Coda File System

Re: Did I do right, Dad?

From: Peter J. Braam <braam_at_cs.cmu.edu>
Date: Mon, 25 Aug 1997 18:28:55 -0400 (EDT)
Thanks for your interest!

On Mon, 25 Aug 1997, Toshio Kuratomi wrote:

> So I tried using the coda client rpm against the testserver and let me
> just say, wow!  This is dreamy!
> I do have a question though: I was playing with cfs and doing hte following
> did something wrong:
> 	$ cfs disconnect
> 	$ cfs reconnect
> because now when I attempt to do things to files/directories that aren't
> cached, I get a Connection timed out error:
> 	$ cd /coda/playground
> 	$ ls junk
> 	ls: junk: Connection timed out

You get this message because you are disconnected, and the file is not in
the cache.  If you hoard it, you can use it while disconnected.


> on reading the file INSTALL, I saw this:

ONLY after a crash.  I haven't had time to write a sigchild handler (not a
lot of work really, but then). So what you see is a healthy Coda
situation. Your system is working. Print out the manual and read about
hoarding then try again.

Good luck.

Peter Braam



>         d) Restarting a client. You should umount /coda, since this 
>            may not get done automatically. If it fails, make sure you
>            kill all processes using /coda; cd out of Coda directories etc. 
> 
>            If ps aux | grep venus displays a Zombie venus, reboot 
>            your machine before trying again.  (If you don't you'll
>            likely have to run fsck by hand later.)
> So I checked for zombie processes and sure enough I have:
>      root     11856  0.5 17.0 12144 10828  p4 S   13:27   0:25 venus -init 
>      root     11865  0.0  0.0     0     0  p4 Z   13:28   0:00 (venus <zombie>)
> So.... What does all this mean?  I take it, I should reboot before
> experimenting with coda further, yes?  But is this supposed to happen?
> Should I be doing something else to disconnect from the coda server
> temporarily?  To reconnect?  Should a normal user be able to disconnect from
> coda?  Or should only root be able to do a cfs disconnect?
> 
> Anxiously awaiting answers (so I can ask more questions!)
> -- 
> -Toshio Kuratomi            // "The Past is always with us."
>  badger_at_prtr-13.ucsc.edu   //=_=~=_=~=_=~=_=~=_=~=_=~=_=~=_=~=_=~=_=~
> 

Peter Braam
Received on 1997-08-25 18:43:37