At State Farm in 1973, a new bank of tape drives were installed in late May, and they ran fine until Jun 15, when we began to see very strange tape ABENDS (starting with Fnnx as I recall), perhaps a dozen each day, that would then not occur until the next evening. After 10 days and much research by IBM, I decided to print the step records with those ABEND codes, and noticed that the time of the first instance of each day's ABEND was one or two minutes later than the prior day's first ABEND, but only up to June 22, when its first time was earlier than the first time on June 21, and subsequent days were also failing earlier by a minute or two on each successive day.
I immediately concluded it must be somehow related to sunset, so the late Tim Wuthrich and I estimated the projected time of that day's first abend, stayed late, and were adjacent to the new drives when we saw the sun come thru the window, and one of the tapes that was being read immediately started to rewind! Those 3420 tape drives had an optical sensor that read the reflection from the silver strip at the end of the physical tape, and the sun got into that sensor, causing a false detection of end of tape. Installed blinds on that window and solved the problem. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@bama.ua.edu with the message: INFO IBM-MAIN
