In reverse order of your questions:
Sorry, I am not free to discuss the actual application details.
And I did not get that from Peter R.'s statement -- quite the contrary in fact.
With the z/OS generated programmable field LPAR value (whatever it may be),
the STCKE result is in fact
Ah... that makes sense. Then they might have been going over that 32 CP
limit fairly soon after the limit was changed. Being the first on your
block to run new code is no fun at 2 in the morning.
On 3/19/2021 8:15 PM, Mike Schwab wrote:
I think the limit was 32 z processors in an LPAR.
Yep. Flight / hotel number + seat / room number + date (/ time) +
customer number as the key with billing details.
On Fri, Mar 19, 2021 at 8:59 PM Bill Johnson
<0047540adefe-dmarc-requ...@listserv.ua.edu> wrote:
>
> Yup, hotel reservation isn’t much different than an airline reservation.
>
>
I think the limit was 32 z processors in an LPAR. They might have
raised it by now. z15 can have 190 in high capacity order.
On Fri, Mar 19, 2021 at 9:09 PM Radoslaw Skorupka
wrote:
>
> Yes, I also heard about quick IPL time, I was even heard about under
> minute times. Of course I cannot test
Yes, I also heard about quick IPL time, I was even heard about under
minute times. Of course I cannot test it.
Regarding 3215 - it is another type of console. In z/OS realm we use
3270 family, but z/VM (CMS) like 3215 and there is TERMINAL CONMODE
command to change the type of emulated
I worked at Braniff Airways back in the late '70s and early 80s' supporting
MVT (yes MVT on a 3033 which required an RQP). They ran ACP (Airline
control Program), which I think is the origin of z/TPF. I watched them do
an IPL once. It seems just a few seconds and the operators were busy typing
Yup, hotel reservation isn’t much different than an airline reservation.
Sent from Yahoo Mail for iPhone
On Friday, March 19, 2021, 9:25 PM, Attila Fogarasi wrote:
Why be surprised at hotels on the list (of zTPF users)? A hotel is just an
airplane without wings :)
zTPF is great for any high
I hardly know anything about it, but about a year ago I watched an IPL
at one of the customers you mentioned. TPF was running on a z15 with
over 30 full-speed CP's. Their support guy mentioned he was concerned
that they may be running more TPF CP's on a single box than anyone else
in the
Why be surprised at hotels on the list (of zTPF users)? A hotel is just an
airplane without wings :)
zTPF is great for any high volume transaction where there are few
transaction types but many per second (IBM claims million tps) and at low
cost. The tradeoff is constrained application function,
On Fri, 19 Mar 2021 22:55:25 +, Farley, Peter x23353 wrote:
>
>The GENERATE_UNIQUE function returns a bit data character string 13 bytes long
>(CHAR(13) FOR BIT DATA) that is unique compared to any other execution of the
>same function. The function is defined as not deterministic. Although
On Fri, 19 Mar 2021 21:50:57 +, Seymour J Metz wrote:
>> Is there any reason to do a WAIT before the CHECK?
>EOV processing.
>
It was long ago and I no longer have the code. IIRC, I coded a CHECK,
but never a WAIT. It was specialized code, and I may not have accounted
for EOV.
What ill
Been around for decades as the airline system but usable by any industry
requiring massive throughput.
Sent from Yahoo Mail for iPhone
On Friday, March 19, 2021, 8:16 PM, Radoslaw Skorupka
wrote:
Old only?
I have read about relatively new and small airlines, not mentioned below.
And of
Old only?
I have read about relatively new and small airlines, not mentioned below.
And of course airline control program is not really applicable to banks
and hotels. In fact I understand banks, but I'm really surprized at the
hotels on the list. What transaction workload do the have???
--
The old yet still used airline control program.
