On Sun, Mar 30, 2014 at 4:51 PM, Clint Kelly <clint.ke...@gmail.com> wrote:
> 1. *How many total unique tokens should I expect to see in my cluster?*
> If I have three nodes, and each node has a cassandra.yaml with num_tokens =
> 256, then should I expect a total of 256*3 = 768 distinct vnodes?
>
Yes. Generally, vnodes are just like nodes, except there are more of them
one of them per physical node.
> 2. *How does the creation of vnodes and their assignment to nodes relate
> to the replication factor for a given keyspace?*
>
The same way that it would if you created the same number of nodes with a
rack-unaware ("simple") snitch. If you have racks configured, it does the
rack thing with vnodes... which is less clear than in the CASSANDRA-3810
non-vnodes rack-aware no-op case, but logically the same.
> 3. What do the values in the "token" column in system.peers and
> system.local refer to then?
>
Node primary range ownership. Each node, v or not, has one and exactly one
token. The space between its token and the next token is the "primary
range" it is responsible for.
>
> - 4. Is there any other way, without using Thift, to get as much
> information as possible about what nodes contain replicas of data for all
> of the token ranges in a given cluster.
>
> I don't know the CQL answer to this, but for JMX there is
getNaturalEndpoints.
=Rob
