This is what I did to migrate my live data accross the the cluster. I
didnt have any major issues. I utilized an existing slave so that I
didnt have to adjust operations on the master, eg stop slave on the
running slave host, dump data from the slave, then restore to cluster,
then start slave on both.
On 6/24/06, Alex Davies <alex@stripped> wrote:
> Hi Guys,
> I'm about to move a rapidly changing database over to NDB. Data loss
> or downtime (or extended locks) are not an option for this DB. Can
> anyone confirm that the following plan will work?
> 1) Dump current DB on live machine with --master-data=1 (this machine
> happens to be a replication master)
> 2) Import dump, alter to NDB
> 3) Setup one SQL node as a replication slave; start it
> 4) Wait for the NDB cluster to catch up on all the rows that changed
> since the dump was made (which will be a several hour period by the
> time we have imported the dump)
> 5) Move the application over to the new cluster
> 6) Stop the slave on the new cluster.
> I'm going to be doing this in the next few hours, so any comments
> would be appreciated.
> Many thanks,
> MySQL Cluster Mailing List
> For list archives: http://lists.mysql.com/cluster
> To unsubscribe: http://lists.mysql.com/cluster?unsub=1