Sent from Yahoo Mail for iPhone
On Friday, March 19, 2021, 8:03 PM, Radoslaw Skorupka
wrote:
I know the IBM-MAIN forum is mostly about z/OS, but I wanted to ask
about about z/TPF - it seems to be on topic.
z/TPF
TPF is the system I have
I know the IBM-MAIN forum is mostly about z/OS, but I wanted to ask
about about z/TPF - it seems to be on topic.
z/TPF
TPF is the system I have never seen.
I'm pretty sure there is no TPF installation in Poland.
However I'm curious about the following:
1. What is typical size of TPF system? I
W dniu 19.03.2021 o 22:23, Pierre Fichaud pisze:
If a security (RACF) violation occurs in a CICS region, where does the
violation get reported?
I couldn't find anything in the CICS SMF records but I'll look again.
Do they get reported in the JESMSGLG or in a CICS ?
Does a CICS exit need to be
Ah! I somewhat misread the question.
Security violations for Db2 are reported as SMF Type 102, IFCID (which is kind
of like a subtype, but not in the subtype field) 140.
They are not really documented in a manual. They are documented in macros in
the Db2 product.
They are not reported in SMF
You indicated RACF which sometimes people use generically for SAF). If you
have Top Secret or ACF2 there is an additional SMF record (type 230 IIRC). TSS
and ACF2 also generate the type 80s for consistency.
Matt Hogstrom
m...@hogstrom.org
+1-919-656-0564
PGP Key: 0x90ECB270
Facebook
Pierre, have you tried to look at the SMF Record Type 80 (Security Product
Processing) record for the information you want?
Thanks..
Paul Feller
GTS Mainframe Technical Support
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Pierre Fichaud
Sent: Friday, March
Thanks Charles, I did already think of that, and it has performance
disadvantages that I can't ignore.
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Charles Mills
Sent: Friday, March 19, 2021 4:47 PM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Contents of TOD
Thanks for the idea Gil, but I need guaranteed alphanumeric (upper case letters
plus digits, nothing else).
Plus invoking z/OS *nix services dynamically is not cheap either.
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Paul Gilmartin
Sent: Friday, March
Yes, but then we have to have a loop to call random, multiply by 100, use the
integer value of the result to pick a random character from the allowed ones
for the field to be filled, with as many iterations as the length of the field.
Not cheap.
One STCKE is far easier and quicker.
Peter
Thanks Salva, but as I said in my prior reply, this is not a case where DB2 is
already in use. The additional overhead to use DB2 for this would be far too
large.
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Salva Carrasco
Sent: Friday, March 19, 2021
Thanks for the reference. I see from the DB2 documentation that:
The GENERATE_UNIQUE function returns a bit data character string that is
unique, compared to any other execution of the same function.
GENERATE_UNIQUE()
The schema is SYSIBM.
The GENERATE_UNIQUE function returns a bit data
There is a DB2 mailing list, run by IDUG. Google can find it for you.
I do not *know* the DB2 answer for certain but I believe all RACF violations
are caught internally by DB2 and reported as SQL completion codes. RACF manages
the whole security process itself -- either (the old way) totally
> Is there any reason to do a WAIT before the CHECK?
EOV processing.
> There's little reason to use BSAM.
NOTE, POINT.
> The consensus in this thread has been,
> QSAM is at least nearly as good as BSAM;
> perhaps better
There's a good deal of overlap between the BSAM and QSAM code.
--
If a security (RACF) violation occurs in a CICS region, where does the
violation get reported?
I couldn't find anything in the CICS SMF records but I'll look again.
Do they get reported in the JESMSGLG or in a CICS ?
Does a CICS exit need to be installed?
There's tons of documentation to go
@Peter do you mean alphanumeric (A-Z, 0-9) or do you mean "any bit combination
00 to FF"? I read it as "any bits" but you did say alphanumeric.
If you mean alphanumeric then you need a table of the 36 or 62 or whatever
characters comprise your set. Then take my method, generate a number 0 to 35
On Fri, 19 Mar 2021 15:58:57 -0400, Joseph Reichman wrote:
>Data is not there after check
>
WTF!?
