root@nidhog.nidhog.cactus.org (Root) (09/25/89)
HALP!
Nidhog's disk is very sick I have a brand new(6 mo.) miniscribe 3085,
an even newer(1 mo.) WD 2010 controller, and a two week old 3b1 power suppply.
Nidhog had a 7300 1 meg on board 2 meg expansion dos board and am running 3.50 OS.
The disk works well most of the time but occasionally the hard disk
makes an awful noise ``brrrrrrrr...spriong-griong-griong''. The status
manager(smgr) flickers or dies sysinfo responst /dev/rfp002 can't
read superblock. The disk is gone. Well diagnostics ran and responded
error on check read when ever this happens. The only thing I can think
of is bad power. If any one out there has had a similar experience :*(.
Or has any helpful suggestions Please Let me know.
The only consistant response I got from the diags was that block # 3168
was bad. How I convert the block that diagnostics gives from surface test
to a logical block that I can give to the badblock table stuff?
Thanks in advance.
--
Henry D. Reynolds @'s: @nidhog.cactus.org
1718 Valeria !'s: {cs.utexas.edu,texbell}!helps!nidhog!hdr
Austin, Texas 78751 IBM: cs.utexas.edu!auschs!leperc.austin.ibm.com!hdr
HOME: (512)448-3617 WORK: (512)823-4348psfales@cbnewsc.ATT.COM (Peter Fales) (09/26/89)
In article <ROOT.89Sep25002448@nidhog.nidhog.cactus.org>, root@nidhog.nidhog.cactus.org (Root) writes: > > HALP! > > The disk works well most of the time but occasionally the hard disk > makes an awful noise ``brrrrrrrr...spriong-griong-griong''. The status > manager(smgr) flickers or dies sysinfo responst /dev/rfp002 can't > read superblock. The disk is gone. Well diagnostics ran and responded > error on check read when ever this happens. The only thing I can think > of is bad power. If any one out there has had a similar experience :*(. > Or has any helpful suggestions Please Let me know. Deja Vu! I had an almost identical experience. Every few days I would come home and find that smgr had died, though the system seemed to be working fine. I was also getting hard disk errors in the unix.log every few days, though I didn't know if the two were related. Finally, I happened to be in the room at the time and I heard the strange noises that you described. I was worried, let me tell you! My first thought was also "bad power." I took the cover off and checked the power supply voltages. Ah ha! Power was going up and down randomly between about 4.5 and 4.8 volts. Then I happened to look at the connector where the cable from the mother board connects to the power supply. Pins on this connector alternate with odds pins as +5 and even pins as ground (or vice versa, I don't remember). In any case, the plastic connector was discolored near each of the five volt connectors, as if those pins had somehow gotten very hot! When I pulled the connector off the pins on the power supply were black (like smoke). As a first try, I polished the connectors and put everything back together. Voila! 5 Volt power solid as a rock. I still don't know what happened to get me into this, but since that time (several months of continuous use) the system has never crashed, smgr has never mysteriously died, and I have had exactly one disk error in the log. -- Peter Fales AT&T, Room 5B-420 2000 N. Naperville Rd. UUCP: ...att!peter.fales Naperville, IL 60566 Domain: peter.fales@att.com work: (312) 979-8031
gil@limbic.UUCP (Gil Kloepfer Jr.) (09/28/89)
In article <3454@cbnewsc.ATT.COM> psfales@cbnewsc.ATT.COM (Peter Fales) writes: >and checked the power supply voltages. Ah ha! Power was going up and >down randomly between about 4.5 and 4.8 volts. Then I happened to look >at the connector where the cable from the mother board connects to the >power supply. Pins on this connector alternate with odds pins as +5 and >even pins as ground (or vice versa, I don't remember). In any case, the >plastic connector was discolored near each of the five volt connectors, as >if those pins had somehow gotten very hot! When I pulled the connector >off the pins on the power supply were black (like smoke). >UUCP: ...att!peter.fales Naperville, IL 60566 My last inspection of the power connector on my UNIX-pc seems to have the same problem, although my system has been working BETTER than normal since I got it over a year ago. I think that polishing the connector will probably improve the problem. One thing to those with 7300s that have upgraded their machines to 3B1s, or to those with perhaps braindamaged 3B1s, where the hard drive is powered from a spot on the MOTHERBOARD -- fix this. If you're handy with a soldering iron and have a multimeter handy, solder a new 4-pin molex connector with relatively thick wires onto the power supply directly. 3B1s usually have this done in the factory. The reason is that the motherboard power supply cable is responsible for powering all the expansion boards, the video display, and the floppy drive! This is a LOT of power, and that power supply line to the motherboard is cheap. My apologies if this procedure is impossible because it just can't be done -- I don't own a 7300 so I have only a small idea how to proceed and where to hook the wires on the power supply (maybe someone out there can supply this information, no pun intended). In any event, this will help preserve the life of the power supply connector. If worse comes to worse, this connector can (with great effort) be removed and replaced with one of those heavy-duty Molex-type connectors, but I wouldn't do it unless you're REALLY having a problem with it. ----- | Gil Kloepfer, Jr. | ICUS Software Systems/Bowne Management Systems (depending on where I am) | ...ames!limbic!gil or gil@icus.islp.ny.us
root@nidhog.cactus.org (Root) (09/30/89)
In article <569@limbic.UUCP> gil@limbic.UUCP (Gil Kloepfer Jr.) writes:
%My last inspection of the power connector on my UNIX-pc seems to have
%the same problem,
This is why I ordered a replacement 3b1 power supply form AT&T for my
7300. The problem is that when using the power strap that goes
directly from the power supply to the disk ... It must have a bad
connection because if i jiggle the wires to the disk power the disk
dies. I've had to reconnect the mother board power strap untill I can
get this ironed out the additional load of the disk on the power to
the mother board has recreated the wavy screen problem that my
other(7300) power supply had.
PS. Is it possible to get one of those connecters the ribbon strap to the
planer board and replace the existing one _without_ doing any
soldering?
PPS. does any one have a status report on the possible SCSI interface?
--
Henry D. Reynolds @'s: @nidhog.cactus.org
1718 Valeria !'s: {cs.utexas.edu,texbell}!helps!nidhog!hdr
Austin, Texas 78751 IBM: cs.utexas.edu!auschs!leperc.austin.ibm.com!hdr
(512)448-3617