Radoslaw,
2)
The restriction isn't in the SAN fabric, or anything to do with FC as I'm not aware of any method in
the FC protocol to actually restrict this on a frame basis. Ficon does work over 2 hops and is
actually supported by IBM in some very specific circumstances/configurations,
Natasa,
I case you haven't fixed this.. one possibility is that if you are using 2 byte link addresses ( ie
cascading) and your directors haven't been configured correctly you will get this status. Check your
insistent Domain ID setting and the fabric security policy.
Cheers
Roy
Natasa
You can reboot the SE remotely, I did this last week on one of our z10s. You just need to use the
Reboot SE option from the HMC. If you try to reboot from a single object session you get the message
cannot be done remotely.
Also you can easily find the SNA names of all CPCs on the same HMC
Todd,
Have you enabled Routing in the HMC network configuration, there is a check box for this, just turn
it off. Also check in Network Diagnostic Configuration to show what UDP ports are listening.. is 520
there?
Cheers
Roy
Todd Burrell wrote:
I got the following info from one of our
Frank,
I think poorly documented is giving it far too much credibility. Not only is the interface
inaccurately described, but if you only use the function calls that are documented then your wasting
your time as you are unlikely to get anything useful out of it - I can't understand why they
Bruce,
This should be no problem, as you are not effecting all of the paths to the devices. The devices
would only need to be offline if all the paths were changing.
To make a change like this all that you need to do is ensure that, at a minimum, all the effected
paths are offline to the
Chris
Chris Mason wrote:
Jorge
Well, I'm glad you've worked it all out.
Now the only mystery for those following this thread in the list is how your
F RESOLVER,DISPLAY
and
//SYSTCPD DD DSN=...(TCPDATA)
do *not* correspond to the resolver trace output you posted before! I even
found where
Jorge,
rather than guess what might or might not be happening, just allocate the SYSTCPT DD and it will
tell you exactly what resolver information you are using and also show where the nslookups are being
resolved against
from TSO just do
ALLOC FI(SYSTCPT) DA(*)or allocate to sysout and
Jorge,
so this shows that your resolver parameters are coming from the local TCPDATA dd statement but is
not showing the actual address resolution
Where did you allocate the SYSTCPT?.. it needs to be in the address space that is doing the
nslookup/ping etc.. Is T99TCPIP the userid of the
It would be even better if IBM stopped using the default of /Service for everything zos, cics,
db2 etc.. It would be a bit smarter if the default was something unique for each order
/service/orderno (with 'orderno' being the actual order number
or maybe
/service/zos/orderno
me laugh - wait
until Friday to make nonsense statements like this.
On Mon, Nov 30, 2009 at 1:43 PM, Roy Hewitt
ibm-m...@frozen.eclipse.co.ukwrote:
...snip
IBM come up with a decent design on how to install and maintain z/OS
...snip
Guy,
Perhaps I didn't explain it very well, I wasn't talking
Ted MacNEIL wrote:
Nothing beats the integrity, reliability, and security of mainframe.
(Of course, I'm biased)
Of course you're biased.. (and Ted I'm not directing this at you, but your comment is typical of
many I see).. We're all biased on this list (typically!!) but a lot of the time to
McKown, John wrote:
-Original Message-
From: IBM Mainframe Discussion List
[mailto:ibm-m...@bama.ua.edu] On Behalf Of Kirk Wolf
Sent: Thursday, November 19, 2009 11:45 AM
To: IBM-MAIN@bama.ua.edu
Subject: Re: XMITMGR for 64 Bit Windows?
It would be very nice, IMO, if there were an
McKown, John wrote:
-Original Message-
From: IBM Mainframe Discussion List
[mailto:ibm-m...@bama.ua.edu] On Behalf Of Roy Hewitt
Sent: Thursday, November 19, 2009 3:34 PM
To: IBM-MAIN@bama.ua.edu
Subject: Re: XMITMGR for 64 Bit Windows?
snip
John
I know you always like a challange
Anthony,
When you say DWDM, was it just a pure optical extender, or something like CNTs USD device. For the
former I would not expect the CHP to come online, but the USD is defined with the relevant CU
information for the DASD at the remote end. I can't remember the exact details but the local
Radoslaw,
R.S. wrote:
Guy Gardoit pisze:
Looks like HCD TEST result answered your question. Since you're not
deleting anything, I'd trust that result.
