Re: MIDRANGE-L Digest V3 #730 -- MIDRANGE-L

> Date: Wed, 27 Sep 2000 14:25:11 -0500
> From: Chuck Lewis <clewis@iquest.net>
> Subject: Communications Question on Frame Relay & CTLU's
>
> Hi Folks !
>
> We are running remote locations via Perle Control Units with an Ethernet
> NIC, ditto in the AS/400 via Frame Relay.

> At from time to time a remote location will call and say they are down
> (we are in the process of installing UPS's at ALL remote locations) and
> I see NOTHING indicating that on the AS/400. I can ping the router and
> ping the card(s) in the Perle control units. But if I vary off a device
> that shows a Signon screen and vary it back on, it shows vary on
> pending. QSYSOPR also gets a message.
>
> So I have two choices:
>
> (1) wait for the system to do something. This SEEMS to take about 10
> minutes. Any have any idea how this works or how to control this ?
>
> (2) start killing jobs and reseting everything. 9 times out of 10 the
> system and I collide while doing this (it starts vary something off that
> I am varying on of vice versa)...
> 
> An APPC Device is created for each card in a Perle control unit as well
>as a Remote Control Unit.
>
>Is there an order in to varying this things on ? Sometimes it can take
> FOREVER to get APPC Device to grab the Remote Control Unit. The APPC
> Device shows Active under the ANYNETAPPC device but has none of the
> QRMTWSC entries that SHOULD be there !
>
> HELP !!!!
>
> Thanks !
>
> Chuck


   We had this problem a couple of years ago and had all morning outages that
brought production to a halt.  Through very painful trial and error, we
figured out that ending the Perle jobs, deleting the controller description,
and letting it be recreated was the only thing that fixed it, which is a
little unbelievable.  We migrated the OS/2 off of the Perle's Anynet to the
OS/2's Anynet, as I recall, and the PC's and their local printers off of the
Perle to a local AS/400 170 that networked to the ERP AS/400, and was left
with only a few line printers on the Perle.

   After that, an outage was not critical, but we still had it down to a 5 or
10 minute outage.  I don't recall everything, but after shutting everything
down and deleting the description, turning the Perle on recreated the
controller and line descriptions.....but you're right, the boot process took a
few minutes.  That was pretty much why it took probably 10 minutes overall to
recover from a lockup, which we never determined a cause for.  It happened
infrequently but spending all morning with network technicians and AS/400
system engineers while your production people are going apeshit is not a
pleasant experience, and it happened often enough that we bought an AS/400 170
to attach local twinax devices that we had configured with third party
products for certain operations.  Of course, our PC's were networked over the
WAN through a router.

   Never had a problem with the 170......:)

 Ralph
+---
| 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
+---