henry@utzoo.uucp (Henry Spencer) (01/03/90)
A little while ago, we were having trouble talking to one of our uucp neighbors. Some investigation turned up a bizarre problem: when talking to their dial-in modem, everything we sent got echoed! It wasn't the host doing it, because during the login sequence, at times when echoing should be on, *double* echos occurred. It happened when signed on from a terminal, too, so it was definitely at their end. Otherwise, everything was fine; characters *were* getting through. I couldn't think of any reason why this should be happening. Many modems have an option to echo on the RS232 line, for use with dumb terminals, but not on the phone line. Some have a remote command mode, but this was happening in normal data mode. There were good reasons for suspecting that it wasn't the host or the interface. Clutching at straws, I suggested a close inspection of the RS232 cable. Turned out the cable was half- unplugged at the modem end, and this was the problem. My guess is that it was far enough out for the ground wire, pin 7, to be disconnected. I don't entirely understand how that would give the observed symptoms, but I do know that a broken ground line can do bizarre things. Uucp, incidentally, responded very badly to this. We run a fairly-recent HDB (aka BNU) version. Despite the unexpected echoing, it successfully fought its way through login and protocol startup! When it asked to send a file, it saw the echo of the request as the response to it, noted that this was a request and not a response, and promptly had a cardiac arrest. It fell over with a very uninformative assertion-blown message. -- 1972: Saturn V #15 flight-ready| Henry Spencer at U of Toronto Zoology 1990: birds nesting in engines | uunet!attcan!utzoo!henry henry@zoo.toronto.edu
flinton@eagle.wesleyan.edu (01/07/90)
In article <1990Jan2.235724.7287@utzoo.uucp>, henry@utzoo.uucp (Henry Spencer) writes: > A little while ago ... everything we sent got echoed! ... I suggested > a close inspection of the RS232 cable. Turned out the cable was half- > unplugged at the modem end, and this was the problem. Not long ago I found a 4 ohm quasi-short between lines 2 and 3 (TxD and RxD) in a modem cable I was trying (unsuccessfully :-) !) to use. Nothing BUT echo! Long cable for a short run, so I cut it in half. One end good, other end still showing 4 ohms. Slit and removed the molding from the connector on the bad half, pried open the shield thereunder, snipped the lead to pin 2: STILL 4 ohms twixt pins 2 and 3! After further disassembly: Turns out the cable manufacturer had used crimp pins and stranded wire (nothing unusual there) and _one_strand_ that should have been crimped with its mates to pin 2 at one of the connectors was brushing up instead against pin 3, deep within the plastic pin-cushion sandwich held together by those press-fitted D-rings. Snipped that strand off, reassembled all (not forgetting to rejoin the severed ends of the lead to pin 2) and ... well, two perfectly usable half-cables needing DB-25's at one end each. | still just | Maybe I'll wire one of 'em up as a nul-modem ... -- Fred | a babe in | | the woods! | Fred E.J. Linton Wesleyan U. Math. Dept. 649 Sci. Tower Middletown, CT 06457 ARPA/Internet: FLINTON@eagle.Wesleyan.EDU (preferred) Bitnet: FLINTON@WESLEYAN[.bitnet] (works too) on ATT-Mail: !fejlinton ( ...!attmail!fejlinton ) Tel.: + 1 203 776 2210 (home) OR + 1 203 347 9411 x2249 (work)