On Fri, 19 Mar 2021 11:25:12 -0700, Ed Jaffe wrote:
>On 3/19/2021 11:09 AM, Joseph Reichman wrote:
>>...
>> I check to see if the first fullword has been populated by the BDW
>
That's called
Hi!
If I have a DB2 data sharing group that has a well defined naming scheme,
is there any effective difference between using:
- an IEADMCxx member that identifies the job names to dump with properly
placed wildcards and using RO (sys1,sys2,sysx,...),DUMP PARMLIB=xx
- an IEADMCxx member
Checks checks for eodad
Thanks
> On Mar 19, 2021, at 4:02 PM, Ed Jaffe wrote:
>
> On 3/19/2021 12:57 PM, Paul Gilmartin wrote:
>>> On Fri, 19 Mar 2021 12:43:38 -0700, Ed Jaffe wrote:
>>>
>>>
>>> Don't forget to issue CHECK after WAIT.
>>>
>> Is there any reason to do a WAIT before the
On 3/19/2021 12:57 PM, Paul Gilmartin wrote:
On Fri, 19 Mar 2021 12:43:38 -0700, Ed Jaffe wrote:
Don't forget to issue CHECK after WAIT.
Is there any reason to do a WAIT before the CHECK? I never did.
You would issue WAIT first if you were in an environment that could not
tolerate
CP DEFINE Reader as 00C
Still valid in 7.2.
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Ed
Jaffe [edja...@phoenixsoftware.com]
Sent: Thursday, March 18, 2021 10:51
Data is not there after check
> On Mar 19, 2021, at 3:57 PM, Paul Gilmartin
> <000433f07816-dmarc-requ...@listserv.ua.edu> wrote:
>
> On Fri, 19 Mar 2021 12:43:38 -0700, Ed Jaffe wrote:
>
>>> On 3/19/2021 12:35 PM, Joseph Reichman wrote:
>>> So this is what I will do
>>>
>>> I’ll do 3
On Fri, 19 Mar 2021 12:43:38 -0700, Ed Jaffe wrote:
>On 3/19/2021 12:35 PM, Joseph Reichman wrote:
>> So this is what I will do
>>
>> I’ll do 3 reads in the first since i need to get myself going I’ll issue the
>> WAIT using the ECB from the DECB
>> Than when I finish processing that buffer and
On 3/19/2021 12:35 PM, Joseph Reichman wrote:
So this is what I will do
I’ll do 3 reads in the first since i need to get myself going I’ll issue the
WAIT using the ECB from the DECB
Than when I finish processing that buffer and need to go to the second I/O I’ll
issue a WAIT for that
etc
On Fri, 19 Mar 2021 15:35:39 -0400, Joseph Reichman wrote:
>So this is what I will do
>
>I’ll do 3 reads in the first since i need to get myself going I’ll issue the
>WAIT using the ECB from the DECB
>Than when I finish processing that buffer and need to go to the second I/O
>I’ll issue a
So this is what I will do
I’ll do 3 reads in the first since i need to get myself going I’ll issue the
WAIT using the ECB from the DECB
Than when I finish processing that buffer and need to go to the second I/O I’ll
issue a WAIT for that
etc
> On Mar 19, 2021, at 3:05 PM, Seymour J Metz
It would be helpful to include the message number in the subject.
As I recall, there is a discussion of SDSF security requirements in both the
RACF and SDSF documentation. In addition to authorizing access to the SDSF
address space, you will probably need to tailor authorization to commands.
The only safe way is to check the ECB. Relying on the buffer without first
testing for completion will lead to errors.
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of
There may be a delay for EOV processing. AFAIK there is never a delay for 7F.
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Ed
Jaffe [edja...@phoenixsoftware.com]
On Fri, 19 Mar 2021 18:20:04 +, Farley, Peter wrote:
>
>The actual application requirement is for alphanumeric random values of a
>certain length. I can’t say more than that without revealing company IP.
