[Nottingham] Various kernel oops

[Martin]

Graeme Fowler wrote:
> On Wed, 2008-05-28 at 09:31 +0100, Graeme Fowler wrote:
>> As an aside, are you using a processor with hyperthreading? If you are,
>> turn it off and see if that makes a difference. Also don't overclock.

It's a dual core AMD: AMD Athlon64 X2 6000

> ...and, if you have it available, try the -hugemem kernel. It's designed
> (well, in 32-bit land it is!) for systems with over 16GB memory but at
> work we've found it useful for systems with 8GB or more (using RHEL4 and
> 5). The headline is that for a small CPU overhead, apps have access to a
> greater per-process memory space; underlying this seems to be much
> improved FS cache handling.

Interesting idea.


Thanks for the comments.

I've minimized activity on that box now. However, what I have noticed is
that:

Mem:   8190988k total,  7589768k used,   601220k free,   573092k buffers

The used inexorably climbs ever higher and rarely drops back.

Checking the processes from top and X has the largest memory footprint
at 800M or so. Everything else is just a few MBytes each. Only 176 tasks
in total. It certainly doesn't add up to 7.5 GigaBytes!

A memory leak somewhere?

And so how to investigate further?


How does top actually report "used"?

Cheers,
Martin

-- 
----------------
Martin Lomas
martin at ml1.co.uk
----------------