How To Restore MySQL Cluster From A Data Node Filesystem Snapshot?

(Doc ID 2411116.1)

Last updated on JUNE 21, 2018

Applies to:

MySQL Cluster - Version 7.4 and later
Information in this document applies to any platform.

Goal

If it is required to recover data from a MySQL Cluster database and the only available backup for it is the snapshot for a data node file system, the following steps may help in recovering data. A data node, usually belonging to a node group with NoOfReplicas=2, stores both primary copies for dataset fragments and secondary copies as well. Because of that, the full dataset is available even in the case only one data node snapshot is available.

Following steps will take into account a MySQL Cluster deployment with only one node group configured. Same principles are valid also in the case of having multiple node groups; in such a case, a file system snapshot per data node in every node group must be available.

This is not a supported method and must be considered as an emergency recovery procedure, as recovered data will likely not be consistent per the intrinsic method used to backup the data.
The only solution to have a fully consistent restore procedure is performing a Point In Time Recovery it from a native backup.

Solution

Sign In with your My Oracle Support account

Don't have a My Oracle Support account? Click to get started

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms