[nycbug-talk] SSH client session multiplexing

[Okan Demirmen]

On Sun 2004.07.25 at 19:59 -0500, Pete Wright wrote:
> Okan Demirmen wrote:
> 
> >On Sun 2004.07.25 at 02:33 -0400, Jesse Callaway wrote:
> > 
> >
> >>On Jul 23, 2004, at 5:57 AM, michael wrote:
> >>
> >>   
> >>
> >>>I was reading about session multiplexing on undeadly.org and was 
> >>>curious  what advantage this would have over running a single session 
> >>>to a host and running screen on the host.
> >>>
> >>>Does anyone have an opinion?
> >>>Michael
> >>>
> >>>     
> >>>
> >>I'd have to assume that running virtual screens on the host would be 
> >>most efficient in all aspects of the case.
> >>   
> >>
> >
> >ssh multiplexing and virtual screens are two different things. you
> >all know what virtual screens are using screen and window, but ssh
> >multiplexing allows one to, say, open up one control session to
> >your server then "piggy back" the rest of them over that channel -
> >the advantage is that one can do anything else over that same
> >channel: tunneling, more ttys, X forwarding, rsync, cvs, whatever
> >leveraging that one channel - net gain is performance. basically,
> >it speeds up all my actions to one server. try it, you'll love it ;)
> >
> > 
> >
> yea, agreed.  but michael was talking about "screen" a gnu utility that 
> allows you to run a console...have it detached, reattach at another 
> location etc...
> -p

right, i got that...i guess my mail didn't make that clear - i was
just trying to show that running "screen" or "window" is different
than ssh multiplexing...different tools for separate actions...i
don't think multiplexing and screen really can be compared... screen
is great for, say, irc'ing and ssh multiplexing works well when you
develop/work on a remote box.

$0.02

-- 
Okan Demirmen <okan at demirmen.com>
PGP-Key: http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xB3670934
PGP-Fingerprint: 226D B4AE 78A9 7F4E CD2B 1B44 C281 AF18 B367 0934