Virtual Device spontaneous combustion!

John,

One reason it works this way is so you can get QPADEVnnnn devices by 
default.  If it did not do this, you would be stuck getting VTxxx default
devices.  The device component is looking for a FREE device, since it 
will delete and recreate to change the device type/attributes if
necessary.

The following assumes we are talking TCP/IP networks ...

Before named device support (TN5250E RFC), it didn't matter, since you
could only get QPADEVnnnn devices.  I suspect you are actually talking
about named device support not working the way you expected.  True?

You can acheive the desired actions, but it will require use of Telnet
Exit programs to accomplish.  The sample exits check for generic 
compatibility, e.g. if you ask for VT100 terminal type then samples will
look for device type of V100 or V220.  Similarly, for 3270, we look for
devices of 3277/3278/3279 type.  Printers are 3812 or 5553.  Anything else
is 5250 terminal type.  If the device attributes being requested are 
not radically different from the existing ones, the device is "updated"
rather than deleted/recreated (if you are changing NLS for a device, 
then it is possible the device will be deleted/recreated).  It will be
deleted/recreated for sure if your switching generic types i.e., asking
for VT100 device which already exists as a 5250.

Using the samples, one can tweak the code to refuse any request that 
is not for a compatible, existing device.  Right now, the samples will
just create a new device, if no free existing ones can be found.  You
would have to modify the action of GetFreeDevice(), and also checking 
the return code to know if you must set Allow Connect to '0' for this
client.

Hope this clarifies things.

-- 
J.S. (Jeffrey) Stevens  AS/400 TCP/IP (Telnet/WSG/LPD)


> Date: Sun, 11 Apr 1999 16:23:52 -0700
> From: John Earl <johnearl@toolnet.com>
> Subject: Re: Virtual Device spontaneous combustion!
> 
> I don't want the system to delete devices, I just want it to leave them alone.
> If I have a 3477 defined, and someone comes along emulating a 3197, I want the
> system to do _nothing_.  Instead it appears that it will delete an existing
> device and then create a new one that will be usable by a 3197 device type.   
>I
> should be able to control this behavior shouldn't I?
> 
> jte
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---