new messaging system
rcalex at home.com
Wed Sep 13 21:43:32 GMT 2000
FWIW, having created several messaging systems over the years, me and my
partner in crime came to the understanding that the following general
structure was ideal.
Length (either inclusive or non-inclusive)
with appropriate sizes (short or long).
The only real debate was if Type should be first, but we settled that for
the following reason.
The 'smallest' program is one that simply reads a stream of data. The only
'data' that you need to know about is the length so you can consume the
record and get to the next. (this was a buffered data stream with about 30
messages per block)
This was for a global financial markets data feed and had to be super fast
(shared memory DB's) as well as 100% reliable.
I have no interest in a debate nor do I wish to claim this is better than
the current proposal, I just wanted to let you know of the experience of
some folks from another systems area. If you do not see any merit in the
concept for your new messaging system, that is fine with me.
From: samba-technical-admin at us4.samba.org
[mailto:samba-technical-admin at us4.samba.org]On Behalf Of Christopher R.
Sent: September 13, 2000 5:16 PM
To: tridge at linuxcare.com
Cc: Samba Technical
Subject: Re: new messaging system
> The next step is for me to write a little program that I'll probably
> call smbcontrol. It will have commands like:
> smbcontrol nmbd force-election
> smbcontrol smbd set-debug 7
> smbcontrol smbd/3267 set-debug 9
It would be nice if this could also be a command-line deal, eg.:
> nmbd force-election
> smbd list pids
smbd user OtherInfo
666 jra Tree id?
3267 tridge How 'bout share name?
10973 crh Perhaps some statistics?
That sort of thing. I'm finishing up my contract stint, so a little time
will be freed up. I have a backlog to catch up on, but I want to finish
the WINS failover (need Jeremy to okay a few things first). After that--
though probably not for the next release--I've got:
- implement the RFC behavior for group names (there was a patch, and we
talked about this at CIFS2000--I'll return the broadcast as the first
list entry as is currently done).
- Consider the multi-WINS-membership requests that came in (with heat and
- Get back into jCIFS some more.
- Work on more documentation.
Christopher R. Hertel -)----- University of Minnesota
crh at nts.umn.edu Networking and Telecommunications Services
Ideals are like stars; you will not succeed in touching them
with your hands...you choose them as your guides, and following
them you will reach your destiny. --Carl Schultz
More information about the samba-technical