Anyone got any thoughts on this? I thought Snapshot and resilience/backup was a cornerstone of the 1.0 release, if so I would have thought someone would have had some idea on how to achieve 100% data safety.
On Friday, 6 December 2013 14:53:17 UTC, davrob2 wrote:
I've been having a look at the Snapshot API in Beta2.
Previously I've always relied on a database to be the primary source of my indexed data, I'm wondering if now is the time to consider going pure elasticsearch.
So, from the docs, it appears that snapshots are incremental, and can be done every 5 minutes or so, if needed.
However, is there a recommended procedure for restoring data that has not made it into the latest snapshot, perhaps by playing back the transaction log(s)?