Hi,

After Jan Harkes's and Ivan Popov's help, I think I'm getting a grip
on coda. Thanks to both.

My current outstanding confusion has to do with hoarding.  I've
created a hoardfile (it's pretty long, here's some of it):

-
clear
add /coda/ramsey.localdomain/home/kreuter 100:d+
add /coda/ramsey.localdomain/usr/local/stow/stow 600:d
add /coda/ramsey.localdomain/usr/local/stow/aclocal 600:d
add /coda/ramsey.localdomain/usr/local/stow/bbdb 100:d
...(more lines like the above)
-

When I run 'hoard -f ~/hoard.hdb', some of the files in these
directories are cached, some aren't.  The file /tmp/_Venus_List_Cache
shows a listing of files in the directories I intend to hoard,
apparently with asterisks at the beginning of lines containing the
names of files that aren't cached.  Naturally, when I disconnect, the
uncached files are unavailable.

If I remain connected, it seems that the uncached files get cached
under some circumstances, but I'm not sure what these circumstances
are: I tried the following to force venus to cache everything below a
certain directory, but got many 'Connection timed out' errors:

$ find . \! -type d -exec cat \{} > /dev/null \;

However, when I try reading individual files for which the connection
timed out, I find that I am able to read them, and thereafter these
files seem to be cached.  It also looks as though individually running
'cfs fl' for each file that's uncached then trying to read the file
will bring the file into the cache.

I've increased the size of my venus cache to 250MB, and I don't seem
to be using up all the cache:

$ df -i /coda
Filesystem            Inodes   IUsed   IFree IUse% Mounted on
coda                   10416    4446    5970   43% /coda

I've also flushed the venus cache and started over a few times.  Each
time, the set of uncached files changes, but it's always pretty
sizable (so far more than half the files in the volumes).  

So, is it possible to coerce hoard to actually cache the whole lot of
files in specified in the hoardfile?

--
Thank you,
Richard