Cluster list,
In recent versions of cluster we have the new feature that works in the
following way.
If a SQL node is connected to the cluster at the time of database
creation by another SQL node then this database is also created on the
local SQL node. To a MySQLD the presence of a database is determined by
the presence of a folder in the filesystem.
What is the mechanism that implements this feature?
Does the SQL node connect as a client to each SQL node and run, create
database ?
Is the management node involved ?
Are the data nodes involved ?
A link to the documentation is fine, I just want to know how this is
implemented.
--
Tom