Well... I don't trust the results g
I understand TEST answers as :
1. Failure. You cannot do it.
2. Success. You can TRY.
your right - a
Hi Phil,
As mentioned previously the only thing that really matters is the FC domain ID, it's sometimes
called the switch ID, but just to be awkward HCD allows a separate switch address. Yes you can make
them different, but unless you're after a headache always make them the same.
In HCD
Ben, Rex
I must admit it was a few years back when I last had to do this in anger (in a darkened machine room
with a mag light!).. It was a z800 and you had to perform the Power On function from SERVICE
before you could do a POR or Activate - I remember having a complete memory failure for the
Do I need to use a more powerful signon than SYSPROG? I don't see any power
type ICONs on the HMC or when I go into Single Object Operations.
--
John McKown
John,
When you say power off/on are you talking about removing/restoring the mains supply to the box?
then I think you will probably
Richards, Robert B. wrote:
About 20 years ago, I dabbled here and there with VTAM, NCP, EP, NCCF,
NLDM, et al. My boss would like me to pick up current TCP/IP skills.
Knowing this august group does not lack in providing excellent advice, I
am seeking recommendations on the best course of
yes you can
for this type of info I find it is best to refer to the Redbook
IBM System z Connectivity Handbook SG24-5444...
http://www.redbooks.ibm.com/abstracts/sg245444.html?Open
or.. just try it in HCD..
Cheers
Roy
GIONFRIDDO MICHELE wrote:
Hi
I have a Z9 mainframe
Michael,
Have you checked the Ficon status on HMC/SE from the Channel Problem
Determination info, also the
Serial Link option will show LP status from host to each dasd CU.
Once the device is online what does a DS Q,,RCD and RDC show?.. does it
work?
On the switches, I'm presuming the
Chris Mason wrote:
Bill
The key word here may be exit.
(much snippage..)
Thus, given the assumptions above, it is no surprise to me that RACF have
not revisited the supplied assumed to be *sample* exit in order to ensure it
conforms to the best coding practices probably several years on
Shmuel Metz (Seymour J.) wrote:
In listserv%200901261716349266.0...@bama.ua.edu, on 01/26/2009
at 05:16 PM, Andy Wood woo...@ozemail.com.au said:
Without a better understanding of what you are trying to do
As I wrote in my original message, I'm trying to get hard coded tests of
specific
Yukus, Mary J CIV USMEPCOM wrote:
I have looked in the HCD documentation but can't find an answer on the
following.
If the only change you make in a configuration is adding an esoteric to some
devices, can you get by with only building a new IODF file or do you need to
create an IOCDS and do a
R.S. wrote:
Roy Hewitt wrote:
[...]
I asked because not excluded is where the number comes from. I looked
at a 2nd CSS on one of our boxes and put a P next to it to show the
partitions and some partitions with a name of * are defined.
Radoslaw,
When you did it from the other machine
R.S. wrote:
I have questions regarding OSA Express3 1000Base-T, FC3367.
There are 4 copper ports per card.
Q1: How many CHPids has the card?
2/4 ?
2
Q2: (assuming 2 chpids per card)
In such case I have to define both ports per chpid as same type.
For example I must have a *pair* of ICC
Mark Zelden wrote:
On Fri, 2 Jan 2009 15:28:54 +0100, R.S. r.skoru...@bremultibank.com.pl wrote:
Mark Zelden wrote:
On Fri, 2 Jan 2009 14:25:52 +0100, R.S. r.skoru...@snip-it.pl wrote:
I found strange thing in HCD.
When type action character U against CSS (or SMP-type CPC), you get list
of
Art,
Arthur Gutowski wrote:
On Thu, 18 Dec 2008 23:10:15, Roy Hewitt ibm-
m...@frozen.eclipse.co.uk wrote:
it's still STP that you need... why not have a look at the redbook.
Server Time Protocol Implementation Guide, SG24-7281-01 (latest draft)
[...]
