Re: Change DC from multi-homed to single homed segmented

Tech Tip: Click here to run a free scan for Windows Errors and optimize PC performance

Thanks Bill,

I understand now. I will be replacing the current 100MB switches for 1GB
ones so will look into the VLAN option.

Regards
Terry

"Bill Grant" <not.available@online> wrote in message
news:urEV0MtWHHA.4832@xxxxxxxxxxxxxxxxxxxxxxx
You don't need to remove the subnets. What you are proposing would make
no difference to your browsing problems. You are just using different
numbers, but you still have multiple subnets and multiple NICs.

The solution is to use some other device as the router between the
subnets instead of using your DC. DHCP will still work OK if the router
can handle DHCP relay. If you move to a new switch you can probably do it
all with VLANs (ie the switch looks after the routing between subnets).

Once you have a routed network independent of the DC, browsing will
work if you configure all the machines to use WINS. WINS allows the
browser service to operate across a routed network (if the DMB isn't
multihomed). Each segment has its own browse master, and these can be
contacted by the DMB through WINS. The DMB then builds a network-wide
browse list so that all machines are in the browse list and all machines
can see the browse list.


"Terry" <new-grps@xxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:%23Z6cLdqWHHA.4188@xxxxxxxxxxxxxxxxxxxxxxx
Hi Bill,

Browser problems is the reason for removing the other subnets. Browsing
is not an issue on the main .1.1 subnet but as you correctly stated, can
be an issue between subnets on a multi-homed DC. However, I would still
like some advice on how to safely remove the other two subnets.

Regards

Terry

"Bill Grant" <not.available@online> wrote in message
news:%23UuSJZgWHHA.2284@xxxxxxxxxxxxxxxxxxxxxxx
If you want to improve browsing, the best thing to do is get the
routing off the DC. A multihomed DC is a major cause of browser
problems. If this is your only DC it automatically becomes the domain
master browser.

"Terry" <it@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:%23UyhV7ZWHHA.392@xxxxxxxxxxxxxxxxxxxxxxx
I have a Win2K3 DC (the only DC) with three NICS with subnets
192.168.1.1, .2.1 and .3.1.

DHCP enabled on all three.
WINS enabled on .1.1
Server is enabled as a Router.

I would like to segment the .1.1 range across at least two NICS, maybe
three.

Are there any 'gotchas' in reconfiguring the Server? DNS, DHCP, WINS,
Routing.
What could the subnet masks be on the Server and clients?

This is in preparation for moving away from 100MB switches to 1GB
switches and to provide better network browsing between the clients.

Regards
Terry















.

Relevant Pages

  • Re: Single domain two IP subnets
    ... VLANs would be the way to go. ... subnets in IP are far more than purely logical. ... VLANs switches, and switches in general, by many people. ... both Routers and Bridges -- we can call the features something ...
    (microsoft.public.win2000.dns)
  • Re: DNS replication on 2 domain on 2 forests
    ... Switches "route" (hate to use that term with a switch since it is ... to use when referring to those switches (layer 3 or router switches.) ... subnets, therefore, the answer is you need a router. ... It makes the network more efficient. ...
    (microsoft.public.win2000.dns)
  • Re: multiple subnets
    ... and a few gigabit switches to connect all hosts directly. ... If both subnets use distinct netmasks and broadcast patterns, ... We will get a router soon, but it is fast ethernet only. ... VLANs create separate "broadcast domains", and you would generally put different subnets on different VLANs so you could route between them. ...
    (comp.os.linux.networking)
  • Re: multiple subnets
    ... and a few gigabit switches to connect all hosts directly. ... If both subnets use distinct netmasks and broadcast patterns, ... compromise all the machines on all the networks. ...
    (comp.os.linux.networking)
  • Re: simplest way to intercept and use right mouse click?
    ... Anyway, the lesson here is that the script's behavior is dependent on browser settings that the script cannot control, and its utility is therefore rather limited. ... Ergo, any code that we write that is intended to be executed by a javascript interpreter in a browser is, in fact totally subject to the browser having the magic box 'allow javascript' ticked, right? ... Is it a pissing contest in which people who have no solutions to offer, attempt to outdo each other in willy waving, and tearing any attempt to solve a real life problem apart with reasons given being that - like javascript itself 'it wont work if the user switches it off'. ... It wont fucking work if they never visit the site either, ...
    (comp.lang.javascript)