Re: fixing coda in -current

[Brett Lymn <blymn%internode.on.net@localhost>]

On Tue, Nov 20, 2018 at 10:18:56AM -0500, Greg Troxel wrote:
> I used to use it, and may again.  So I'd like to see it stay, partly
> because I think it's good to keep NetBS relevant in the fileystem
> research world.  I am expecting to see new upstream activity.
> 

I guess it is obvious that I use it too.  There is still some activity
going on at CMU around coda so it has not been abandoned by the
upstream.  It is difficult to automate the testing because you need the
userland support there too - the kernel driver does a bunch of upcalls
to a userland daemon to perform most of the work.

I use coda to keep my work in progress stuff on so I can work on my
laptop or my desktop, have the changes shared but then be able to take
my laptop off network and keep working, when I get back home the changes
can be synced without me having to do anything.  Sure, there are other
ways of doing this but trying to maintain a shared tree potentially
changing either side is difficult when you don't have constant network.

> But, I think it makes sense to remove it from GENERIC, and perhaps have
> whatever don't-autoload treatment, so that people only get it if they
> explicitly ask for it.  That way it should not bother others.
> 

Yes, this would be fine.  Mind you, you have to explicitly install the
userland from pkgsrc so the driver is dormant until that happens.

-- 
Brett Lymn
Let go, or be dragged - Zen proverb.