RE: Client Access Dropping Out at the desk but the job remains active?

Hi John,

I had exactly the same problem with one of my clients a couple of years ago.

User connects to AS/400 running 4.1 using client access - fine while they
use it, but if they leave it for a couple of minutes it "disconnects", only
the job on the AS/400 remains active until either cancelled or its dumps
because the display file couldn't write to the PC.

Same PC could connect to an AS/400 running 3.2 with no problem!!!

The client were tearing their (and my) hair out because this was part of a
CISC to RISC that was due to go live.

I checked everything out my self and then I spent an age on the phone to IBM
assist and we went through everything again - and I mean everything!!!

Finally, in a fit of desperation, I asked the client if out of hours I could
set their RSIC box to the same IP address as their CISC box (with the CISC
box down). Guess what? - it worked perfectly! Then I tried another free IP
address - this worked fine as well, in fact every other IP address worked
except the one it had been given.

We changed the IP address of the RISC box and to this date it hasn't given
them another problem.

I passed the problem with the IP address on to their network guys but I have
never received an explanation about why it happened. I had check all the
obvious conflicts etc. and the original address was amongst a range
specifically reserved for servers.

To sum up, try changing the IP address and if anybody know IP inside out,
perhaps they could shed some light on why the problem happened in the first
place.

HTH

Alan


-----Original Message-----
From: owner-midrange-l@midrange.com
[mailto:owner-midrange-l@midrange.com]On Behalf Of John Taylor
Sent: 02 March 2000 15:16
To: 'Midrange List'
Subject: Client Access Dropping Out at the desk but the job remains
active?


Long time no post.

Midrange list you are my last hope. On a regular occurrence (5-6 times a
day), we are experiencing Client Access dropping out at the desk top,
however the actual job remains active and we do not notice a problem until a
user calls us. We then manually cancel the job unlock any record locks and
away the user goes.

This is now at the annoying stage, not only for the user but for us. I have
checked the archives here (once they were made available again - thanks
David) and the IBM knowledgebase to no avail. Any help would be appreciated.

We are running a 170 at v4r3 over TCP/IP (INACTTIMO = 0) No messages or
logs. This only occurs with 4 users out of a userbase of 60 clients. (I have
also closely followed the recent "Inactivity time out" thread - which is a
similar but not identical problem)

Thanks

John Taylor
UK
+---
| 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
+---

+---
| 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
+---