The NTP ability of STP (via the SE
Edward Jaffe wrote:
Arthur Gutowski wrote:
If I read correctly, the OP later clarified he currently has no
sysplex timer - single CEC, multiple LPARs - i.e., all monoplex. If
that is true, then I read nothing in STP Implementation that states
SIMETRID with an external dial-out to the atomic
Richard Peurifoy wrote:
Roy Hewitt wrote:
Last time we discussed this..(check the archives it comes up quite
regularly..) I wondered whether the TOD clock Steering Facility (aka
instruction PTFF) is only *fully* valid on machines with STP, (I know
you can call the query request form of PTFF
Hal,
Hal Merritt wrote:
I thought I read that while z/os cannot get time as a NTP client, the
z/10 box can with some NTP software on the HMC. That, in turn, sets the
z/10 clocks and, I think, that percolates up to z/os.
There was some discussion on what was required, and I was wondering if
R.S. wrote:
Not necessarily. 1000 FICON connections sounds seriously, but in fact it
means 2000 connectors clicks. It has to be done be some human - hasn't
it? So the only additional task for the human is to note cable number
and port number. 2000 records. IMHO spreadsheet is sufficient as an
R.S. wrote:
Roy Hewitt wrote:
R.S. wrote:
Not necessarily. 1000 FICON connections sounds seriously, but in fact
it means 2000 connectors clicks. It has to be done be some human -
hasn't it? So the only additional task for the human is to note cable
number and port number. 2000 records. IMHO
Martin Packer wrote:
My mantra is machines should be self-documenting and in that vein I
perceive value in RMF learning how to get things like:
Installed Memory and how much of that is purchased.
HSA Size (and yes I know it's fixed 16GB on z10)
How much memory each LPAR has reserved.
Martin
Hal
The current active IOCDS is a bit of a misnomer. It refers to that fact
that it will be the default IOCDS used for the next POR. You always have
to set it manually.
The important thing to remember is that the IOCDS is totally separate
from the IODF and dynamic actvate processing.
Radoslaw,
R.S. wrote:
1.
I read in some Redbook that I should mark ISL ports as occupied.
From the other hand I can connect the switches using action character W
against the port. The actions above are mutally exclusive.
What's better?
Doesn't really make any difference, it is purely
Radoslaw
Go for option 3 .. it works and simplifies things. I've done it that
ways for OSA/ICC etc...
Although it appears fake (well perhaps in HCD) it is a genuine
definition. The CU statement is only ever processed when building the
IOCDS or doing the HW side of an Activate - and during
Shane wrote:
Nice green stripe.
Big page only 1 Meg. Interesting.
HiperDispatch ... way overdue.
16 Gig HSA.
anything else of interest ??? ... :0)
There is no ETR feature listed.. so I presume this is STP only..
Roy
--
For
Paul [EMAIL PROTECTED] wrote:
One of my disappointments with TSO TRANSMIT is that I know
no safe way to imbed INMR text in an instream data set:
the format seems to allow any digraph to appear in columns
1-2. But today, attempting to test a trial-and-error routine
to scan a TRANSMITted IEBCOPY
Paul Gilmartin wrote:
On Tue, 29 Jan 2008 09:33:23 +, Roy Hewitt wrote:
See the following as an example of how to include such data instream
using standard IBM utilities. This example includes Terse output (but
could easily be Transmit), that has been reformatted as 64 bytes and
offset
Paul Gilmartin wrote:
On Tue, 29 Jan 2008 19:29:47 +, Roy Hewitt wrote:
Hmmm. The largest divisor of 80 I could use is 40. This cuts the
efficency to 50% from your 80% with TERSE. That's still acceptable,
and I don't need TERSE which is not yet a standard IBM utility at
all supported
A good starting point is here...
http://www.ibm.com/developerworks/linux/linux390/whatsnew.html
Binyamin Dissen wrote:
I have been looking around but have not found anything current.
I am particularly interested in looking how LSM is integrated.
--
Binyamin Dissen [EMAIL PROTECTED]
George,
Yes you could define them as 9490, and as you have discovered you will
need to install various modules to install the HCD UIMs, as HCD does not
understand '9490'
However, there is little point in doing this as you can just as easily
define them with Control Unit 3490 and Device Type
Miklos Szigetvari wrote:
Hi
We have a one pack RESCUE system generated under z/OS 1.4
and the actual IODF generated under z/OS 1.7.
