z/OS 2.3 announcement

[John McKown]

http://www-01.ibm.com/common/ssi/ShowDoc.wss?docURL=/common/ssi/rep_ca/6/897/ENUS217-246/index.html&lang=en&request_locale=en

Some niceties

When RACF is invoked to map UID(0) to a user ID, it returns the same value
defined in the SUPERUSER keyword of BPXPRMxx. This provides a consistent
mapping to the user ID owning an object, such as a file or directory, as
displayed by the UNIX™ 'ls -l' command.

== does not address other "shared" UID values -> which sharing is EVIL!

​In z/OS V2.3, RACF is designed to allow the RACF Data Set Names Table
(ICHRDSNT) and Range Table (ICHRRNG) to be specified by means of a
SYS1.PARMLIB member. Currently, these are installation-defined load modules
that are written, assembled, and link-edited by clients. Defining the RACF
Dataset Names Table and the RACF Range table as PARMLIB members will help
simplify customization and maintenance for these tables.

===

z/OS UNIX System Services supports dynamic installation exits for its
callable services. z/OS UNIX callable services, sometimes referred to as
*syscalls*, can be individually enabled to call installation exits both
before the entry and after the exit of the callable service. Installation
exit routines can be dynamically added, deleted, and modified through the
use of the z/OS Dynamic Exits Facility. Unlike z/OS UNIX process exits,
callable service exit routines are allowed to call other z/OS UNIX callable
services from within the exit routine. This enhancement enables customers
to implement callable service-level usage policies, resource auditing, and
recovery routines along with a host of other possibilities.​

​===


- IBM Java for z/OS will exploit RMODE64 to place JIT code cache above
the bar, enabled by default in IBM SDK for z/OS, Java Technology Edition,
Version 8.0.5 (IBM Java for z/OS V8 SR5).

​===
​z/OS V2.3 eliminates the need for Generation Data Group (GDG) Bias
usermod. Device Allocation is designed to optionally maintain the
relationship between relative and absolute generations of a GDG on a
job-step basis in addition to a job basis to improve the usability of JCL
job restart processing.​


​===​

​New support in JES2 enables email notification to a user in addition to
the current immediate notification when the user is logged on. Otherwise,
notification is deferred until the user does log on. With this support,
notifications can be issued to multiple email addresses and filtered based
on job return codes.​

​===​

​Removal of support of YES setting for VSM ALLOWUSERKEYCSA DIAGxx parmlib
parameter: z/OS V2.3 will be the last release of z/OS to support the YES
setting for the ALLOWUSERKEYCSA DIAGxx parmlib parameter. If you run any
software that requires the setting of this parameter to YES, the software
will need to be changed to no longer require the setting of this parameter
to YES. All IBM provided software should not require this setting. If you
have any other non-IBM provided software that requires this setting,
contact the owner of the software regarding this usage.​

​=== above means no user key CSA/ECSA/ESQA or SCOPE=COMMON data spaces.​

​For several decades, z/OS has offered two spooling subsystems: JES2
(formerly HASP) and JES3 (formerly ASP). JES2 is used by the majority of
z/OS customers and has evolved into nearly a superset of functionality over
JES3. IBM is affirming that JES2 is the strategic Job Entry Subsystem for
z/OS. New function in spooling subsystems will be primarily developed only
for JES2. JES2 supports unique features in the area of availability such as
spool migration, online merging of spool volumes, and in the area of
function such as support for email notification when a job completes and
soon in the area of security with encryption of spool data.

​===​ JES3 is moribund.

Other stuff, but not of general interest to me personally.


--
Veni, Vidi, VISA: I came, I saw, I did a little shopping.

Maranatha! <><
John McKown

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

[Clark Morris]

[Default] On 17 Jul 2017 10:34:08 -0700, in bit.listserv.ibm-main
john.arch...@GMAIL.COM (John McKown) wrote:

>http://www-01.ibm.com/common/ssi/ShowDoc.wss?docURL=/common/ssi/rep_ca/6/897/ENUS217-246/index.html&lang=en&request_locale=en

If you had a relatively small organization why would you get a z
rather than a windows, IBM i (AS400 follow-on) or unix/linux server
and would you be able to get similar capability at a similar cost? I
your organization went with non-z, especially non-EBCDIC as it grew
would it ever go to z? Given the volatility of organizations these
days, not having a good entry level offering seems to be long term
suicide.

Clark Morris

>
>Some niceties
>
>When RACF is invoked to map UID(0) to a user ID, it returns the same value
>defined in the SUPERUSER keyword of BPXPRMxx. This provides a consistent
>mapping to the user ID owning an object, such as a file or directory, as
>displayed by the UNIX™ 'ls -l' command.
>
>== does not address other "shared" UID values -> which sharing is EVIL!
>

>?In z/OS V2.3, RACF is designed to allow the RACF Data Set Names Table

>(ICHRDSNT) and Range Table (ICHRRNG) to be specified by means of a
>SYS1.PARMLIB member. Currently, these are installation-defined load modules
>that are written, assembled, and link-edited by clients. Defining the RACF
>Dataset Names Table and the RACF Range table as PARMLIB members will help
>simplify customization and maintenance for these tables.
>
>===
>
>z/OS UNIX System Services supports dynamic installation exits for its
>callable services. z/OS UNIX callable services, sometimes referred to as
>*syscalls*, can be individually enabled to call installation exits both
>before the entry and after the exit of the callable service. Installation
>exit routines can be dynamically added, deleted, and modified through the
>use of the z/OS Dynamic Exits Facility. Unlike z/OS UNIX process exits,
>callable service exit routines are allowed to call other z/OS UNIX callable
>services from within the exit routine. This enhancement enables customers
>to implement callable service-level usage policies, resource auditing, and

