70 - 059 TCP/IP with nt 4
Okay, Okay, so its an adaptive test now. Lets not overreact. I just passed it yesterday and thought Id give some positivte feedback as a counterbalance to all the negative voices out here. The good news comes in a couple of areas. Number one the questions arent any different.
The format of the test has changed in terms of which question will be your next one, but if youve been studying using resouces such as the sybex book, transcender, exam cram and of course this web site, then you wont need to change study habits in mid stream.
A second positive is that the test is still 90 minutes whereas the questions have be cut down to roughly half the original number. You used to have to wade through 58 questions in 90 minutes; now its only 25 to 35 questions in the same amount of time, so take a few deep breaths and dont feel rushed. I was watching the clock and anticipating a full 35 questions when after just 24 or 25 the test was over and I had another 45 minutes that could have been used but werent. In the other tests I always felt the time squeeze much more.
On my exam they actually broke it up into two sections, where the first possible 25 questions would be adaptive followed by an additional 10 non adaptive questions. I had assumed this meant that with my last 10 questions I would be able to back up and revisit them, but this turned out to be a naive assumption on my part. There was no going back, its just that the answers to those dont determine which question will appear next. I also couldnt tell when the transition took place between the two types of questions, although the last 10 were pretty standard a,b,c,d questions with neither graphics nor scenarios involving required and optional results.
A third positive is that the passing score has been lowered. It used to be 750, now its 570. Of course if you follow the inverse logic of these matters you might conclude that the test must be harder for them to lower the passing score, but who understands inverse logic anyway.
One question involved a graphic, but I couldnt figure out where the answeres where located. They didnt appeare after the question setup and the graphic didnt seem to have a way of providing answers. I kept looking at the more button in the lower left corner thinking that it was required to screen down somehow, even though that was normally the position where the next button was usually located. Finally gambled and pushed the more button. Turned out to be a faily major oops on my part, since it sent me on to the next question and I couldnt get back.
There is NO going back. Of course it was my own d*** fault for not being aware of the new methodology of answering the questions. Still it was frustrating to say the least. None of us can aford to throw away a full question like that, for whatever reason.
A few questions late I got another of those types (I had three altogether) and as many of you are aware the idea is to drag selected blocked answers in the graphic to the appropriate rectangles elsewhere in the graphic.
For studying, I used the exam cram and the transcender tests.
Personally I would never consider taking one of these test without first studying the transcender. At least three or four of the questions on the test were exact duplicates of ones in the transcender tests, and with the smaller nuber of questions now, that works out to about 10%.
Aside from giving you a feel for the types of questions to expect, they give you extensive feedback on the individual answers in terms of why B and E may have been correct and A,C and D were not. Once I begin to understand not only why a specific answer is correct, but more importantly why the other answers are bogus then I feel Im just about ready to take the test. Thats when I come visit this site to hear the feedback from others and look for other sample questions.
Some of the brain dumps I found useful for this test are: Laurence 13-4-98, Carmen 14-3-98, Studyguide 2-8-98, chilipepper 9-25-98, and rooster 23-8-98.
In terms of specific questions here are some I can remember:
1) Be aware of push pull scenarios where two servers are replicating to each other by count only rather than count and time criteria. In this case would they be pull partners, push partners or both and be prepared to drag and drop the appropriate label (i.e. push or pull, etc.) to the designated server in the graphic.
2) Another scenario was a graphic showing multiple subnets with workstations scattered around on each, all with a common subnet mask.
Each workstation had its host id and default gateway listed. The idea was to decide who was going to have trouble communicating and at first glance it didnt look like anyone would have a problem. The trick is to use the subnet mask to determine where the boundaries were for ip addresses within each subnet. (for 240 it would be groups of 14, for 224 it would be 30. Youve all seen the table by now and should have it memorized) Naturally they clump the addresses together to make it look good, but if you figure out where the boundaries are youll find one or two ip addresses with a default gateway that wornt work for it.
3) A scenario question involving subnetting again, with required and optional results. Youv probably already seen the advice on these types of questions. At first ignore the options and conccentrate on the required results. The questions are structured such that if the required results are not met the options are irrelevant and the answer is it doesnt meet the required results. Its too easy to get bogged down with the options, especially when you have to keep screening up and down to reference back and forth between the proposed solutions and the results. In this particular case the solution involved recognizing that with a specified subnet mask, some of the proposed host ids were in the invalid range just below w.x.y.254.
4) Had a question about static mapping for a ras server which totally threw me. Had to guess and wasnt too suprised to see another ras question a bit later. Youve all probably seen references to a scenario with a ras server connected to an ISP and how you configure the default gateway with a blank ip. Now the same scenario with 4 answers relating to default gateway configuration; two for the ras server - both wrong, and two from the client configuration perspective. Clients will use the ras servers local inteface ip as their default gateway. And incidently the ras servers ip configuration, with the exception of the default gateway being left blank, will conform to the subnet mask and ip addressing used by the local subnet to which it is attached. In other words its ip address and subnet mask wont have anything to do with the ISP.
5) There is a familiar scenario of not being able to connect to an FTP server due to permissions. However an alternative situation is being able to connect to the FTP service with nt explorer, but not with the command "ftp ftpserver.abccorp.com". None of the answers address permissions, but one possiblity is that the client is not set up to use DNS.
6) Understand the router tables and of course the route add command syntax. One question concerned two multihomed servers being used as routers between subnets a, b, and c with a real router connecting subnet c to subnet d. Was then asked to show how the router tables looked for each of the servers by dragging and dropping selected addresses to either the gateway address or network address column with each table for each server.
server A server B
network ip gateway ip network ip gateway ip
x.x.x.0 x.x.x.x x.x.x.0 x.x.x.x
x.x.x.0 x.x.x.x x.x.x.0 x.x.x.x
For the most part this is fairly straigtforward if you are familiar with the route add command, however every example Ive come across has only had two routers and three subnets. This question involved a third router and fourth subnet. I was unsure of the second entry for server A. The subnet d address I was confident of, but should the gateway ip be the nearside server b interface or the nearside real router interface? I chose the nearside server b, but it was a total guess.
Good Luck
five sixths