I copied this new IODF to the RESCUE, but got a B0 wait state.
(saying IODF is in a higher version as the system attempting to use )
Can I solve this without
Roy Hewitt wrote:
Miklos Szigetvari wrote:
Hi
We have a one pack RESCUE system generated under z/OS 1.4
and the actual IODF generated under z/OS 1.7.
I copied this new IODF to the RESCUE, but got a B0 wait state.
(saying IODF is in a higher version as the system attempting to use )
Can I
Radoslaw,
Simply answer is yes.
For any CU, you can define whatever subset of devices you need, whether
base devices or aliases.
Just ensure that the CU definition is defined for maxiumum devices.. ie
00.256
R.S. wrote:
I need to define DASD array to MVS.
It will be escon connected, CU
John
McKown, John wrote:
Bottom line: Why did the S/A IOCP allow me to create a 2086 IOCDS on a
2096 machine?
I presume you are referring to the SYSTEM keword on the ID statement, It
is probably ignored by SA iocp, as its only purpose is to allow machine
type verification of
Dave Kopischke wrote:
On Tue, 5 Jun 2007 11:28:42 +0100, Roy Hewitt wrote:
You dont even need to create the IOCDS, remember the IOCDS is only ever
used at POR time...so if PORs are hard to come by, then you can defer it
until you really need to...
If you don't create and activate a new
Andy,
Simple answer is yes.. Just think of the OSA-ICC as an internal 2074..
works ok for both MVS console and VTAM access (ie TSO). I would imagine
that the code base from a 2074 was used to build the ICC function - if
you've used 2074 before then you'll recognise the very similar config
Roger,
Sure, even deleted a few in my time!.. Apart from IPL it isn't used, or
even allocated by the system until you come to do an activate...so if
you do rename it you might have some difficulty Activating...
May I ask why you need to rename?
Roy
Roger Lowe wrote:
Hi Listers,
to the correct name, then do another soft activate... shouldn't be
any need to IPL...
Roy
Roger Lowe wrote:
On Tue, 6 Feb 2007 12:30:43 +, Roy Hewitt [EMAIL PROTECTED]
wrote:
May I ask why you need to rename?
Roy,
Have a 1.4 system that is a totally separate/isolated environment
Have a 1.4 system that is a totally separate/isolated environment that
doesn't share any dasd with the 1.7 systems I have on the same CEC. All
HCD work is done on the 1.7 system and a copy of the IODF is shipped
across to the 1.4 system. For some unknown reason, when the IODF V5 was
unloaded
Steven,
You can, you need to apply some toleration maintainence. You can IPL and
activate from a V5 iodf, but you can't use HCD to update it.
From the z/OS 1.7 Migration guide GA22-7499-08
Other system (coexistence or fallback) requirements:
To read from, IPL with, and dynamically activate
Robert,
Forget the Activate, you're gonna need an IPL!.. 3286 just don't support
dynamic activate.. no way around it that I know of.
(Last time I got caught out by this, I deleted a whole bunch of old 3274
entries without looking too carefully at all the OSConfigs.. then
started activating
Neubert, Kevin (DIS) wrote:
I'm not certain about z890/z990 and beyond, but if the device does not
support dynamic capability it certainly appears a power-on reset will be
necessary.
Kevin,
A POR is not usually required, it's only the definition in the OSCONFIG
that causes the problems.. so
Edward Jaffe wrote:
Clear as mud. Conflicting statements. It sounds as if Brian Peterson's
speculation is correct. Specifically, some hardware/software handshake
functionality, that got backed out because it never really worked, was
prematurely documented in z/OS MVS System Commands. I
Howard,
If you've only changed OS config details, then that's normal that you
would get will get the 100 etc..
How did you test the config, HCD or MVS command.. and what syntax did
you use?
Providing you DON'T do anything to the existing IODF10.. then you should
be ok with what your doing, ie
Edward Jaffe wrote:
Hal Merritt wrote:
We are converting everything to z/os.e but running in the roadblock of
needing a POR to change the partition name. I am told that, under z/os
1.7, I can 'rename' a partition presumably via a dynamic change.
