Questions on some replication designs

Hi,

I have some questions regarding replication.

Question 1:

Assuming we have Database A at Server A, which has lots of transactions going on. Then we replicate the whole database to Server B. And then, we replicate the same database from Server B to Server C. The topology is as below:

1) Server A -> B -> C

When Server B is down, we need to replicate from Server A to Server C, must we do the re-initialization? Is there a way to ensure that the data are in sync without re-initialization? We understand that re-initialization is the recommended way, but just hope to hear some other suggestions here.

Question 2:

Assuming we have Database A at Server A, which has lots of transactions going on. Then we replicate the whole database to Server B and Server C. And then, we replicate again from Server B and Server C to Server D. The topology is as below:

1) Server A -> B

2) Server A -> C

3) Server B -> D

4) Server C -> D

We understand that this topology sounds weird, but also hope to hear from other suggestions here.

Thank you.