On Thu, Mar 30, 2000 at 11:51:31AM -0500, Andrew Gillham wrote:
> Thor Lancelot Simon writes:
> > Have we both arrived at the hypothesis that, sometime between the 
> > introduction of the pciide code in mid-1.3, and 1.4, the filesystem code
> > was broken so as to cause this problem on some systems (long hangs under 
> > heavy I/O load)?
> 
> FWIW, I have seen this type of behavior on a server at home.  It is a
> PII-400 with 128MB, and a single 20GB UDMA/33 EIDE.  I will try to test it
> some more when I get home.  What I have seen pretty much what you guys
> have been describing, under certain loads (cleandir/sup/cvs??) I can barely
> do a 'ls -l' or similar in a reasonable time.

Does the behaviour you're describing appear as a momentary freeze of all
disk activity? In 1.4 on i386 I noticed some odd pauses (about 1 1/2 seconds)
that I attributed to thermal recalibration. The machine it stood out on the
most had a 2940UW with UW Scsi drives, but I think I've seen it on IDE also.

I noticed this primarily during installs, using wscons, and doing installs
from pkgsrc on one screen, and commands on another screen which generated
multiple screens of output would sometimes pause (1 1/2s). Not often, maybe
once every 3-5 minutes.

I assumed hardware at the time - could it be related?

-- 
David Maxwell, david@vex.net|david@maxwell.net -->
All this stuff in twice the space would only look half as bad!
					      - me