Has anyone verified this? Has anyone actually
Schiradin,Roland HG-Dir itb-db/dc wrote:
Hi,
in same cases I seen a NEDCLASS X'0C' but IHACDR contains only
the following definition
NEDCUNSP EQU 0 Unspecified
NEDCDASD EQU 1 DASD
NEDCTAPE EQU 2 Magnetic tape
Ed
I think your undestanding of the both the Sysplex Timer and STP is a
little incorrect.. comments inserted..
Ed Gould wrote:
Tim:
Just curious about the offering. IBM seems to think (if I understand
what is being offered correctly). That most mainframes are hooked up to
the internet.
Paul Gilmartin wrote:
In a recent note, Roy Hewitt said:
Date: Tue, 14 Nov 2006 11:24:19 +
Ok, so the price might have been on the high side, and I *fully*
understand the frustrations of single-CPC sites and external time
syncronisation, but remember the primary function
Ted MacNEIL wrote:
This is not one of my strengths.
;-)
I have read all the redbooks on it.
try this one..
http://www.redbooks.ibm.com/abstracts/sg246297.html?Open
Networking with z/OS and Cisco Routers: An Interoperability Guide
My understanding is that TCP/IP on z will/can use
Howard,
A bit more info would be useful to be able to help
1) You created IOCDS A3 from IODF05. Correct?
2) Did you POR or Activate.
3) If Activate, from which LPAR.
4) What config is in use in each system. Issue a D IOS,CONFIG, paste
the reply
5) What IPL parms issue D IPLINFO paste reply
Ed,
it's actually a small reprieve.. it was previously announced to go end
of June 2006..
Roy
From today's announcements
906137 Hardware withdrawal: IBM Sysplex Timer Model 2 and Sysplex
Timer LED/Laser Extender -- Replacement available
Mark,
There's something called EZANS, written by an IBMer (Alfred
Christensen). It is a rexx and panels that issues Netstat commands and
then browses the output. you can filter on various parameters, very useful.
I'm not sure where you can download it. I think it might be from a
redbook,
Sam,
I wouldn't hold out much hope with the SA/390 version. It still burns a
large amount of CPU at startup. From what I've seen this is due to the
size of the IO config i.e. how big the IODF is. I get the impression
that ESCM reads the IODF and builds a various internal tables and cross
Bill,
I agree its a size thing, but the ironic bit is that the creating of
these tables is usually a waste of time as the size of the zseries
complex increases.
Why? Well, this internal representation of the IO config that is
maintained by ESCM is only really used if you want to use ESCM to
Hal,
The best place to check this is the IOCP manual for the specific
processor type, one of the appendixes has a list of machine Limits.
These manuals are available on Resourcelink. Also check the zSeries
Connectivity Redbook..
Your right that 16k is the maximum device limit on a Ficon
Ted MacNEIL wrote:
One of the datacenters is using Linux under IFL. 2-4 images. Is it really cost
effective to buy z/VM for this poor z/800 machine ? Is it really needed at all ?
I think it is.
4 images isn't worth $125,000 USD for an IFL.
How do you work that out? Surely the number of
Ted,
How do you work that out?
$125K is what an IFL costs.
I wasn't referring to the purchase cost, I meant how do you calulate
that 4 images isn't worth $125,000 USD for an IFL. i.e when does it
become worth it?
Surely the number of Linux images is
irrelevant, the only real purpose
Howard,
There is a way to define devices/CUs via some macros exits but it's
primarily for vendors, and having looked at it i would suggest that's
not what you want to do.. (See z/OS Device Validation Support..).
You should be able to use Control unit of 2105 as far as HCD is
concerned, it
Chris Mason wrote:
Luis,
When I saw your post - for the second time - I hadn't a clue. But I asked
Google and Google told me the following:
quote
The Cisco IOS for S/390 command class is AC#CMD. Cisco IOS for S/390 uses
the SAF interface for general resource class checking.
/quote
This was
Richard,
Thats something like what I tried with FDRAPPL, and as you say the disk
datasets end up with generated names, but it doesnt work when the output
is on SMS volumes. neither will it allow an output datsaset to exist on
any of the input volumes.
