Details
Description
Customer request to allow for a node that has not had a catastrophic failure to re-use it's data on disk when rejoining a cluster (after it has been failed over).
This goes a long with being able to support larger datasets on each node. By forcing a rebalance after failover, we are potentially copying 100's of GB of data over the network when it is seemingly unnecessary. Not only does this take quite a bit of time over the network, it is using network bandwidth unnecessarily (potentially impacting performance) and resynchronizing from disk would alleviate this and speed up the process considerably.
Attachments
Issue Links
- is duplicated by
-
MB-9979 Delta node recovery after failover: a failed node should be able to catch up instead of being considered a new node
- Resolved