I don't think there is any check for that. I believe that z/OS isn't that
smart when it comes to figuring that stuff out, and actually I can't really
think of a reason why it wouldn't work. Remember, just because IBM doesn't say
they "support" something, doesn't necessarily mean it doesn't wor
Slightly outside my area of knowledge but since z/VM is the OS that is
attached to the DS8K, it's doing the physical IO. And this is usually FICON.
The Linux guest is being provided a virtual disk from z/VM so the disk
protocol is probably not relevant. Plus I don't think a Linux guest can
actuall
>From the Linux documentation
https://www.kernel.org/doc/Documentation/s390/DASD it certainly seems that
ECKD disks are well supported. That aligns with my recollection of
experimentation of testing various zLinux in Hercules.
On Tue, Dec 17, 2019, 09:32 Laurence Chiu wrote:
> Slightly outside m
CKD are well supported for Linux guest.
But having FBA disk to linux guest will improve any sort of performance ?
On Tue, 17 Dec, 2019, 12:32 PM Laurence Chiu, wrote:
> Slightly outside my area of knowledge but since z/VM is the OS that is
> attached to the DS8K, it's doing the physical IO. And
You will need to convert to using Linux containers without the z/VM
layer and FCP to open system volumes. I did this on ESS F20s, should
be similar for DS8xxx.
On Tue, Dec 17, 2019 at 2:44 AM Peter wrote:
>
> CKD are well supported for Linux guest.
>
> But having FBA disk to linux guest will imp
Until some shared resource breaks. Not saying it'll happen, but it might.
Mark Jacobs
Sent from ProtonMail, Swiss-based encrypted email.
GPG Public Key -
https://api.protonmail.ch/pks/lookup?op=get&search=markjac...@protonmail.com
‐‐‐ Original Message ‐‐‐
On Tuesday, December 17, 2019
Add to this the H/W and CFCC levels of the different systems. While the
'newest' release of z/OS in the Sysplex will support the highest level of the
CFCC level in the Sysplex, does 1.13 support it?
IBM normally tests up to n-2 levels/generations of OS and H/W. Combination of
older levels MAY w
Hello, I am working with an application team, and they are creating an ISPF
application. One of the options is to use an ISPF table for the data in one
component, but they will have between 50,000 and 80,000 rows in the table.
What are your experiences with large ISPF tables, and is a table of 80,
Hi Billy,
Permanent ISPF tables reside in a PDS data set which will have to be
sized appropriately. If the ISPF table is keyed then you have big
problems. ISPF tables are basically just linked lists and not dictionaries,
so if they are keyed I would advise you to use VSAM.
On 2019-12-17 10:1
ISPF Tables are great but not ideal for the volume that you are planning. If
you decide to use ISPF Tables then be sure to create them in anything other
than REXX which is extremely slow due to an inefficient design in this area.
Table serves will work but your performance may not be what you re
With a table of that size, do NOT make it a keyed table.
That would just makes it extremely slow and almost unusable.
Dan
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu
RXVSAM from Rexx is a possible replacement that may provide much better
performance than an ISPF table for that volume of records.
File 268 on the CBT Tape website.
HTH
Peter
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Billy Ashton
Sent: Tuesday, December 17,
Billy,
you can access DB2 table (or VSAM with a small trick) directly from your
REXX ISPF programs.
IMHO the table size you've described is better in a "DB-like" object.
Regards.
Max
Il giorno mar 17 dic 2019 alle ore 15:19 Billy Ashton <
bill00ash...@gmail.com> ha scritto:
> Hello, I am workin
I am not at all an expert on Sysplex compatibility. I bring to the table only
the knowledge of the IBM policy and years of experience with "how things work."
I think the answer is very simple: it might work and it might not. Or, more
correctly, it might IPL, and if it does, then everything YOU c
Everything depends on what your application needs to do.
1_ What type of performance is excepted from the application? As others have
stated, the number of rows will make it very slow to use
2_ how many users will be in the table at one time?
3_ Is it going to be shared in multiple LPARs or o
If you try and add an LPAR to a sysplex without having applied the relevant
coexistence maintenance, you get a 082 wait state at IPL. See
description below.
If you try to add a system that is not even in the compatibility matrix, I
would guess the same...
