No domain controller found after replacing switch...

[TheRaidersRock]

Hi everyone. We have a problem that we can't figure out- maybe someone here can help us? About two weeks ago, our main switch blew. We replaced it and reconfigured it, but ever since then we've been experiencing the following problem: when we start up our machines and log on, an error shows up in the event viewer stating that no domain controller can be found. Sometimes at logon a message pops up on the screen stating that we are being logged on with cached credentials, but usually it just shows up in the event viewer. The weird thing is, we can still ping the DC and access all network resources (except in the case where we wind up logging on with cached credentials- then removing it from the domain and rejoining it fixes it temporarily). A netlogon error also appears in the event viewer on the domain controller; basically, no domain controller can be found for domain y...
We thought that it was possible that the trust relationship between the x and y domains had been broken so we restarted both domain controllers- it didn't change a thing. We made sure to enable portfast on the switch, but that hasn't helped either. We're thinking that the issue must be related to the switch since that's when this problem first started, but we don't know what else we can do to fix it. It's been two weeks now, and we would really like to get this issue resolved. Can anyone help us out? Thank you very much in advance for your assistance- it's greatly appreciated.
Raider Fan
P.S. We're running Windows NT on the domain controllers with NT domains, and we've got NT, 2000Pro, and XP on the client workstations. The switch in question is a 2900. Thanks again.

 

[vipergg]

I can't think of anything with the switch that would cause this . You said that you had portfast on for all ports , I would just verify that it shows up in the config . You could probably try to debug ip packet on the switch and see if that gives you any idea of what is going on . Can't imagine what it would be , a 2900 is just plain L2 switch . Default gateway is ok???

 

[BuckWeet]

if you just have a single switch with a single vlan do a "no spanning vlan 1"


that will totally disable spanning tree


BuckWeet

 

[TheRaidersRock]

Thank you both; I really appreciate your help.
Does portfast need to be on for all ports, or just the one that the domain controller is plugged into?
Also, we have several switches with more than one vlan- is there anything we can do in that case?

 

[MTandSAV]

with the portfast command you expect only workstations on those ports. Basically the port starts to forward data directly, instead of going through blocking, listening, learning steps, which can take up to a minute.
I presume when those pc's had this problem, you did reboot them again, so by then the port will be in forwarding state for sure, so i don't think the problem is with that. It doesn't really matter if the port to the DC is portfast enabled
Make sure the port to the DC is in forwarding state. Don't know anything else..sorry

 

[vipergg]

Portfast should be turned on for all user ports so that they immedietly go to forwarding , this will eliminate dhcp ,ipx logon type problems .

 

[TheRaidersRock]

The one thing that still really concerns me though is the fact that the error message also appears on the domain controllers for both the x and y domains.
What would happen if we disabled the spanning tree algorithm?

 

[TheRaidersRock]

Three questions: 1. Would enabling portfast and disabling spanning tree be redundant? 2. Would disabling spanning tree allow loops to start occurring? 3. And what's the best way to do it?