colin.surprenant at gmail.com
Wed Nov 24 10:53:53 EST 2010
>> 1. Take down one node at a time and innodump the data so I have a
>> known good backup
> If you're taking each node down, why not simply backup the filesystem
> instead of using innodump? Restoring the filesystem should work just as
> well, right? That way you'd have a backend agnostic procedure that would
> also be faster and you'd always backup the ring state (less steps).
fileystem level backup will be fine for disaster recovery if shit hit
the fan but if a straightforward upgrade does not work, data
extraction and restore in the new upgraded cluster will be required.
thanks all for your input, I will be testing a 0.10 to 0.13 (with
innostore) upgrade soon and let you guys know how that worked out.
More information about the riak-users