mark@cbosgd.UUCP (Mark Horton) (07/04/85)
cbosgd went down again Tuesday night when a thunderstorm came through and the power flickered. It wouldn't come back up, and could not seem to access either of the disks with a root filesystem. Wednesday morning, DEC came in and did some serious rearrangement of the massbus cables. (We suspected that the massbus had been too long ever since we added an RM05 on the end, and the RM05 would work with a 10 foot massbus cable but not with a 15 foot cable.) After tracing all the cables down, we discovered that our massbus was 130 feet long. (The massbus spec says 60 feet is the limit.) After removing two disks from the chain (which we had to give up anyway for political reasons) and shorting one cable, the chain is down to 55 feet and is working fine. We'll know for sure next thunderstorm. Fortunately, the data on the disks was not damaged this time. It seems that if you have all new massbus transceivers all along your chain, you can exceed the spec pretty far. However, as soon as you take a power hit, some of them degrade and mess up thw whole bus. We claim the record for the longest working massbus chain at 130 feet. (It worked in between DEC service calls and thunderstorms.) Before we added on the RM05, there were 120 feet of massbus, which worked reliably.