Re: linux-kernel-digest V1 #4408

Robert Dinse (nanook@eskimo.com)
Sat, 4 Sep 1999 19:35:24 -0700 (PDT)

On Sat, 4 Sep 1999 owner-linux-kernel-digest@vger.rutgers.edu wrote:
>
> From: Alan Cox <alan@lxorguk.ukuu.org.uk>
> Date: Sat, 4 Sep 1999 20:41:54 +0100 (BST)
> Subject: Re: 2.2.12 Memory Leak?
>
> > As I can see it so far it behaves just like the clean 2.2.12,
> > buffer cache is filled, then free mem gets filled and the
> > swap memory increases also more and more. A few runs of innd
> > crunching the newsspool/-databases will show it.
> >
> > I'll try it one or two days in addition to see if it's really
> > this thing we talk about, but I'm very afraid it is.
>
> Ok so the three people still seeing an apparent leak are running innd. Thats
> probably not co-incidental.
>
> Alan

I've got two machines that are essentially identical save for the fact
that one has slightly more RAM and the other a 2nd disk controller, one running
innd, the other running Apache.

The Apache web server actually runs itself out of memory faster under
2.1.12, despite the fact that it has more physical RAM, so this problem is not
limited to INND.

I am back on 2.2.10 now as 2.2.11 and 2.2.12 have proven completely
unusuable here, while 2.2.10 with half a dozen patches people provided is
approaching usability.

Incidentally, David Miller, the patch to Sparc assembly you provided does
not seemed to have materially helped. I applied it to 2.2.10 as well, hoping
it might help there (since the memory leak in 2.2.12 made it unusable), but it
still will do that same spinlock thing.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/