If you’ve encountered an error message like the following:

Then you have a data corruption situation. This section describes how to address the issue.

If most of the replicas on the disk went bad, that means the disk is unreliable now and should be replaced.

If only one replica on the disk went bad, it can be a situation known as . In this case, removing the replica is good enough.

The reason for this is probably that the bad bit was written from the workload the volume attached to.

To revert to a previous snapshot:

1. Revert to a snapshot. You should start with the latest one.
2. Detach the volume from maintenance mode to any node.
3. Mount the volume from /dev/longhorn/<volume_name> and check the volume content.
4. If the volume content is still incorrect, repeat from step 1.