70-068 NT Server in the Enterprise

This material is best read by first reading Yancy's addition to this site.    The following is all my opinion, of course, and I admit that it's conceivable that I might be wrong. Just my two cents.  Haven't taken the test yet but...

With regards to the question about migrating Netware accounts to NT and no accounts be overwritten-

If the question is stated correctly in Braindumps, then why not choose "start migration with default options"?

By default the migration tool won't overwrite anything. This won't transfer any duplicate accounts but the question doesn't say anything about requiring that.

Remember, if West trusts Corp then Corp users can log on from Corp or West (because of pass-through authentication) but West users can only log on from West. They have no option to logon from a computer connected to Corp domain. So what is really important is where the user account resides. If the question says that each user's account resides in their own domain and a West user is logging on to Corp, then they  have rights of corp\domain guests because they have no account in corp. If a corp user is logging on to Corp then they will have rights of corp\domain users. If the question doesn't say anything about where the user  accounts reside, as in this question according to Yancy's site, then it should be safe to assume that they have an account in that domain and have all rights of domain users. Guest account is disabled by default anyways.

The question about having just captured 400 frames. You've already captured the frames so you can't use a capture filter. You need to use a display filter which is option d, use sort by address in capture window.

Why does everyone choose PerfMon and NetMon for performance and baseline analysis? Netmon and Response Probe are not tools for establishing or examining a baseline.   Netmon especially is only used for realtime analysis, not for establishing baseline of performance. I've done a lot of research on this and I believe  PerfMon and Excel are the correct answers.

The Caracas, Chicago, Paris scenario. The single domain model meet required and one result. The master domain model, whether you assume it is single master or multiple master, does not meet the required results.

How can you have centralized users and still have trust relationships of chicago trusting paris and caracas and seattle, dallas, and new york trusting chicago. There must be users in paris, caracas, and chicago and chicago  has both users and resources.

You have single domain and 1 PDC and 10 BDC's. Users say logon is slow.

Here is the key- PerfMon says

that average logon/sec is 5 or higher. MOC says that for logon/sec you desire a high number. That means it's probably operating within acceptable bounds so the reason why logon is slow isn't because there aren't enough BDC's. There is too much synchronization between the controllers.  To free up some bandwidth you  have to lower the replication governor parameter from the default 100%.

Remember, if there is a virtual private network or intranet you must install a wins server and dns server in addition to setting up your ip addresses and www folders. If the question mentions no need for intranet then  ignore wins and dns. Never choose dhcp.

DHCP relay is for doing dhcp over subnets. Rip for IP is to have a dynamic

table that tells you instantly if a

router is or is not attached.

For fault tolerant wins servers use one for each domain and make them

push/pull partners. If no fault

tolerance is needed you can just use a wins proxy.

Well, that's all I can think of. I only touched upon what I thought were

the most controversial questions. And

remember, it's all just imho so research this stuff yourself!

Good luck and may the force be with you!

-GlassJoe