Re: [gomd-devel] <GOMD2GOMD> preliminar code works...

[Matthias Rechenburg]

Hey JP,

On Mittwoch 30 Juli 2003 03:17, Gian Paolo Ghilardi wrote:
> Hi all.
>
> Preliminar gomd2gomd code seems to work.
> The code is experimental (some hardcoded stuff, as usual ;) ) and in these
> days I'll work to make it stable for CVS commit.
> BTW the change was _really_ minimal to the "client request processing"
> sequence (yeah! :) ).

yeaah, a good design (concept + code) can save masses of work :)))

>
> My test used:
> - 2 gomds executed on the same machine (first one listening on port 9889,
> the other one on port 9887)

ok, why not asking the "other" one on the default port too ?  (9889)

> - a telnet session.
>
> 1)I've asked "get cpuload 2" to gomd on 9889;
> 2)gomd on 9889 asked gomd on 9887 "get cpuload";
> 3)gomd on 9887 sent the answer to gomd on 9889;
> 4)gomd on 9889 sent the answer to me.

coool !

>
> Now the problem is: how can we get the list of nodes running gomd?

get nl    ;)    -> very easy because we can assume that the gomd is 
running on all nodes in the cluster. Maybe we can have a shorter
timeout if some gomd's are not reachable.

>
> PS: Matt => nice pic! Now I'm waiting for a kind of "migration gloves" from
> you... ;)

yep <lol> .... will see and keep you updated ;P


>
> CU.
>
> <rejected>
>
>
>
>
> _______________________________________________
> gomd-devel mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/gomd-devel

great progress again JP, very good.

enjoy,

Matt
-- 
E-mail  :  address@hidden
www     : http://www.openmosixview.com
an openMosix-cluster management GUI

Be safety conscious.  80% of people are caused by accidents.