>recovery routines along with a host of other possibilities.?
>
>?===

>
>
> - IBM Java for z/OS will exploit RMODE64 to place JIT code cache above
> the bar, enabled by default in IBM SDK for z/OS, Java Technology Edition,
> Version 8.0.5 (IBM Java for z/OS V8 SR5).
>

>?===
>?z/OS V2.3 eliminates the need for Generation Data Group (GDG) Bias

>usermod. Device Allocation is designed to optionally maintain the
>relationship between relative and absolute generations of a GDG on a
>job-step basis in addition to a job basis to improve the usability of JCL

>job restart processing.?
>
>
>?===?
>
>?New support in JES2 enables email notification to a user in addition to

>the current immediate notification when the user is logged on. Otherwise,
>notification is deferred until the user does log on. With this support,
>notifications can be issued to multiple email addresses and filtered based

>on job return codes.?
>
>?===?
>
>?Removal of support of YES setting for VSM ALLOWUSERKEYCSA DIAGxx parmlib

>parameter: z/OS V2.3 will be the last release of z/OS to support the YES
>setting for the ALLOWUSERKEYCSA DIAGxx parmlib parameter. If you run any
>software that requires the setting of this parameter to YES, the software
>will need to be changed to no longer require the setting of this parameter
>to YES. All IBM provided software should not require this setting. If you
>have any other non-IBM provided software that requires this setting,

>contact the owner of the software regarding this usage.?
>
>?=== above means no user key CSA/ECSA/ESQA or SCOPE=COMMON data spaces.?
>
>?For several decades, z/OS has offered two spooling subsystems: JES2

>(formerly HASP) and JES3 (formerly ASP). JES2 is used by the majority of
>z/OS customers and has evolved into nearly a superset of functionality over
>JES3. IBM is affirming that JES2 is the strategic Job Entry Subsystem for
>z/OS. New function in spooling subsystems will be primarily developed only
>for JES2. JES2 supports unique features in the area of availability such as
>spool migration, online merging of spool volumes, and in the area of
>function such as support for email notification when a job completes and
>soon in the area of security with encryption of spool data.
>

>?===? JES3 is moribund.

>
>Other stuff, but not of general interest to me personally.

[John McKown]

On Mon, Jul 17, 2017 at 2:44 PM, Clark Morris <cfmp...@ns.sympatico.ca>
wrote:

> [Default] On 17 Jul 2017 10:34:08 -0700, in bit.listserv.ibm-main
> john.arch...@GMAIL.COM (John McKown) wrote:
>
> >http://www-01.ibm.com/common/ssi/ShowDoc.wss?docURL=/
> common/ssi/rep_ca/6/897/ENUS217-246/index.html&lang=en&request_locale=en
>
> If you had a relatively small organization why would you get a z
> rather than a windows, IBM i (AS400 follow-on) or unix/linux server
> and would you be able to get similar capability at a similar cost? I
> your organization went with non-z, especially non-EBCDIC as it grew
> would it ever go to z? Given the volatility of organizations these
> days, not having a good entry level offering seems to be long term
> suicide.
>
> Clark Morris
>

​I am guessing that you're really asking the above generically. Because I
agree with you. The IBM Z is _not_ for a small shop or even a medium sized
shop. The IBM i (AS/400) is a fantastic​ machine for a smaller shop,
despite its use of EBCDIC instead of UNICODE (note UNICODE != ASCII in this
instance). If you want Linux, then a "p" (RS/6000) is better, although
there is a Linux for the "i" too. And, as is likely well known, the "i" is
really just a slightly modified "p" system any way. And if you're tiny or
love complicated clustering solutions, go with Intel and Windows. [puking
up guts]

--
Veni, Vidi, VISA: I came, I saw, I did a little shopping.

Maranatha! <><
John McKown

[Scott Chapman]

On Mon, 17 Jul 2017 16:44:10 -0300, Clark Morris <cfmp...@NS.SYMPATICO.CA> wrote:

>would it ever go to z? Given the volatility of organizations these
>days, not having a good entry level offering seems to be long term
>suicide.

Exactly, there needs to be an on-ramp that starts at zero (or very nearly so). Specifically for z/OS. (Does Linux for Z survive without z/OS cash to help back new machine development?)

I know it happens on occasion, but "on occasion" doesn't seem like a long-term survival (let alone growth) strategy. It's too easy to get started on other platforms. Which also have way more mind share among the smaller/startup organizations.

Scott Chapman

[Dana Mitchell]

The hardware is called IBM Power processors (currently Power 8) and they can run AIX, Linux, and IBM i LPARs concurrently. They do have Linux only models ($cheaper) but basically the same, just like IFLs.

Dana

On Mon, 17 Jul 2017 15:04:28 -0500, John McKown <john.arch...@GMAIL.COM> wrote:
> If you want Linux, then a "p" (RS/6000) is better, although
>there is a Linux for the "i" too. And, as is likely well known, the "i" is
>really just a slightly modified "p" system any way.