082
Explanation
The system entering th
Performance will kill you long before you hit the size limit. Have you
considered using, e.g., DB2, to store the data?
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
From: IBM Mainframe Discussion List on behalf of
Billy Ashton
Sent: Tuesda
I have 2 ZFS files I would like to compare. I have tried the diff command but
that has length limit of 2048 and does not compare binary files.
What else is available ?
z/OS v2.2
Thank You
*** Disclaimer ***
This communication (including all attachments) is solely for the use of the
person to
Bill Schoen's FSCP is what I used in the past:
https://github.com/IBM/zos-tools-and-toys/blob/master/fscp/fscp.rexx
Regards,
Leo
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf
Of Steely.Mark
Sent: Tuesday, December 17, 2019 12:23 PM
To
Periodic mandatory recompiles would have saved us a lot of grief. Take
companies that lost their source code - please!
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
From: IBM Mainframe Discussion List on behalf of
Charles Mills
Sent: Monda
For that many rows I would use a Database, or SQLite 3 (CBT Tape #965 and there
is compile JCL on the CBT Tape the can be used against the current version of
SQLite from the sqlite.org website). BPXWDYN makes a great interface to SQLITE
via REXX (my preference for ISPF Apps), or there is an incl
Lots of OS/360 code can still be run unchanged. Some will take an ABEND S0C4
due to accessing unassigned storage; that problem shouldn't exist for code
written for an OS/VS2 or MVS/SP system.
ISAM may be an issue for some code, as may BTAM and QTAM..
--
Shmuel (Seymour J.) Metz
http://mason.gm
Hi all
After some paramters were changed dynamically , these paremeters didn't be
changed in members .
Next IPL, we will lost these changes.
Is there any way to compare active parameters with the parameters in members?
Any thoughts/comments/suggestions would be greatly appreciated
Best
Do you need real SCSI LUNs or will a VM minidisk do? The latter is easier to
manage.
--
Shmuel (Seymour J.) Metz
http://mason.gmu.edu/~smetz3
From: IBM Mainframe Discussion List on behalf of
Peter
Sent: Monday, December 16, 2019 1:28 PM
To: IBM-MAIN@
We experienced a lot of brouhaha during Y2K preparation over the ancient and
venerable TSO Utilities: COPY/MERGE/... Their link edit dates were in the early
1970s. We had to fight up the chain of command to avoid deleting them. Of
course they work as well--or poorly--today as they did in 1999. S
On Tue, 17 Dec 2019 17:22:45 +, Steely.Mark wrote:
>I have 2 ZFS files I would like to compare. I have tried the diff command but
>that has length limit of 2048 and does not compare binary files.
>
>What else is available ?
>
What is your requirement? Would the simpler (and minimally featur
If you have CICS, use REXX/CICS. Do end users go through ISPF? There's an
overhead already.
Under CICS you have a neat interface to DB2 or VSAM with the full set of
API calls.
On Wed, Dec 18, 2019, 04:46 Al Ferguson wrote:
> For that many rows I would use a Database, or SQLite 3 (CBT Tape #9
On Tue, 17 Dec 2019 12:16:38 -0600, Paul Gilmartin wrote
>
>I believe POSIX requires text files, ...
>
I was undeservedly kind to IBM. In:
The Open Group Base Specifications Issue 7, 2018 edition
IEEE Std 1003.1-2017 (Revision of IEEE Std 1003.1-2008)
Copyright © 2001-2018 IEEE and Th
If you only care about whether they're the same, not where any differences
might be, try the cksum command. If they both come back with the same values,
they should be the same.
Mark Jacobs
Sent from ProtonMail, Swiss-based encrypted email.
GPG Public Key -
https://api.protonmail.ch/pks/look
I guess the point I was trying to make somewhat imperfectly is that while
the Linux guest might be able to access CKD disks and use SCSI protocols
that is all emulated by z/VM so it's difficult to see the advantages of
what the OP is asking.
On Wed, Dec 18, 2019, 12:39 AM Mike Schwab wrote:
> Yo
There is a product called Image Focus from NewEra Software that has that
capability.
Alan Schwartz
--
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO I
Most if not all the changes I have done with the SET Command. The one I have
always
Changed is the LNKLST, however I have always been very careful to LNKLST for
the next IPL.
