RE: Troubleshooting interrupted Telnet connections

["John.BresinaJr" <John.BresinaJr@xxxxxxxxxx>]

James,

We were having a similar issue here.  I contacted IBM and after running some TCP/IP traces from the IBM 'I' we found that the system was being sent a RST 'RESET' request from the network.  I then contacted our network team and found that they have 5 minute inactivity timers set on the routers/load balancer that my system is attached to.  So when I looked at the CHGTELNA command and looked at the Session Keep Alive Timeout ...TIMMRKTIMO parameter it was set to *CALC.  Apparently *CALC equates to about 10 minutes.  I changed that parameter to 240 seconds or 4 minutes so the system sends a "I'm here" message to the network to avoid their 5 minute inactivity timeout.  After changing that parameter you also need to Stop and restart Telnet services.  Now we no longer have the issue of our users sessions timing out when they are on the phone or on another application.

John Bresina Jr| Engineer | TTS Server Tech iSeries |  Target |  33 S. 6th St.| Minneapolis, MN 55402 | 612 304 3665 (ph) |

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of James H. H. Lampert
Sent: Thursday, April 22, 2010 5:29 PM
To: midrange-l@xxxxxxxxxxxx
Subject: Troubleshooting interrupted Telnet connections

We've long learned to live with having our Telnet connections to our 
customers interrupted after a relatively short period of inactivity, 
whether by a short QINACTIV, or by some intervening piece of hardware. 
In most cases, except with the one customer whose box is set to disable 
concurrent sessions, we never have a bit of trouble reconnecting and 
picking up where we left off.

But that particular customer is having the same problem with THEIR OWN 
remote users (and it's clearly NOT QINACTIV-related), and they've 
started blaming US for it.

Anybody know a good way to track down this sort of glitch?

--
JHHL