NetNews Usenet Archive 1992 #20/NN_1992_20.iso/spool/comp/dcom/modems/13201

Internet Message Format | 1992-09-10 | 1.8 KB

Xref: sparky comp.dcom.modems:13201 biz.comp.telebit:911
Path: sparky!uunet!cs.utexas.edu!sdd.hp.com!nigel.msen.com!emory!mathcs.emory.edu!km
From: [e-mail] (Ken Mandelberg)
Newsgroups: comp.dcom.modems,biz.comp.telebit
Subject: Worldblazer loses session stats & modem drops
Message-ID: <[e-mail]>
Date: 10 Sep 92 14:18:25 GMT
Sender: [e-mail]
Reply-To: [e-mail]
Followup-To: comp.dcom.modems
Organization: Emory University, Dept of Math and CS
Lines: 29

I've been trying to track down a problem involving frequent modem drops
between two WB's. These are V32BIS/V42BIS sessions. There really isn't
very much info to get from the modem to find out what caused the
drop, except for the S74 sessions stats. S74 should at least tell you
whether the line conditions were bad enough for retransmits.

However, even that meager info doesn't exist half the time. Although I
don't see this in the doc, experimentally I find that if the remote
modem causes the drop, S74 is cleared. If the local modem causes the drop
S74 is preserved until the next connection is made.

Does anyone know why S74 is getting cleared, or how to prevent it.
Is there any other diagnostic info I can get after the session drop
from the local side that would be helpful in tracking the problem.

I suspect that there is some undoucmented command for retreiving some
finer diagnostic info that Telebit is not telling us about. There was
on the TB+ (ATJ6J0).

Incidentally, both sides run with S10=30, which should be high enough
that line noise doesn't look like a short carrier loss and cause the
drop.

---
Ken Mandelberg      | [e-mail]          PREFERRED
Emory University    | {rutgers,gatech}!emory!km    UUCP
Dept of Math and CS | [e-mail]              NON-DOMAIN BITNET
Atlanta, GA 30322   | Phone: Voice [phone], FAX [phone]