Thanks
Roy
Richard Pinion wrote:
Take
Jack,
I've tried CATDSN, but although it identifies the datasets, DSF still
requires the volumes to be specifed in JCL and I dont want to do this
Thanks
Roy
Jack Kelly wrote:
i think that the secret is 'select catdsn'. pretty sure tha tuse to
work for me
//RESTORE EXEC
Howard,
Yes, in as much that you can recreate an IOCP deck from the IOCDS and
then import that into an IODF. However, it will only have the minimal
Partition, Chpid, Cntlunit, and Device details. Items such as
descriptions, serial numbers etc will not be there neither will any
OSconfig
Bruce,
thanks for the confirmation.. I started out many years ago on FDR and
the volume-orientaion was just something I accepted - the number of DASD
was a lot smaller, no SMS etc. (In fact its 'volumeness was always
something I missed when i was at a site - I'm contractor - and trying to
do
Micheal,
I'm still a bit confused? are you talking about the cu groups within the
USP, or CU definitions within HCD (ie CUADD=). Yes they are related, but
I dont understand what you mean by 'Logical CUs accessing the same
disk'. AFAIK each disk (ie 3390) is defined to only one CU image within
Ron,
When you say 'genned as 3990' I presume you mean how the USP interface
emulation has been defined.
Roy
Ron and Jenny Hawkins wrote:
Michael and I have talked offline and your comment regarding toleration for
2105 devices genned as 3990 may be what he is looking for.
Ron
BTW the
Matt
There are a couple of things I guess I don't understand about this
process. In a normal situation, I have a hardware configuration that
was loaded from an IOCDS at POR, and an IODF that matches it. Then I
can activate a new IODF which is different from the currently loaded
config,
hardware is emulating a 2105 controller. That's what I thought
you meant.
Ron
-Original Message-
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Roy Hewitt
Sent: Tuesday, 4 April 2006 9:53 PM
To: IBM-MAIN@BAMA.UA.EDU
Subject: Re: 2105 Question
Ron,
When you say
Matt,
The simple answer is yes, you need to do a POR with your correct IOCDS
However, how desparate are you at avoiding an POR...
If you have a test partition on the same machine you might be able to do
something, as you only need one of the lpars to be in sync to be able to
do dynamic
Dave,
If by processor you mean the sZeries CPC, then your understanding
might be a little off...
My understanding is that the timer just sends timing signals to the CPC,
it is unaware of what the CPC is doing. Basically it is a one way
process. Whether there is a CLO or not, the 9037 never gets
Paul Gilmartin wrote:
What initially sets the content of the TOD clock at IPL (or
is it POR) time. IIRC, there is some setting that can only be
accomplished at POR. Is there a clock in the HMC, so the
data path for that initial setting might be:
ETR to HMC (at POR)
HMC to TOD (at
BTW: 3494 Library ID is not hardcoded in metal. It can be easily changed
from operator panel (although requires Service logon). It is quick and
easy. Is it the option for you to chaneg LIB ID instead of IODF ?
BTW2: AFAIR It is not required to code LIB ID in IODF at all. Having ID
is more
R.S. wrote:
snip
I'll try to rephrase my doubt again:
IODF constains definition fo three CPCs: CPC1 CPC2 and CPC3.
All three CPCs are same type (i.e. 2064-1C6), but each of them has
different I/O devices connected. To be more complex, Each CPC have
LPAR1, and LPAR2 defined. Note, the same
I'm not sure I completely understand the question.
When you ask Question: How system recognize relevant CPC definition?
are you referring to the creation of the 1)IOCDS or 2)at IPL time? If
its 2, then CPC information is not used at IPL time. You dont need CPC
information in the IODF to IPL,
Radoslaw,
I can't remember where I saw this but they are abbreviations for:-
NTV New Test Vehicle
FTV Fibre Test Vehicle
STV Spanned Test vehicle
I am presuming from the names that they are used during hw testing
during zseries development.
Have a look at IOSDPATH or CBDZHCEX in
Sam,
I came across this the other week after an upgrade. Our problem was that
on the image profile we had Initial processors set to 16 and then
Reserved set to 1, total 17 thus blowing the 16 limit, we set the
reserved to 0 and all was ok..Check you image profiles
Roy
Knutson, Sam wrote:
89 matches
Mail list logo
