Re: Size matters was Re: notification charter proposal

[Andy Bierman <ietf@andybierman.com>]

Phil Shafer wrote:

> Andy Bierman writes:
>  
>
> > Balazs Lengyel wrote:
> >    
> >
> > > Please dont set a max message size generally for netconf !!!
> > >      
> > We are discussing the notion of setting a minimum value
> > that conforming implementations can use.
> >    
>
> Setting a minimum maximum effectively sets the maximum an implementation
> can count on the other side implementing, which quickly turns into
> a hard restriction.
>  

This is possible, but IMO NMS developers from a company
will pressure the agent developers in that company to provide them
with a reasonable max-size.

You set the bar at 64K, and it's not such a bad restriction.
I don't think we should try to set a limit, especially if it is
a really low number like 1500.  I stated in the previous mail
that there isn't consensus to set a limit, so I think this topic can be 
closed now.


> Instead we should have any message size restrictions be passed in
> to the server at subscription time.  If the client gives a number
> larger than the server can send, it's ignored.  If the number is
> smaller, then the server must truncate.  If the number is not
> provided, the server is freed from worrying about the client's
> message size considerations.
>  

Well put.

>  
>
> > If anybody out there is planning an implementation with
> > a message size bigger than 4 billion bytes, please let me know :-)
> >    
>
> You are _so_ 32-bit, man ;^)
>  

Hey! I switched over from ASM to C in '86, but I still hate to
waste memory or CPU cycles.  ;-)

> Thanks,
> Phil
>  

Andy


--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>