Re: nfsiod issues?

Bryn Paul Arnold Jones (bpaj@gytha.demon.co.uk)
Wed, 17 Apr 1996 02:06:26 +0100 (BST)


On Tue, 16 Apr 1996, Olaf Kirch wrote:
> On Mon, 15 Apr 1996 23:18:31 CDT, "Larry 'Daffy' Daffner" wrote:
[......]
> > It seems strongly against the concept of loadable modules to me,
> > especially with kerneld.
>
> In this case, probably kerneld should be fixed to run `killall -TERM nfsiod'
> before unloading the module.
>
> Olaf

Hmm, shouldent the nfs module have no kernel threads running when it's
not in use (compiled in/insmoded, but no mounts (yet/anymore) ).

Ie only have nfsiod(s) about when nfs is activly (sp?) being used, or
activly trying to mount one (activly means we have a nfs tree imported
now).

You (we) could have one nfsiod per imported tree, instead of 4 all at once
which would make more sense from a user point of view, but I don't know
from an implementation point of view.

I would have thought that keeping n processes up to date with each other
would be a real pain, but thing's must be differant in a kernel thread,
but letting a one process watch the activity on a nfs mounted tree.

Bryn

--
PGP key pass phrase forgotten,   \ Overload -- core meltdown sequence 
again :(                          |            initiated.
                                 / This space is intentionally left   
                                |  blank, apart from this text ;-)
                                 \____________________________________