FC3 and Drive Errors

[mylar]

On Mon, 2006-09-18 at 14:22, Rick Stevens wrote:
> On Mon, 2006-09-18 at 14:04 -0400, mylar wrote:
> > Hi,
> > 
> > I am running Fedora Core 3 on a Dell Dimension 4700 desktop. It's a dual
> > boot machine. Windows resides on a 80 Gb SATA drive and Linux on a
> > regular 200 Gb ATA IDE drive attached to the PATA controller. When
> > running FC3 the system has been working fine for well over a year but of
> > late has been freezing up every now and  then. Everytime it freezes and
> > I go over the system logs I am seeing the following drive errors
> > occuring just prior to the system crashing...
> > 
> > 
> > > Sep 18 04:02:20 brooklyn kernel: hdb: dma_intr: status=0x51 { DriveReady
> > > SeekComplete Error }
> > > Sep 18 04:02:20 brooklyn kernel: hdb: dma_intr: error=0x84
> > > { DriveStatusError BadCRC }
> > > Sep 18 04:02:20 brooklyn kernel: ide: failed opcode was: unknown
> > > Sep 18 04:02:20 brooklyn kernel: hdb: dma_intr: status=0x51 { DriveReady
> > > SeekComplete Error }
> > > Sep 18 04:02:20 brooklyn kernel: hdb: dma_intr: error=0x84
> > > { DriveStatusError BadCRC }
> > > Sep 18 04:02:20 brooklyn kernel: ide: failed opcode was: unknown
> > > Sep 18 04:02:20 brooklyn kernel: hdb: dma_intr: status=0x51 { DriveReady
> > > SeekComplete Error }
> > > Sep 18 04:02:20 brooklyn kernel: hdb: dma_intr: error=0x84
> > > { DriveStatusError BadCRC }
> > > Sep 18 04:02:20 brooklyn kernel: ide: failed opcode was: unknown
> > 
> > Any ideas what might be going on ? Is the drive going bad on me ? 
> > I am going to doi a check for bad blocks and see what happens. Any  other ideas,
> > hints, suggestions would be appreciated.
> 
> Those kinds of errors are usually indicative of a dying drive.  You're
> receiving CRC errors on the DMA transfers and seek incomplete errors.
> 
> This can also be caused by a dying power supply--particularly the +12VDC
> side of things, since that's what's used to turn the spindle and move
> the heads.  If you have the technical competence and a multimeter, check
> the voltages AT THE DRIVE using the meter.  You want the +12VDC level at
> about 12.7 to 13.0 volts.  IIRC, the 12V should be between the black
> (ground) and yellow wires on the drive connector.  If they're low, you
> should think about a new power supply.
> 
> If the voltages seem OK, then plan on a new drive in the near future.
> While you're in there, also make sure all the fans are running well
> (they do wear out).  You could also blow dust out using a can of Dust
> Off.  Use care so you don't force dirt down into any of the connectors
> such as the CPU, memory, PCI bus, etc.  It's amazing how much crud
> accumulates in a case.
> 
Hi Rick,

Yep, I was thinking along the similar lines, it could be the drive is on
it's way. Thanks for the tip regarding the power supply. Yeah, I can
handle the process of checking the voltages. I'll connect up my Fluke
multitester and see what kind of voltages I'm getting. Actually I was
kind of wondering about a possible power supply  problem as I have seen
similar behavior on other machines shortly before a power supply  crap
out. Unfortunately the machine ain't no longer under the Dell warranty
so I'll have put out the cash if it needs a new drive. If it's a power
supply I have plenty  of those  laying around all over the place. :)

Good advice on cleaning the machine out. It's well overdue for a
dusting. These machines tend to collect more dust than a vacuum cleaner
and I haven't been getting around to cleaning them as often as I should.

Much Thanks.

John a/k/a mylar






-- 
Email-> micros50 at computer.net
Website-> http://home.computer.net/~micros50/n2ize_index.html

I speak Mathematics