When Does Folklore Begin 4185

re:

then there is the story about getting operating system deployed in the disk engineering lab (bldg. 14) and disk product test lab (bldg. 15). their attempts at it before hadn't been very succesful (they tried it with mvs and found it to have something like 15min mtbf). i had to nearly completely rewrite the i-o supervisor to handle the significant error conditions generated by hardware in development and test.

the side-effect was they got engineering own time-sharing service ... the down-side was i got the trouble calls when anything went wrong, even when it was their own hardware.

one such was some monday mid-morning (from bldg. 15) when they called to complain that the performance had gone all to pieces and what had I done over the weekend (and they had done nothing). turns out that they had string of 16 drives that they had put together for their own use ... and had actually exchanged an existing 3830 disk controller with a 3880 disk controller (still in test) over the weekend. after some amount of diagnostic, it was shown that there had been no software changes over the weekend, and the (performance) problems were isolated to the newly installed 3880 disk controller. after some more diagnostic the nature of the problem was characterized ... and they went off to try and figure out how to redo the hardware to fix the problem. fortunately, first customer ship was still six months off and there was some time to fix things up.

lots of past telling of various activities in bldgs 14&15