Otherwise you are doomed to keep redoing the change.
-Original Message-
From: IBM Mainframe Discussion List [ma
ISPF 3.12/3.13 only accepts uppercase DSN's.
If downloading to Windows or Linux is an option there are command line and
visual diff programs there.
Charles
-Original Message-
From: IBM Mainframe Discussion List [mailto:IBM-MAIN@LISTSERV.UA.EDU] On Behalf
Of Paul Gilmartin
Sent: Tuesda
Hello.I'm Not a RACF person.
.
I am looking at the RACROUTE macro, and don't see a RACROUTE REQUEST=RESUME
option.
.
I would like to resume an end-users password by submitting a job.
Can this be accomplished by using RACROUTE macros or another Assembler
interface ?
.
Any examples would be appreci
Thanks for the information. This seems to do what I needed.
Thanks
-Original Message-
From: IBM Mainframe Discussion List On Behalf Of
Leonardo Vaz
Sent: Tuesday, December 17, 2019 11:30 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: ZFS Files Compare
You would be better off asking on the RACF list (RACF-L)
But I do not believe that user definitions are under RACROUTE - you would
probably have to code ICHEINTY - or simply issue RACF commands in the batch
job under the TMP.
(On Tue, 17 Dec 2019 21:39:01 GMT "esst...@juno.com" wrote:
:>Hello.
I thought I know this stuff, but my employer's z/OS V2.2 system is saying I
don't.
I have a simple one-step PROC which executes a COBOL program but has no keyword
parameters on the EXEC statement, like this:
//MYPROC PROC OUT=*
//MYPGM EXEC PGM=MYPGM
//SYSOUT DD SYSOUT=&OUT
// PEND
MYPG
I don't even know how to spell COBOL, but LE programs typically must have a
"/" either before or after the application's parms. LE execution parms go
on the other side of the "/", but can be null. So possibly
PARM='/PARM,DATA' is what you need. If that's the problem, I'd expect the
program to ac
I have an ISPF application that is built around a table of about 90,000
rows. Performance is acceptable, but the app is read-only except for the
overnight batch process that builds it. The build process is a bit of a
pig.
The Locate command was a problem. My first attempt was a brute force sc
On 2019-12-17 17:42, Farley, Peter x23353 wrote:
I thought I know this stuff, but my employer's z/OS V2.2 system is saying I
don't.
I have a simple one-step PROC which executes a COBOL program but has no keyword
parameters on the EXEC statement, like this:
//MYPROC PROC OUT=*
//MYPGM EXEC
Tried that, same results, BUT . . .
I found the problem and it is COBOL-related, no JCL or PARM weirdness applies.
The PROCEDURE DICISION statement did not specify the PARM-FIELDS variable in
the LINKAGE SECTION as a parameter.
The program just has:
PROCEDURE DIVISION.
Instead of th
I had build a new plex on zos 1.13, new CF, XCF CDS, SMS, RMM, RACF, Mcat, etc,
with only 1 LPAR in it.
IPL’d the same LPAR with zos 22 sysres/Mcat: OK
IPL’d the same LPAR with zos 23 sysres/Mcat: OK
IPL’d the same LPAR with zos 22 sysres/Mcat: OK
IPL’d the same LPAR with zos 1.13 sysres/Mcat
It probably depends on which parameters for what subsystem. JES2 has a series
of $D commands to display the current parameters which you can then compare to
the text in the parameter member. z/OS has a similar set of operands to the
DISPLAY command that somewhat mirror the various parameter me
On Tue, 17 Dec 2019 22:42:45 +, Farley, Peter X23353
wrote abour Adding a PARM to a PROCSTEP
where no PARM is coded:
> I thought I know this stuff, but my employer's z/OS V2.2 system is saying I
> don't.
>
> I have a simple one-step PROC which executes a COBOL program but has no
> keyword
Showzos (CBTTAPE) will report some current values
> -Original Message-
> From: IBM Mainframe Discussion List On
> Behalf Of retired mainframer
> Sent: Tuesday, December 17, 2019 4:08 PM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: how to compare active parameter with the parameter in
> m
-Original Message-
> From: IBM Mainframe Discussion List On Behalf Of
> David W Noon
> Sent: Tuesday, December 17, 2019 7:16 PM
> To: IBM-MAIN@LISTSERV.UA.EDU
> Subject: Re: Adding a PARM to a PROCSTEP where no PARM is coded
>
> On Tue, 17 Dec 2019 22:42:45 +, Farley, Peter X23353
>
Search the system log for the updating USERID and select the commands
you entered and search for the static location of those values?
