Re: [Vrs-development] Cluster Management

[Eric Altendorf]

On Wednesday 10 July 2002 20:31, Ian Fung wrote:
> example if an LDS leaves the vrs, its going to take a discrete
> amount of time before that information is propagted to other LDS's.
> so what is considered as the current version of the LDS list?

This is pretty much the same problem that we have with the distributed 
filesystem.

Any time that you need to have state (be it a filesystem or a list of 
LDS nodes) that is consistent among a number of distributed nodes, 
you need to use a two-phase commit protocol...  You *can* guarantee a 
single consistent shared LDS list, it's just tricky to do.

It does complicate things when the act of synchronization (the 
two-phase commit protocol) depends on the data we're trying to 
synchronize (the list of LDS nodes).  I haven't quite figured out how 
that will work.  Probably time to pull out my transaction processing 
book again and start reading. :-P

eric

-- 
"First they ignore you.  Then they laugh at you.
 Then they fight you.  And then you win."             -Gandhi