Hi, We are planning to upgrade our hadoop cluster from 0.10.1 to 0.14 version. Any chance of not loosing data just by changing jars on all the nodes or should we think about writing a migration script to transfer all the data from the 0.10.1 cluster to the 0.14 one knowing we have around 1.5 TB of data which means we cannot store them temporarly on a local file sytem.
Has anyone experimented this ? Thanks for any help. Cheers, Thomas.
