ferguson@BKNLVMS.BITNET (02/17/88)
I have a DSP80, and a CDC 300 Mbytes storage drive, and have been having
a similar problem with disk errors. It's been happening for some time,
and I've swapped disk controllers, CPU boards, CPU extensions, etc.
many times.
I also often get a crash with the error:
floppy drive has been opened or storage module has been stopped.
This seems to coincide with the block header errors on my disk.
WARNING: I don't know whether this is related or not, but shortly after
these errors occurred on my system, we had a head crash, and the disk was
ruined. This may be just a coincidence, but it may also be a scary flaw
in CDC drives, or something else. My hardware support guy and I have tried
many things, and still I gen an occasional crash, and an occasional newly
born badspot that must be found using salvol -db -talk, and then must be
manually typed in using invol. Another thing you can do to check a spot on
a disk, (if you know the daddress from salvol -db -talk) is to use RWVOL
to list the contents of that address. One field it should return is the
address again, the same as you type in:
>EX RWVOL
(I'm not exactly sure of the prompts)
Read or Write: R
DADDRESS: (Type in the address you wish to verify)
Then it returns some stuff sort of like this:
Head: 4fe2
Cylinder: blah blah blah
DADDR: (This should be the same as the address you typed in)
If the output DADDR has a lot of F's (hexadecimal) in it, there is probably
a badspot, which may be in the middle of a file. If so, you will have to
do something screwy to back up the file, and then enter the newly found
badspot using INVOL. Then you must salvage the disk with the new
badspot list.
I'm not a complete expert on this, but I'm sure most Customer services
hardware support people can help answer questions on this.
Scott Fergusonkrowitz@mit-richter.UUCP (David Krowitz) (02/18/88)
If you're getting sporadic DSP80 (not DSP80A) crashes, you might try replacing the power supply. We had a similar problem some time ago, with the machine randomly crashing, and we replaced every single board in the machine twice! We replaced both of the MSD500's (due to failures of the old style HDA's!!!) and still had the problem. Then our service rep measured the power supply output. It was something like 4.95 volts. The spec was 5.0 plus or minus 0.1 volts. He replaced the supply anyways with one that put out 5.1 volts. The problem went away and has never been seen again. We bought our service rep a case of his favorite liquid refresher as a token of our appreciation. -- David Krowitz krowitz@richter.mit.edu (18.83.0.109) krowitz%richter@eddie.mit.edu mit-erl!mit-richter!krowitz@eddie.mit.edu mit-erl!mit-richter!krowitz@mit-eddie.arpa krowitz@mit-mc.arpa (in order of decreasing preference)