>
708 $ head -c16 /dev/random | uuencode -m Random.string
begin-base64 644
If it is like the crypto things I am familiar with, you could simulate it by
generating a RANDOM value and multiplying by 255 or 256 (depending on the exact
specs for RANDOM, which I do not have open at the moment) to get a value
between 0 and 255. Do that repeatedly until you have a string of
Peter,
Cobol UUID4 (available in 6.3) had a horrible performance on a z13. In z15, the
performance is excellent. We opened a SR and they tell us about the random
number generator.
If you can't wait and have Db2, I wrote a Db2 Funct to generate UUID based on
TOD. It is available in
First byte Of DECB/ECB was x’7F’ 40 bit ( posted bit was on )
> On Mar 19, 2021, at 2:25 PM, Ed Jaffe wrote:
>
> On 3/19/2021 11:09 AM, Joseph Reichman wrote:
>> Hi
>>
>> When doing overlapped I/O is there a way to tell if the I/O had completed on
>> the subsequent reads
>>
>> I check to
On 3/19/2021 11:09 AM, Joseph Reichman wrote:
Hi
When doing overlapped I/O is there a way to tell if the I/O had completed on
the subsequent reads
I check to see if the first fullword has been populated by the BDW
The DECB that you issue CHECK against contains an actual embedded ECB at
Peter, GENERATE UNIQUE Db2 function, warrants the uniqueness aceoss a Sysplex
and it is based on TODE.
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message:
Charles,
The actual application requirement is for alphanumeric random values of a
certain length. I can’t say more than that without revealing company IP.
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Charles Mills
Sent: Friday, March 19, 2021 1:31 PM
Running under test check doesn’t seem to wait after check the buffer is hex
zeros
Maybe I issue wait of the DECB->ECB
> On Mar 19, 2021, at 2:15 PM, Farley, Peter x23353
> <031df298a9da-dmarc-requ...@listserv.ua.edu> wrote:
>
> Joe,
>
> That is what the READ ECB is for. When that ECB
Joe,
That is what the READ ECB is for. When that ECB is posted then the READ is
complete.
Use the CHECK macro to determine if a particular READ is complete.
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Joseph Reichman
Sent: Friday, March 19, 2021 2:10
Hi
When doing overlapped I/O is there a way to tell if the I/O had completed on
the subsequent reads
I check to see if the first fullword has been populated by the BDW
--
For IBM-MAIN subscribe / signoff / archive access
> that only returns a fraction between 0 and 1, which could be useful but quite
> a bit more work
What do you need? An integer between 0 or 1 and 'n'?
Multiplying the result of RANDOM times 'n' should give you that integer pretty
readily, no?
This may not give you crypto quality, but the idea
Did you consider
https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.3.0/com.ibm.zos.v2r3
.e0zm100/SDSF_SDSFAUX_V2R3.htm
Charles
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On
Behalf Of Steve Beaver
Sent: Friday, March 19, 2021 8:27 AM
I forgot about the ISFPARM part, yes I set;
CONNECT AUXPROC(SDSFAUX), AUXNAME(SDSFAUX),
AUXSAF(NOFAILRC4),
DEFAULT(YES)
Carmen Vitullo
-Original Message-
From: Todd <0316e668f7df-dmarc-requ...@listserv.ua.edu>
To: IBM-MAIN
Date: Friday, 19 March 2021 10:32
It's been a long time since the SDSFAUX address space has been used for certain
SDSF functions, When I first saw this message, I want to say back in z/OS 2.2 I
used the migration guide + the SDSF operations and customization Guide to
define all the new security resources I needed. once defined
Update: It seems we are on z13 boxes at the moment, and they do not have the
Message-Security-Assist Extension 7 feature necessary to use the TRNG functions
of PRNO. I could try the DRNG functions of PRNO but they seem to be a lot of
work to use the right way (seeding, parameter blocks, etc.).
Try doing a SET SECTRACE in SDSF and then re-attempt the failing command. You
should get a good amount of info in the SYSLOG to help resolve.
We ran into some similar issues when going to 2.4.
