Subject: Re: vnode usage and reclaimation - feels like deadlocking
To: Stephen M. Jones <firstname.lastname@example.org>
From: David Brownlee <email@example.com>
Date: 01/16/2004 20:37:30
On Fri, 16 Jan 2004, Stephen M. Jones wrote:
> What would be a good value for kern.maxvnodes? Is that even an issue?
> Can it go beyond 65536? (should I even be setting it?) Does it have a max
> value? I realise a vnode represents all sorts of objects (files, sockets,
> symlinks .. fifos) so it will vary if a machine is totally pigged out and
> in that case, should you have your kern.maxvnodes set to take care of
> all your potential resources? Is reclaimation done any differently on
> an NFS client or vnodes that map to NFS objects?
On all but very low memory machines I tend to set NVNODE to
131072 or 262144. It does consume quite a bit of memory, but
it does help dealing with large directory structures (pkgsrc,
large websites, rsyncing).
The way thinsg are handled has be reworked significantly in -current
leading to (as I understand) less effective memory cost per vnode.
As a data point I'd definitely try it very high on one of
the problem machines. You also might want to try dropping in
a -current kernel and keeping everything else the same on one
David/absolute -- www.netbsd.org: No hype required --