| Option | AZURE_RECOVER_FROM_BUCKET | |
| Description | Name of the Azure Blob container that holds the data to recover from during disaster recovery. | |
| Introduced Version | 1.231 | |
This environment variable specifies the Azure Blob storage
container name to recover data from. It is deployed only on the
secondary (standby) cluster when dr="standby" and
is removed when promoting to dr="active". Removing
this variable changes the HumioCluster spec and may cause the
operator to restart pods during promotion.
During reverse recovery (failback), this variable points to the secondary storage container.
This variable tells the cluster that it starts in a new
environment. The cluster checks it only when it finds no global
snapshot in the container that
AZURE_STORAGE_BUCKET specifies. If that container
holds a global snapshot, the cluster uses the snapshot and
ignores this variable.
The container can be a replicated copy of the primary container.
For example, you can replicate the primary container to a second
region and set this variable to the replica. Set
AZURE_STORAGE_BUCKET to a new, empty container in
the recovery region. LogScale migrates the segments from
the recovery container to the new container in the background.
Do not point AZURE_STORAGE_BUCKET at the
replicated container instead. The global snapshot in the replica
describes the old cluster, including its nodes and digest
assignments, so the new cluster cannot use it. This setup is not
supported.
Setting this variable requires
BUCKET_STORAGE_MULTIPLE_ENDPOINTS to be
true.