We also set this in SDSF to resolve when some things are not defined:
CONNECT DEFAULT(COND),/*
Has anyone seen this message before know how to fix the prblem
ISF458E Not authorized to connect to the SDSF server. Verify read
access to the ISF.CONNECT.system resource in the SDSF class.
--
For IBM-MAIN subscribe /
Thanks Charles. As I indicated in one of my earlier responses, I agree with
you and am taking that route. Too many chances for failure in the future.
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Charles Mills
Sent: Friday, March 19, 2021 10:27 AM
To:
@Peter, I have given this some serious thought. Thanks for the interesting
problem!
I believe you should eschew the clever "assembler in working storage"
approach. If the UUID4 approach works for you, then of course, by all means
use it.
If not, then you should go with a conventional external
Regarding the questions about order of clock values in a multi-system
environment, if you want/need such ordering, you should use the STCKSYNC
service. It has never been clear to me how useful that is since the saving
of the time and the processing of the event are not atomic.
There are likely
Tom has a Very Good point.
Keep the ‘tape’ and save all the searches. Not to mention all the web page
maintenance.
Just my 2 cents.
Doug
.
On Mar 18, 2021, at 23:35, Ron Wells
<02ebc63ff5ef-dmarc-requ...@listserv.ua.edu> wrote:
Good idea
-Original Message-
From: IBM Mainframe
W dniu 19.03.2021 o 12:59, Paul Gilmartin pisze:
On Fri, 19 Mar 2021 14:43:05 +0800, Timothy Sipples wrote:
...
Paul Gilmartin wrote:
The latter suggests that a pseudo RNG is periodically reseeded
by the TRNG
Yes, that's right. CPACF on the IBM z14 and LinuxONE II models, and
higher, have
Yes, of course! It is just I mentioned: COCOM. Coordinating Committee
for Multilateral Export Controls.
I even remember times when (some) hard disk drives were under CoCom
restrictions.
Funny fact: devices under CoCom could not be legally exported to Poland,
but some devices (oscilloscopes)
W dniu 19.03.2021 o 09:10, Stefan Skoglund pisze:
ons 2021-03-17 klockan 10:13 + skrev Seymour J Metz:
Yes, z/OS on z is attractive for a medium or large customer, but it
is priced out of the market for a hobbyist or a "Mom and Pop"
business. The entry cost is much lower for, e.g., openSUSE
On Fri, 19 Mar 2021 14:43:05 +0800, Timothy Sipples wrote:
>...
>Paul Gilmartin wrote:
>>The latter suggests that a pseudo RNG is periodically reseeded
>>by the TRNG
>
>Yes, that's right. CPACF on the IBM z14 and LinuxONE II models, and
>higher, have this feature. If you try to use the TRNG
Sonny,
Thank you very much! I have forwarded your email to the student. Here's hoping
good things will come from this.
Have a great weekend!
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to
Thanks Ed, I hadn't actually looked at the RNG capabilities in CPACF in recent
times. From an initial once-over of the TRNG capability described in the
latest PoOP, it may be a better source of data than STCKE for my application.
I will need to make some performance measurements to make sure
I actually do not care what the value is, so long as it is guaranteed to be
unique. I am curious about the technique and value used, but I actually don't
*need* my curiosity satisfied.
Documentation of a guarantee of uniqueness is all I think a management review
would require, and in truth
ons 2021-03-17 klockan 10:13 + skrev Seymour J Metz:
> Yes, z/OS on z is attractive for a medium or large customer, but it
> is priced out of the market for a hobbyist or a "Mom and Pop"
> business. The entry cost is much lower for, e.g., openSUSE on a PC. I
> miss the days of 80% discounts
Ed Jaffe wrote:
>IBM Z was recently enhanced with a true random source in CPACF.
>For all the many decades before that, all "random" numbers on the
>mainframe were actually pseudo-random...
The IBM Crypto Express features have had TRNGs aboard for many years (and
still do). This is a fairly
71 matches
Mail list logo
