View Full Version : TCP Timeout
01-01-1995, 02:00 AM
Should there be any concerns with changing the following TCP/IP values. CHGTCPA TCPKEEPALV(40320) CHGTELNA TIMMRKTIMO(0) Other than the TCP/IP Connection Status looking like a big mess, what other problems could these changes cause?
10-02-2000, 12:30 PM
I'm curious as to why you are trying to assure that a session never times out? Certainly there are more times that it should time out rather than never timing out? Let us know what you are trying to accomplish with these changes and we can provide better repsonses.
10-02-2000, 12:37 PM
James, I agree with Chris. Why WOULD you not want the session to ever timeout? Doesn't this start looking pretty ugly in the DSPJOB and heaven forbid, an abnormal end of the job? I would think the enormity of a JOBLOG for this, would take forever to generate, much less read. -bret
10-02-2000, 01:38 PM
Bret, I agree you and Chris. I don't want the Telnet Keep alive set zero, but the plant I'm working with thinks differently. Here's the story. PC telnet sessions work fine, but they also have RF (900mhz Radio Freq) units that keep dropping. (This doesn't happen at all 14 plants just a few.) The RF ping response time ranges from .300sec to .1600sec. The RF vendor says the slow response is caused by the RF units swithing access point within the plant. When the TIMMRKTIMO was set to 600sec the RF's would drop every 10min. Since the TIMMRKTIMO has been set to 0sec they haven't had any problems, but I don't want to leave it that way. So, I'm trying to come up with some good info as to why it can't stay at zero and they should bring in someone in to look at their RF network. My guess is that the RF's are not responding fast enough to the keep alive packet, so they are getting disconnected. I have other plants that work fine, but their RF ping response is .35sec. -James
10-02-2000, 01:52 PM
Wow. Were using Intermek hand held/RF units and we dont have a problem with dropping on any sort of regular basis. One gets dropped or the cherry picker installing new lights crunches our antenna, and that sort of thing, but never had a problem with pinging. You don't have a lot of cardboard on the shelves in this warehouse do you? Corrugated cardboard 'eats' RF like like my kids eat M&Ms. :) Cardboard was placed around certain sites during WWII. The idea was that since cardboard abasorbed RF, they could be directionally silent by blocking off the area around the antenna array. -bret
10-02-2000, 01:54 PM
The TCPA TCPKEEPALIVE is specified in minutes. There is a lot of room between 600 secs(5 minutes) and 40320 minutes(28days). You might try some interim settings to come up with a more workable work around.
10-02-2000, 02:23 PM
Bret, Thanks for the tidbit on the cardboard. I will definitly look at it when I goto the plant. My thought is to change the TCPA TCPKEEPALV to 120min which I don't think will cause any problems, but if I changed the TELNA TIMMRKTIMO to 1800sec or to 3600sec I think the RF's will drop every 30min or 1hr. I was just curious as to what the ramifications of leaving the TIMMRKTIMO set to zero would be. I feel this is not a AS/400 issue, but a RF network issue. The plant doesn't seem to understand this, because it started happening after I upgraded them from a model 200 to a 720. Outside the RF issue they are very impressed with the performace of the new AS/400. James
Powered by vBulletin® Version 4.1.5 Copyright © 2013 vBulletin Solutions, Inc. All rights reserved.