Re: mdconfig unable to allocate memory

On Sat, Feb 28, 2004 at 03:53:16AM -0500, Chris BeHanna wrote:
> On Friday 27 February 2004 09:17, Chris Elsworth wrote:
> > On Fri, Feb 27, 2004 at 02:08:48PM +0000, Colin Percival wrote:
> > > At 13:58 27/02/2004, Chris Elsworth wrote:
> 
>     There may be *some* tuning to do, but honestly, this is what the
> buffer cache was designed to do, *automatically*.  Data that have been
> fetched from disk is kept on one of several LRU lists, which are in

I think I'm misunderstanding which cache we're talking about exactly
here. By buffer cache, do you mean the figure under "Buf" in top,
which is also vfs.maxbufspace? This is currently 200MB on my 4GB
machine, which feels excrutiatingly small. Is stuff cached elsewhere,
too?

>     I have 1GB of RAM in my workstation.  I had a wild hair when I
> installed it, and made a 700MB swap-backed scratch area, and did a
> buildworld to see how fast it would be.  The system survived just
> fine.  I would think that you should be able to make a 2GB swap-backed
> file system with 4GB of RAM without a problem, but there may be some
> autoscaled tunables that don't scale that high correctly.

It seems to work, although I managed to produce a hard freeze (of the
variety that not even hitting soft-power worked, I had to hold it for
4 secs to get the machine to switch off). I have 4GB, and 12GB of swap
(4GB on each disk, of 3 disks). I piled 10GB of zeros into it
(dd if=/dev/zero of=/md_disk/zeroes) and it performed exactly as
expected; once all 4GB of memory was filled it started moving into
swap. After it had got through about 10GB (so 6GB of swap used, ish) I
stopped it and rm'ed it. Someway through this rm, the thing froze. If
I can recreate it I'll try some kernel debugging, but I've never
really done much with DDB before.


Cheers for your comments so far, everyone, they've been most helpful.

-- 
Chris