Re: Device names and interactive subsystem descriptions

Bob might be right. The cursor in the upper left hand corner means the system is communicating with the device (usually anyway) so really it's waiting for a subsystem to allocate it to provide a sign on screen.

Some more obscure things to check:

Is the signon screen DDS the default provided by IBM ? If not is it in the system portion of the library list so the subsystem can find it?

Is there a workstation type entry for *ALL? What are it's attributes? Most will will job description *USRPRF, Maximum active jobs at *NOMAX, and control job at *SIGNON. (the workstation entry can override the last setting)

Is there a workstation type entry for *CONS? It should match the *ALL entry with the exception that control jobs at should be *ENTER.

It sounds like a workstation type issue at this point, assuming the vary on is not the problem.

Jim Oberholtzer
Chief Technical Architect
Agile Technology Architects


On 2/6/2012 11:03 AM, Bob P. Roche wrote:

Vary them back on?




From:
John McKee<jmmckee@xxxxxxxxxxxxxx>
To:
Midrange Systems Technical Discussion<midrange-l@xxxxxxxxxxxx>
Date:
02/06/2012 11:01 AM
Subject:
Re: Device names and interactive subsystem descriptions
Sent by:
midrange-l-bounces@xxxxxxxxxxxx



OK, I varied off a device. Then removed the WSE fro both subsystems.
Added the entry back, to both with *SIGNON.

Result is that the cursor is in the upper left corner on the display.

What else am I missing?

John McKee

On Mon, Feb 6, 2012 at 10:42 AM, Jim Oberholtzer<midrangel@xxxxxxxxxx>
wrote:

> The subsystem where the entry is set to *SIGNON should allocate the

device.

>
> Verify that you do not have a workstation TYPE entry that is getting in
> the way somehow. I think that's option 5 on the WRKSBSD screen.
>
> Consider trying:
> As Scott suggests, change all the workstation entries to *ENTER
> Vary the suspect devices off.
> Change the workstation entries you want to allocate the devices back to
> *SIGNON
> Vary the devices back on, start with one at a time.
> Verify they get allocated by the correct subsystem.
>
> If they do not chances are the workstation name entry on the telnet
> session is not set correctly, the workstation type is not what your
> expecting, or the device is damaged, in which case it should just be
> deleted and let the system recreate it.
>
> Jim Oberholtzer
> Chief Technical Architect
> Agile Technology Architects
>
>
> On 2/6/2012 10:32 AM, Ingvaldson, Scott wrote:

>> IIRC *SIGNON should be used for the primary subsystem that you want the

workstation to be allocated to and *ENTER is used to allow TFRJOB to
change the subsystem an interactive job runs in. On the other hand, if a
workstation is allocated to one subsystem it can be tricky to change it
with the WSE's.

>>
>> You can try changing the WSE for both subsystems to *SIGNON, then

signing on manually and using TFRJOB to put the workstation into the
correct subsystem. Once that is done use *SIGNON for the primary
subsystem and *ENTER for the other.

>>
>> Hopefully you don't have too many of these.
>>
>>
>> Regards,
>>
>> Scott Ingvaldson
>> Senior IBM Support Specialist
>>
>>
>> -----Original Message-----
>> From: John McKee [mailto:jmmckee@xxxxxxxxxxxxxx]
>> Sent: Monday, February 06, 2012 9:06 AM
>> To: Midrange Systems Technical Discussion
>> Subject: Re: Device names and interactive subsystem descriptions
>>
>> I checked the entry of C0* in both subsystems. Both are C-number zero.

One is *SIGNON, and the other is *ENTER.

>>
>> Both are using the same job description and libraryand both are set to

*NOMAX for maximum jobs.

>>
>> Still doesn't work. If a different workstation is used, all is well.
>>
>> I looked at the job logs for the two subsystems, BINTER and BINTER2.
>> Nothing there except entries where I changed the WSE.
>>
>> Both subsystems are active.
>>
>> What next?
>>
>> John McKee
>> On Fri, Feb 3, 2012 at 5:35 PM, Jim Oberholtzer<midrangel@xxxxxxxxxx>

wrote:

>>> > The subsystem does not need to be restarted for a workstation entry

to

>>> > take effect.
>>> >
>>> > I am confused about the "C0". Workstation IDs cannot start with a
>>> > number. Did you start them literally with "C0"
>>> >
>>> > In any case the subsystem does not need to restart to enable a
>>> > workstation entry. One thing that might be happening. If another
>>> > subsystem grabs the device and allocates it, then it might hang

onto

>>> > it until the device is varied off and back on. I just tried that

on

>>> > our system at iDevCloud and was able to mimic what you are

describing.

>>> > Try varying the devices off and then back on with the new subsystem
>>> > entry active.
>>> >
>>> > I would also look into the other workstation entries to be sure

there

>>> > is not a similar entry out there.
>>> >
>>> > Jim Oberholtzer
>>> > Chief Technical Architect
>>> > Agile Technology Architects
>>> >
>>> >
>>> > On 2/3/2012 5:11 PM, John McKee wrote:

>>>> >> For a long time, there were standards where I work as to naming
>>>> >> conventions for devices. But, since the system is on the way out

in

>>>> >> maybe a year, other people are dictating things.
>>>> >>
>>>> >> In the case of the above, the person who assists others

establishing

>>>> >> a remote connection through the firewall has decided that it

would be

>>>> >> easier if the device name was same as the user id.
>>>> >>
>>>> >> We have two interactive subsystems. Each has a number of

workstation

>>>> >> entries - composed of three characters followed by an asterisk.
>>>> >>
>>>> >> I had forgotten about those entries when I created device
>>>> >> descriptions that start with C0 (a digit zero).
>>>> >>
>>>> >> My day off today, and I received a call from work that the remote
>>>> >> users only got a blank screen. So, I added the C0* entry. I was
>>>> >> told later that it still did not work.
>>>> >>
>>>> >> I am guessing that the subsystem needed to be restarted for the
>>>> >> change to take effect. Am I right, or is there something else

that

>>>> >> needs to be done?
>>>> >>
>>>> >> John McKee
>>>> >> --

>>> > --

> --