[OpenAFS-devel] Re: [OpenAFS] freezes acessing /afs/.git

[Andrew Deason]

On Mon, 11 Aug 2014 07:07:17 -0400
chas williams - CONTRACTOR <chas@cmf.nrl.navy.mil> wrote:

> Before I go ahead and actually write something again how about a
> consensus on where this negative cache should be?  I don't think the
> dnlc currently supports caching negative information (the comments say
> that anyway) but that shouldn't be too hard to fix.  Caching negative
> entries in afsd would be easier of course.

I expected this to be cached as a 'struct cell' in libafs. We already
cache that information in the global list of cells, and fallback to an
afsdb query if we can't find one. So instead, to have a negative cache
entry add a flag or some other indicator to the structure that says
"this cell doesn't exist". Either a magic cell number or a 'states' flag
or something.

Doing that with the existing lookup logic might not be great, though...
it would be better if the cell lookup routines used a hash table or
something. That would be good to do anyway, but caching negative entries
I think makes it more required.

-- 
Andrew Deason
adeason@sinenomine.net