All of a sudden, two separate hard drives in the same box have 
started to cause the kernel to print out messages like the following:

	hdb: dma_timer_expiry: dma status == 0x60
	hdb: DMA timeout retry
	hdb: timeout waiting for DMA
	hda: dma_timer_expiry: dma status == 0x21

Are these drives on their last legs? 

James H. Markowitz <noone@nowhere.net> wrote:
>        All of a sudden, two separate hard drives in the same box have 
> started to cause the kernel to print out messages like the following:
> 
>        hdb: dma_timer_expiry: dma status == 0x60
>        hdb: DMA timeout retry
>        hdb: timeout waiting for DMA
>        hda: dma_timer_expiry: dma status == 0x21
> 
> Are these drives on their last legs? 
> 
> 
> 

Seems very unlikely that you have two disks that are dying
at the same time. much more likely to be an issue
in your software or in your controller hardware.

Stan

On 02/08/2010 01:03 PM, James H. Markowitz wrote:
> 	All of a sudden, two separate hard drives in the same box have
> started to cause the kernel to print out messages like the following:
>
> 	hdb: dma_timer_expiry: dma status == 0x60
> 	hdb: DMA timeout retry
> 	hdb: timeout waiting for DMA
> 	hda: dma_timer_expiry: dma status == 0x21
>
> Are these drives on their last legs?

Does not sound good.  I try to replace my drives
every three years or so.  Usually winds up being four.
How old are your drives?  Hopefully, they are not
Western Digital -- my customers have shed a lot
of tears (and cuss words) over them.  I refuse to
sell them.

I love Seagate's "Enterprise" level hard drives.
They never go bad.  Cost about U$D 30.00 more.

http://www.seagate.com/www/en-us/products/servers/barracuda_es/

If you have a SAS controller anywhere nearby, SAS
drives really work well on scattered multiple reads
and writes.  They make a big difference.

-T

On Mon, 08 Feb 2010 21:03:58 +0000, James H. Markowitz wrote:

> Are these drives on their last legs?

Install smart monitoring tools and use smartctl to check the
state of the drives to find out how bad things are with

smartctl --all /dev/hda

smartctl --all /dev/hdb

This will tell you everything there is to know about your disks,
assuming that it has SMART capability.

On Mon, 08 Feb 2010 16:03:58 -0500, James H. Markowitz <noone@nowhere.net> wrote:

> 	All of a sudden, two separate hard drives in the same box have
> started to cause the kernel to print out messages like the following:
> 	hdb: dma_timer_expiry: dma status == 0x60
> 	hdb: DMA timeout retry
> 	hdb: timeout waiting for DMA
> 	hda: dma_timer_expiry: dma status == 0x21
> Are these drives on their last legs?

Try replacing the ide cable with an 80 wire one. See
http://www.pcguide.com/ref/hdd/if/ide/confCable80-c.html

Regards, Dave Hodgins

-- 
Change nomail.afraid.org to ody.ca to reply by email.
(nomail.afraid.org has been set up specifically for
use in usenet. Feel free to use it yourself.)

On 2010-02-08, Todd <todd@invalid.com> wrote:
>
> Does not sound good.  I try to replace my drives
> every three years or so.  Usually winds up being four.
> How old are your drives?  Hopefully, they are not
> Western Digital -- my customers have shed a lot
> of tears (and cuss words) over them.  I refuse to
> sell them.

YMMV, obviously.  I've used WD for years, with minimal problems.  Of
course not *zero* problems, because drives die.

> I love Seagate's "Enterprise" level hard drives.
> They never go bad.  Cost about U$D 30.00 more.

I've had Seagate drives go bad at about the same rate as WD.  Seagate
used to have a basically unusable RMA system (this was many years ago),
which is why I leaned toward WD.  Now that Seagate has fixed their RMA
system they're more or less equal when I'm ordering drives.