On Tue, Dec 17, 2019 at 11:48 AM Jason Cai wrote:
>
> Hi all
>
> After some paramters were changed dynamically , these paremeters didn't be
> changed in members
Abend0B0 is rarely a z/OS defect, more likely an error in the program
requesting a SWA service. It can even be due to a JCL error (e.g. if you
specify SUB=MSTR and the datasets are in the wrong catalog it will
abend0B0). Check the reason code in register 15 which will give a fairly
precise cause.
On Tue, 17 Dec 2019 17:59:14 -0500, Steve Smith wrote:
>I don't even know how to spell COBOL, but LE programs typically must have a
>"/" either before or after the application's parms. LE execution parms go
>on the other side of the "/", but can be null. So possibly
>PARM='/PARM,DATA' is what y
On Tue, 17 Dec 2019 19:44:37 -0600, Dale R. Smith wrote:
>..
>Followup for those who might be interested. For COBOL Programs, parms to the
>program are before the "/" and parms to LE are after the "/".
>
What if the parm to the program needs to contain a "/", as e.g.
a UNIX pathname?
-- gil
"/" in program parms is fully supported and only gets complex for old Cobol
programs which are a special case. Normally LE parms come first,
PARM='leparm/programparm' with any chars allowed by LE for program
parms for compatibility with old cobol (OS/VS Cobol and couple of other
compilers, all
PL/1 works the same way. Been like that for a while...(JCL parms with
slash).
On Wed, Dec 18, 2019 at 1:29 PM Attila Fogarasi wrote:
> "/" in program parms is fully supported and only gets complex for old Cobol
> programs which are a special case. Normally LE parms come first,
> PARM='leparm/pr
That's a tool waiting to be developed.
D OMVS,F
D PROG LINKLIST
D PROG APF
Plus any others...
mash the output of those and compare to the parmlib members.
On Wed, Dec 18, 2019 at 12:08 PM Mike Schwab
wrote:
> Search the system log for the updating USERID and select the commands
> you enter
Heres how we do it:
Using a PARM= in your JCL.
LINKAGE SECTION.
01 RUN-PARM.
05 RUN-PARM-LENPIC S9(4) COMP.
05 RUN-PARM-DATA PIC X(100).
PROCEDURE DIVISION USING RUN-PARM.
000-MAIN.
UNSTRING RUN-PARM-DATA
Hello All,
we are facing an issue on prod and EE connection is active but CPNAME seems to
be inactive and its active on other one system in the sysplex.
Even after the IPL, the connection didn't establish and it seems like dynamic
activate will not aslo help.
EESW020T LU STATUS= ACTIV
CPNAME is inactive ?? Do EEDIAG command and verify how it shows up the
connection.
On Wed, 18 Dec, 2019, 8:56 AM Vinoth M, wrote:
> Hello All,
>
> we are facing an issue on prod and EE connection is active but CPNAME
> seems to be inactive and its active on other one system in the sysplex.
> Ev
If it's a EE connection between two LPARS
Then verify all SWNET and XCA nodes are active.
Then check the status IUTSAMEH in IP stack
Do this in both LPAR
On Wed, 18 Dec, 2019, 9:18 AM Jake Anderson,
wrote:
> CPNAME is inactive ?? Do EEDIAG command and verify how it shows up the
> connectio
That wait state only applies due to some APAR inconsistencies which don't apply
to this OP. They already have 1.13 through 2.2 running together so we can
safely assume that they complied with the new function APAR issue. I don't see
how they can get the wait 082 with just 1.13 and 2.3 (with 2.
Hi
Cross posted
I have installed zosmf 2.2. i don't see TCPIP configuration assistant on
the zosmf welcome Page.
I believe a plugin have to be imported. Is there a sample plugin that i can
point and use it ?
Peter
--
For IBM-M
59 matches
Mail list logo
