Hi Lamar,

On Mon, 30 Jun 2014 13:18:12 -0400
Lamar Owen <[email protected]> wrote:

> On 06/30/2014 11:48 AM, Andras Horvath wrote:
> > Hi Everyone,
> >
> > I've been having an issue for a month or so now that I seem unable to track 
> > down.
> >
> > I have an external USB drive for backup. When I start copying from the 
> > computer to the external disk, it starts to fail sooner or later. Whether 
> > at a large file or after several small ones, but it fails dropping messages 
> > like this (taken from dmesg):
> >
> > ...
> >
> >
> > I tested the disk on 2 different computers (one of them is brand new 
> > server, the other is old one serving for years) in 3 different USB covers 
> > (they're new ones too) with 2 different brand new disks. The disks are 2 TB 
> > in size (a bit less actually, around 1.8 TB formatted).
> 
> When the errors occur, can you tell if the drive spins down and then 
> back up?
> 
> If so, I have seen similar errors with a pair of bus-powered 2.5 inch 
> 1TB Seagate USB3 drives; it seems like the drives actually draw more 
> power than the port can deliver for an extended amount of time.  I have 
> to put them on a USB 2 port instead of a USB 3 port to get them to be 
> reliable for writing (reading seems to never be the problem).  You 
> didn't say if the drives were USB-powered or not, but the port spinning 
> the drives down would be the first thing I would check.
> 
> As to Debian 6 being able to work with it and SL6 not, that could be a 
> kernel difference where the Debian kernel is waiting and retrying longer 
> in the case where the drive is spinning down.
> 
> If the drive is not bus-powered, it could still be spinning down due to 
> its green features, and maybe the Debian 6 USB stack is disabling those 
> features (or at least not complaining about those features) whereas the 
> SL6 kernel is not as forgiving of drives spinning down or isn't 
> disabling those features.


Actually the drive has its own power so it is not USB powered. I cannot tell if 
the drive spins down (did not get the idea to check it), but the CPU is in 100% 
I/O wait all the time after this happens.

I was told the disk is a WD RED, but I'll check the power mode later with 
hdparm.

Thank You for the suggestions.

Andras

Reply via email to