Re: [Smartmontools-support]ATA errors

[Jan S. <jan...@we...>]

Sorry - I wanted to forward the mail instead of replying to it, but I
think I hit the wrong button.
OK, here we go!

-----------------------------------------------------------------------

Subject:
Re: [Smartmontools-support]ATA errors
From: Bruce Allen <ba...@gr...>
Date: Mon, 21 Apr 2003 17:17:07 -0500 (CDT)
To: Jan Statzner <jan...@we...>

Hi Jan,

This is very interesting -- how about forwarding your mail to the
smartmontools mailing list and to the user who was having problems.

I find the temperature issue very curious -- perhaps Maxtor is using a new
normalization (factor of two??) for their drive temperatures.

Cheers,
	Bruce

On Mon, 21 Apr 2003, Jan Statzner wrote:


 >> Hello again!
 >>
 >> I hope I can give you some useful information regarding this report:
 >>
 >
 >>>> >>194 Temperature_Celsius     0x0032   253   253   000    Old_age
 >>>> >>-       44
 >>
 >>> > This is kind of warm -- you might want to arrange better cooling.
 >
 >> I'm running a Maxtor 4R120L0 for about three weeks now. I tested this
 >> drive before I mounted it into its intended target system, and after a
 >> few minutes of operation the drive told me that it reached a 
temperature
 >> of 44°C, but the surface was just hand warm - and both the temparture
 >> sensor's output and the temparture of the drive's surface didn't change
 >> even after several hours of operation. I ran the test with the drive
 >> lying on top of a computer case - there was no fan or other kind of
 >> ventilation around.
 >> Could this be an error in the temparture mesaurement system of the 
newer
 >> Maxtor drives?
 >>
 >
 >>>> >>SMART Error Log Version: 1
 >>>> >>Warning: ATA error count 91 inconsistent with error log pointer 5
 >>
 >>> > I haven't seen this inconsistency before in Maxtor disks -- 
interesting...
 >
 >> I don't know if it has to do something with this inconsistency, but
 >> another Maxtor drive - a 4D080H4 that is in operation for a bit more
 >> more than one a year now - tells me the following:
 >>     SMART Error Log Version: 1
 >>     Warning: ATA error count 81 inconsistent with error log pointer 5
 >>
 >
 >>> > Command register CR contains 0xca.  This is the "write DMA" 
command.  It
 >>> > may be that there is some type of DMA problem -- though it 
appears that
 >>> > similar commands didn't cause errors before.
 >>> >
 >>> > So I'm not sure what is causing these occaisonal errors.
 >
 >> I experienced DMA (multiword DMA - the system the problems occurred in
 >> doesn't know about UDMA) problems with Maxtor drives before: The 
4D080H4
 >> heavily crashes the entire IDE system when run in DMA mode (you have to
 >> reboot the machine to get IDE working again), and running the 
4R120L0 in
 >> DMA mode results in continuing IDE resets - both drives work reliably
 >> when run in PIO mode.
 >> I made these observations on an i430FX-based system running SuSE Linux
 >> 7.2 with kernel 2.4.7; an upgrade to 2.4.18 (kernel supplied by SuSE)
 >> didn't change this. A friend of mine experienced similar problems on an
 >> i440BX-based board with a new 120 Gig Maxtor drive - it makes 
trouble in
 >> any DMA mode under Debian Woody, but in PIO mode everything works fine.
 >> Another Maxtor drive (a 4K060H3) runs relaibly in DMA mode in my 
system,
 >> just as the IBM drive (DJNA-351520) running in the same machine until I
 >> installed the 4R120L0 did.
 >>
 >> Jan
 >>
 >>