In #1, we are given an example of a token ring, and are told that the wire linking C and D are broken. Then, we are asked which letters could connect to which.
The location of the token is only relevant to the behavior of the network in a very short period immediately after the damage. What I really want in your answer is to know which stations can "continue" to communicate, where "continue" implies for more than a few milliseconds. This makes the token position irrelevant.
I take it from your answer that the token is NOT a unique thing, ie. computer D CAN make another token. (and another and another etc.) If what I just said is not true, then the token would just "fall off" at the broken wire and all communication would stop. This seems like an unlikely way for a real network to operate.
Hmm. In the question I did mention that real networks have procedures for recovering from bad situations like this but that you should ignore such procedures for this question. So, it should not surprise you that the behavior the network described exhibits would seem unlikely for a real network.
Now, what is the position of the token when the wire is cut? For example, if E had the token when the damage was done, the token could be passed along from it to F, A, B, or C. However, if C had the token when the wire is cut, then, unless it could pass backwards, it coudn't pass the token along anywhere.
Is this relevant to the question, or am I just lost?
By Tom Murtagh (Admin) on Sunday, October 25, 1998 - 02:50 pm:
Tom
By Lichtman, Karen M. (02kml) on Monday, October 26, 1998 - 10:58 pm:
By Tom Murtagh (Admin) on Monday, October 26, 1998 - 11:12 pm:
The token must be a "unique thing" if any station could make up more tokens when it felt like it, the token would no longer ensure that only one station at a time would transmit.
Tom