Re: reality check on two-way replication among peers
You are correct, I did not implement the slave of A1, A2, A3... correctly.
the mistake I made was not remembering that while a master can have
multiple slaves, a slave can have only ONE master.
I guess I have two options:
o Have a D1 acting as master for TD1, and a D2 acting as slave for
A1. Then chain A1 as slave of A2, A2 as slave of A3, and so on.
o Have a D1 acting as master for TD1, and a DA1 acting as slave of
A1, DA2 acting as slave of A2, and so on.
In my scenario the D1 and D2; or the D1, DA1, DA2, ... would be
multiple instances of mysql running on the same machine. Each
A1, A2, ... would be a seperate machine.
I think this would work but am not sure it is a good solution, and would
otherwise still have similar questions as in my original post. Comments?
Thanks,
mlk