Steve
Sorry, as far as the z/OS Communications Server IP part of this redbook is
concerned, it's on a downhill trajectory.
Something pretty unforgivable comes to my attention and so many - thread:
"Redbook d/l issue" - appear keen to use this redbook!
Is there some product or product function which has replaced Automatic Restart
Manager (ARM) from one of the supposedly IBM subsidiaries but which has
sufficient clout that it can forbid the use of a "competing" function from
within IBM[1]?
I noticed this possible phenomenon since this much-in-demand redbook,
"Considerations for Transitioning Highly Available Applications to System z",
happens to be so cack-handed in its description of "dynamic VIPA" (DVIPA) that
the VIPARANGE flavour of DVIPA gets overlooked - in favour of the VIPABACKUP[2]
flavour which is next to useless unless a component of the VIPADISTRIBUTE
("Sysplex Distributor" or "Distributed DVIPA" (DDVIPA)) flavour.
I'm willing to consider contradictory evidence but, to my mind, the VIPARANGE
flavour of DVIPA and the ARM function[3] were made for each other as a means to
maximise the availability of server applications[4] - which is what this
benighted redbook is supposed to be all about for <insert your favourite
expression> sake. So I "scanned" for the letters "ARM" and then the phrase
"Automatic Restart Manager". Did I get a "hit"? I did not!
The only explanations can be as follows:
1. The redbook authors have been made an offer they can't refuse *not* to
mention ARM.
2. The redbook authors are more ignorant of relevant z/OS Communications Server
facilities than seems credible.[5][6]
-
As far as passing any comments on to the redbook folk are concerned, it's
pointless if they don't propose to produce updated versions in the future. *If*
there is a prospect for producing updated versions in the future, as for the
now 4-volume "IBM z/OS V1Rn Communications Server TCP/IP Implementation" series
where n is the release number, I'd be more than happy to try to help them get
it right next time. This indeed is precisely what happened regarding the R12
set of volumes and a certain dread three letters which had been corrected in
fact - but *overcorrected* (thanks to Mark Regan for bringing the nonsense to
my attention) so I had to get the *correct* usage reinstated - paraphrasing the
good Sir Walter "Oh what a total mess we create when first we refuse to employ
correct terminology!".
Incidentally I do propose corrections from time to time in the regular manuals
- and some are in train at the moment. The problem is the written versions of
"blank stares" I get back so that I am faced with yet more treacle-wading
before the Swan-Edison moment dawns!
-
> [top posting]
Incidentally, this recent vogue for adding comments following a reproduction of
the post upon which comments are being made is quite out of sorts with the
recent "enhancement" to the list, haven't you noticed? When I "hover" over a
post I'd like to see something of what the post is about.
-
[1] I could name one but I would be open to libel!
[2] Some installation may still use VIPADEFINE statements - and the design of
the installation workload may just still happen to suit the use of the
VIPADEFINE statement - and so may still think of it as the VIPADEFINE flavour -
as, indeed, the redbook authors do!
[3] "z/OS Automatic Restart Manager" for those wanting a quick
reminder/acquaintance regarding ARM:
http://www.redbooks.ibm.com/abstracts/redp0173.html
[4] Specifically the sort of server applications where only one instance can be
run at any one time. Where many instances can be run and all available are to
be employed at any one time, "sysplex distributor" suggests itself.
[5] VTAM's and SNA/APPN's support for "generic resources", "multi-node
persistent sessions" and "RTP path switches" but the explanation for that
doesn't take much working out as evidence of the usual blight of anti-SNA
bigotry, "proprietary" indeed!
[6] I haven't finished with this 258-page redbook yet but I think I've noticed,
"paging" through, that they may actually have something useful to say in some
of the sections other than the z/OS Communications Server "networking" section
which really doesn't reach "pass" grade.
Incidentally what first alerted me to this less than satisfactory understanding
of z/OS Communications Server "networking" which I covered in the previous post
was the emphasis on the "routing" capability of OSA features. The correct
resolution of that problem was simply not to bother mentioning anything
whatsoever regarding "routing" since it contributes absolutely nothing at all
to the topics under discussion.
-
Chris Mason
On Mon, 10 Oct 2011 10:00:27 -0600, Steve Comstock <st...@trainersfriend.com>
wrote:
>[top posting]
>
>When I made my post I forgot to mention the one qualm I had
>right away: 'PL/1' instead of 'PL/I'. I quickly sent off an
>email to the redbooks folks. You might do the same with your
>comments on OSA.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
