--- SODA Noriyuki <soda@sra.co.jp> wrote:
> >>>>> On Fri, 11 Mar 2005 23:49:33 +0900 (JST),
> 	SODA Noriyuki <soda@sra.co.jp> said:
> 
> > And you say yours is more limited than
> Dragonfly's.
> > Well, That's not better, rather it's worse.
> 
> BTW, one of problems of your distribution
> (mypatch.tar.gz) is that it
> doesn't contain a patch from DragonFly's original
> source, rather it
> contain already patched source. This makes it
> difficult to see which
> part is your contribution.
> 
> So, I made a diff from DragonFly's work, and put it
> in the same
> directory:
> 
>
ftp.netbsd.org:/pub/NetBSD/misc/martin/prasad/prasad-ckpt.tar.gz
> 
> 
> IMHO, this work focuses on very limited application,
> and it's not
> generic enough to incorporate to NetBSD.
> --
> soda
> 
Its application is program recovery -but the types of
userland applications (if that is what you meant)
which can use this feature is not limited. 
 If you want to use checkpointing for process
migration, that would require a substantial amt. of
work on checkpointing the kernel side of the process
aka the file descriptors, sockets, pipes etc.. It
would require prior info on whether the filesystem is
shared or not. I mean, I would pass from machine X to
Y info about the process's fds so that when the
checkpointed file is moved to machine Y, the kernel
has setup everything for him as if he is actually
running on X.

regards
-kamal


------------------------------------------------------------
Kamal R. Prasad
UNIX systems consultant 
kamalp@acm.org

In theory, there is no difference between theory and practice. In practice, there is:-).
------------------------------------------------------------


		
__________________________________ 
Do you Yahoo!? 
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/