----- Original Message -----
> 
> In our case it would be impossible for the University's public web
> presence and the AD domain controllers to be the same machines.  It
> is
> conceivable that we could do some magic in load balancers to divide
> traffic appropriately, but I'd rather not do that if I don't have to.
> 
> Sam
> 
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
> _______________________________________________

But, assuming that your web presence is on the load balancer...there wouldn't 
be any trick to putting AD controller(s) on the same IP...since AD controllers 
listen to ports other than 80/443.

At our university (www.)ksu.edu is 129.130.8.49 and (www.)k-state.edu is 
129.130.8.50....on this IP, the load balance has port 80 mapped to a pool of 
webservers handling http, and port 443 is mapped to a different pool of 
webservers handling https (they should be the same servers now, but there was a 
time when the webteam was switching webserver apps, that SSL continued to be 
handled by the old servers since the private keys were internal to that 
application.)  The instability of our web presence was attributed the high 
activity content that was largely http..... until about 2.5 years ago, we were 
still using Netscape Enterprise Server v4.1!  And, there were things specific 
to that version that precluded moving to newer NES/iPlanet/SunOneWS....  
finally with to apache when a mod was written to recreate those features....and 
bugs.

Though our AD controllers are not behind our load balancer, but someday the 
windows group might....now that they want to be considered an enterprise server 
tech group....and cause all sorts of confusion with the already existing 
enterprise server tech group (unix/linux)...and shed their old name of lantech, 
from when they were the netware group

What we do have on this IP, is ports 5222 and 5223 being sent to another pool.

OTOH, I am doing some magic on the load balancers...because different URI paths 
are going to different pools, because some important section was mocked up 
using technology that is not our standard webserver but then is announced to 
the world as a path under our main web site.

The web team is has been talking about replacing our main web presence with 
varnish caches, which would give them the ability to do this 
themselves...rather needing me to maintain the TCL file that makes the magic.  
But, its been taking them a long time for some reason....(years).  I have a 
personal setup, which is a pair nginx servers reverse proxying to various other 
servers that's working pretty slick....

The use of separate IPs for ksu.edu & k-state.edu is a left over from how 
things used to be done....but the site now uses a multiname cert with those 4 
names and others...  since it was cheaper to cram as many different names into 
a single cert....  (and we're doing SSL proxy on our load balancer -- so the 
load balance can works its magic...)
_______________________________________________
Please visit https://lists.isc.org/mailman/listinfo/bind-users to unsubscribe 
from this list

bind-users mailing list
bind-users@lists.isc.org
https://lists.isc.org/mailman/listinfo/bind-users

Reply via email to