercial product and
IBM seems to expect a path which will lead
to quantum processor up to thousands of
entangled qubits before 2030. Such a modular
quantum technology let to foresee some sort
of "quantum mainframe", an architecture which
will mix the classical model with a bunch
of quantum registers of growing length and depth
for advanced quantum computing.

It looks like a pattern we have already seen
at beginning of the sixty, when batch mainframes
was just rented, for a lot of money, to big
companies. At that time IBM was almost
the only owner of the mainframe technology.

If this even become true, mainframes would
lose economical interest for IBM in a very
short time, it may happen, as happened for PC,
that some sort of "Lenovo Company" will replace
BigBlue in the commercial field of classical
mainframes.

Wondering if what I've in mind may really happen
in the timeframe from our days to 2040.

Peppe.

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia Phone:+39.075.585-5518
-----------------------------------------------------------------------------

On Wed, 17 Jan 2024, Greg Price wrote:

On 17/01/2024 6:38 am, Giuseppe Vitillaro wrote:

The word "puntglio" is when you insist on a
point just because you are not likely to admit
you were wrong, even against your own interest.

I'll suggest "bloody-minded"...

Cheers,
Greg

Just googled for this words, Greg:

"the behaviour of someone who is very determined and refuses to give up, to change their mind, or to do what others want them to do: He was losing patience with her bloody-mindedness. Our neighbour refused to chop down the tree out of sheer bloody-mindedness"

Yep, we are almost there, italian is definitely
more ambiguos than american or british (is this
british or american?) english.

It may add to the word the shadows of obstinacy
and stubborness, often about a "point of honor"
which completely disregard if something is true
or false, real or not, just because the point
had already been took for granted in the past.

For what I'm reading the same broad sum of meanings
just doesn't exist in english in a single word.

But I would think a large corporation which is
up and running from a century would be not
capable of these feelings.

When I joined IBM Italy, in 1985, the very first
thing which the company said to me was: we are
here for doing businness (I've read that as "money"
in my naive view of things) and nothing else.

How such a company, with this main goal, may
keep insisting, beside PL/S, speaking now in
a broader sense about things which are dead
from decades, even agaist what rationality
may suggest?

Peppe.

---

Giuseppe Vitillaro | E-Mail : giuseppe@...
ex-IBM, ex-CNR | 06123 Perugia Phone:+39.075.585-5518
-----------------------------------------------------------------------------

On 17/01/2024 6:38 am, Giuseppe Vitillaro wrote:

The word "puntglio" is when you insist on a
point just because you are not likely to admit
you were wrong, even against your own interest.

I'll suggest "bloody-minded"...

Cheers,
Greg

When I worked for IBM I saw how protective IBM was of the PL/S, PL/AS, PL/X compiler. All of the manuals were marked IBM Confidential. There were security audit police that would check a particular MVS system where the compiler was installed to make sure it was installed according to corporate instructions. I think it had to reside in its own linklist load library and the compiler was marked with a RACF attribute of "execute only", meaning that the OS could read the load module from disk for the purpose of executing it but a user couldn't make a copy of the load module. I don't remember for sure but it might have been the case that anyone who wanted to use the compiler had to be a member of a particular RACF group.

Charles Bailey