Path: utzoo!utgpu!water!watmath!clyde!cbosgd!ucbvax!v1.DARTMOUTH.EDU!deh%kiewit.DECnet From: deh%kiewit.DECnet@v1.DARTMOUTH.EDU ("KIEWIT::DEH") Newsgroups: comp.os.vms Subject: RE: Problem with PRO-380 console REV 5 on 8530 Message-ID: <8801221419.AA10999@dartmouth.EDU> Date: 22 Jan 88 14:15:00 GMT Sender: daemon@ucbvax.BERKELEY.EDU Reply-To: "KIEWIT::DEH" Organization: The Internet Lines: 35 >It must be at least 6 months now since we upgraded our 8500 cpu >to an 8530 with the replacement of console (PRO-380) microcode to >version 5 (DEC call it Revision E) by swapping the 32 megabyte >RD52 disk drives. > >The problem arises when the console lose power: While powering >up, at least what I have determined (guessed) is the RD52 has not >come up to speed before the PRO-380 CPU try to access it, thus >the PRO-380 thinks there is no hard drive and try to boot from >the floppy and draw a the picture of a floppy with a question >mark on the PRO-380 screen with some error code. We have swapped >the controller, several RD52's and even a new PRO-380, the problem >still persist. Has anyone seen this symptoms before on their >8500, 8530, 8700 or 8800 ? > We had a similar problem. After a power failure, the 8530 would not boot. All indications from the PRO 380 were that the disk or disk controller were failing. Replacing the RD52 with the REV 4 disk that had been used after the 8500 to 8530 upgrade did not solve the problem; replacing the PRO disk controller did not solve the problem... The system finally did boot and the FE left, but it failed a few hours later and sat most of the night. The next day the system would sometimes boot, but not reliably. Another FE got involved and looked at the board errors that were reported by the 8530 when it did try to boot, but failed. CSC told him to solve the PRO 380 problem first, since the 8530 problems were likely being caused by bad microcode being read. He ignored them and began swapping out one board at a time. Sometimes on boot it complained about a different board, so for a while it looked like we might have multiple errors. Finally he replaced the 8530 clock module and the system booted. He put everything back the way it had been (except that the Rev 4 RD52 was left in the PRO) and everything has continued to work for the past 3 months. The PRO uses that clock signal from the 8530 during boot. ------