Messages in this thread | | | Subject | Re: WIP Patch: Cleanup of knfsd export handling. | Date | Sun, 05 Sep 1999 12:32:57 +0100 | From | David Woodhouse <> |
| |
(with apologies for quoting in public) cks@utcc.utoronto.ca said: > > I suggest that the simple way to handle this is to consider the > kernel's list to be purely a cache, and to have exportfs (or mounted) > flush it entirely whenever the export list is changed. This then > automatically handles removing machines from the exports, and lets you > treat negative entries just like positive ones for space/expiry/etc.
That's my intention, eventually, but currently that would cause some requests to get dropped - when the kernel finds no entry, but has to call up to mountd, it just drops the request until mountd gives it a definite export entry, be that negative or positive.
I want to find some way of keeping those requests, so we don't have to wait for the client to resubmit them.
I was thinking of just feeding the whole thing up to mountd, and then trying to find some way for mountd to stick it back into the request queue if it's authorised.
I don't know if that's going to be possible without a fairly evil hack to the sunrpc layer. Perhaps it would be better to have an alternative return response from the *_dispatch routines which tells sunrpc to 'try me again in <N> jiffies' instead?
---- ---- ---- David Woodhouse David.Woodhouse@mvhi.com Office: (+44) 1223 810302 Project Leader, Process Information Systems Mobile: (+44) 976 658355 Axiom (Cambridge) Ltd., Swaffham Bulbeck, Cambridge, CB5 0NA, UK. finger dwmw2@ferret.lmh.ox.ac.uk for PGP key.
- 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/
| |