--keith

-- 
kkeller-usenet@wombat.san-francisco.ca.us
(try just my userid to email me)
AOLSFAQ=http://www.therockgarden.ca/aolsfaq.txt
see X- headers for PGP signature information

On 02/08/2010 02:08 PM, Keith Keller wrote:
> On 2010-02-08, Todd<todd@invalid.com>  wrote:
>>
>> Does not sound good.  I try to replace my drives
>> every three years or so.  Usually winds up being four.
>> How old are your drives?  Hopefully, they are not
>> Western Digital -- my customers have shed a lot
>> of tears (and cuss words) over them.  I refuse to
>> sell them.
>
> YMMV, obviously.  I've used WD for years, with minimal problems.  Of
> course not *zero* problems, because drives die.

The main problems I have seen are the ones customers purchase
from Best Buy: low bid, cheapie.

>
>> I love Seagate's "Enterprise" level hard drives.
>> They never go bad.  Cost about U$D 30.00 more.
>
> I've had Seagate drives go bad at about the same rate as WD.  Seagate
> used to have a basically unusable RMA system (this was many years ago),
> which is why I leaned toward WD.  Now that Seagate has fixed their RMA
> system they're more or less equal when I'm ordering drives.
>
> --keith

Keith,

You missed part of what I said.  I too will not sell
the regular Seagate drives.  They are not much better
than the WD ones.  I sell the "Enterprise" level drives.
They are specifically designed to run 24/7 in data centers.
In other words, I recommend the drives build for servers
not workstations.

Here is a link to them:
http://www.seagate.com/www/en-us/products/servers/barracuda_es/
I have them spread over two counties: zero defects

And in prior years, I would not sell Seagates at all.  They
use to be such trash.  And, WD use to be so good too, but
they pissed that away.  The main problem I see is the
cheapie (low bid) drives purchased at Best Buy.

Hope I cleared that up.
-T

p.s. WD now sells an "enterprise" drive too.  I do
believe it is called the "Raptor", but I am not sure.
I am still too chicken to use WD with all the troubles
I see coming in from customers.  (They are cheap
for a reason!)

On 2010-02-08, Todd <todd@invalid.com> wrote:
>
> You missed part of what I said.  I too will not sell
> the regular Seagate drives.  They are not much better
> than the WD ones.  I sell the "Enterprise" level drives.
> They are specifically designed to run 24/7 in data centers.
> In other words, I recommend the drives build for servers
> not workstations.

Okay, but WD sells ''enterprise'' drives too.  I've been using this
model with few problems:

http://www.wdc.com/en/products/products.asp?driveid=610

They make SAS enterprise drives as well.  But I've also used their
''desktop'' drives, also without major problems.  But then again, I've
always used these drives in a redundant RAID, so if a drive happens to
fail I don't lose data.  But it would still be a PITA to RMA drives
regularly, and I would not use drives that I had bad experience with.

> Here is a link to them:
> http://www.seagate.com/www/en-us/products/servers/barracuda_es/
> I have them spread over two counties: zero defects

I have some of these as well, but not with zero defects.  Zero
defects is an unreasonable expectation, especially if you have a large
data center (I do not, but I do have a fair number of drives).

--keith

-- 
kkeller-usenet@wombat.san-francisco.ca.us
(try just my userid to email me)
AOLSFAQ=http://www.therockgarden.ca/aolsfaq.txt
see X- headers for PGP signature information

On 2010-02-08, James H. Markowitz <noone@nowhere.net> wrote:
> 	All of a sudden, two separate hard drives in the same box have 
> started to cause the kernel to print out messages like the following:
>
> 	hdb: dma_timer_expiry: dma status == 0x60
> 	hdb: DMA timeout retry
> 	hdb: timeout waiting for DMA
> 	hda: dma_timer_expiry: dma status == 0x21
>
> Are these drives on their last legs? 

How often does that happen?

A few years ago, I got one of those every month or three.  IIRC,
after a later distro release, they quit happening with _NO_
change in disks or disk-related hardware.  Actually, there might
have been a power supply replacement in there, too.  If they're
more than a week apart, my bet would be on a kernel bug.  If
they're more often than every day, my bet would be on hardware.

-- 
Robert Riches
spamtrap42@verizon.net
(Yes, that is one of my email addresses.)