> Thanks David. I thought I saw this working sometime before and the
> previous
> response from Lea says it works for him, so are you saying this will
not
> work?
> That could be a problem because we are trying to run RDS(Resource
> Dependency Service)  which uses NMAP from the mainframe.

Probably wasn't very clear there, was I? My bad. 

No, I'm saying that it depends on how your network is configured as to
what type of nmap probes your Linux guests can respond to. The nmap TCP
and UDP-level probes will probably work fine. Some of the lower-level
stuff that nmap uses to identify host types, etc are layer 2 frames,
which the default layer 3 configuration won't pass, so your guests never
get an opportunity to reply. 

That's why I mentioned the TYPE ETHERNET guest lan and the layer
2/VSWITCH -- if you want all the nmap tricks to work, you need to ensure
that you're using one of those two basic network engineering
philosophies to get all the probes to work right. 

The "recommended" configuration (and the one that RDS assumes) is that
all the guests are either directly attached to an OSA or are on a
VSWITCH trunked to an outside switch, which makes them layer 2 clean,
and should allow nmap to work. This is probably a lousy assumption for
the Real World, but RDS (IMHO) isn't really living in the real world. On
the other hand, using a VSWITCH and pushing all those nasty networking
decisions outside the box is a Big Win, so it's probably a good idea to
do it. 





 

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to