Reply to post:

Bright spark dev irons out light interference

Anonymous Coward
Anonymous Coward

The comms front end processor for a mainframe service had problems. Something about one customer's Remote Job Entry terminal kept finding esoteric holes in the software.

This was in the days when network protocol monitoring consisted of whatever could be stored on one screen. Even at low line speeds that was not a lot - and not enough to allow for the delay in the customer ringing to complain.

A new home toy was a luggable VHS camera and recorder - so it was set up to video the line monitor's screen. I taped my analogue watch to the side of the screen so that we had wall-clock time on the recording. This worked well - and a bonus was that the ringing of the telephone was caught on the soundtrack.

Using fast forward/reverse and slow motion allowed me to see what had happened at the point of failure. It became clear that there were often prolonged bursts of corrupted packets - the protocol struggled to get some useful throughput.

It transpired that the RJE terminal was connected to its modem by line drivers. These were interconnected by a long piece of unshielded cable - that was laid across the floor of the arc welding workshop.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2019