[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Disk I/O stalling whole system ..



On Mon, Sep 08, 2003 at 10:42:28PM +0200, Mariano Kamp wrote:

| when copying from one ide disk to another ide disk on the the same
| controller the whole systems stalls. I am running 2.4.19 on
| unstable.

I've noticed similar effects here.  Kernel 2.4.20, but only 1 disk.
If something pushes the disk hard then most thigns, including xmms
(which is streaming, not using the disk), becomes unresponsive until
the heavy disk usage subsides.  (I had figured old hardware had
something to do with it, but mabye 2.6 will improve this, as someone
else suggested)

| When renicing the copying processes it works ok, but I am not sure
| if that is the solution. And if that is, how would I do this
| efficiently. Looking a top and renicing by hand is a bit annoying to
| me ;-)

Run the command with 'nice' in the first place.  Eg :

    $ nice -n 19 cp foo bar

-D

-- 
For society, it's probably a good thing that engineers value function
over appearance.  For example, you wouldn't want engineers to build
nuclear power plants that only _look_ like they would keep all the
radiation inside.
    (Scott Adams - The Dilbert principle)
 
http://dman13.dyndns.org/~dman/

Attachment: pgp9QD2RMam9m.pgp
Description: PGP signature


Reply to: