Estimate and Backup Plugin Parameters

CommunityEnterprise

These plugin parameters are relevant only for Backup and Estimate jobs:

Cluster Resources

namespace=<name>

specifies a Kubernetes namespace name to be backed up. Multiple namespace=<name> parameters are allowed if you want to backup multiple namespaces. If a namespace with the specified name cannot be found, it will be silently ignored. If this parameter is not set, all the objects in all namespaces will be saved. When performing a pvcdata backup, the namespace(s) must be specified. The Persistent Volume(s) in the specified namespace(s) will be included in the backup.

This parameter is required for pvcdata backup. Otherwise, this parameter is optional.

persistentvolume=<name>

specifies a Kubernetes Persistent Volume configuration to be backed up. Multiple persistentvolume=<name> parameters are allowed if you want to back up multiple volumes. Standard shell wildcard pattern matching can be used to select multiple volumes with a single persistentvolume parameter. If a persistent volume with the specified name cannot be found, it will be silently ignored. If this parameter is not set, all Persistent Volume configurations will be saved unless pvconfig=0 is set as described below or pvcdata is defined. If the pvcdata parameter is defined (in any form), only the Persistent Volumes linked to the defined pvcdata will be backed up.

This parameter is optional.

pvconfig=[0|1]

disables Persistent Volume configuration backups when pvconfig=0 is set. By default, Persistent Volume configurations are backed up.

This parameter is optional.

Note

Starting with Bacula Enterprise 18.1, if the parameter pvcdata is set, the plugin exclusively performs backups of the required Persistent Volume configuration.

storageclass=<name>

specifies a Kubernetes Storage Class configuration to be backed up. Multiple storageclass=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple volumes using a single storageclass parameter. If a storage class with the specified name cannot be found, it will be silently ignored. If this parameter is not set, all storage class configuration will be saved unless scconfig=0 is set as described below.

This parameter is optional.

scconfig=[0|1]

disables Storage Class configuration backups when scconfig=0 is set. By default, StorageClass object configurations are backed up.

This parameter is optional.

clusterrole=<name>

specifies a Kubernetes Cluster Role configuration to be backed up. Multiple clusterrole=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple cluster roles using a single clusterrole parameter. If a cluster role with name cannot be found, it will be silently ignored. If this parameter is not set, all Cluster Role and Cluster Role Binding configuration will be saved unless crconfig=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

crconfig=[0|1]

disables Cluster Role and Cluster Role Binding configuration backups when crconfig=0 is set. By default, ClusterRole object configurations are backed up.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

clusterrolebinding=<name>

specifies a Kubernetes Cluster Role Binding configuration to be backed up. Multiple clusterrolebinding=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple cluster role bindings using a single clusterrolebinding parameter. If a cluster role binding with name cannot be found, it will be silently ignored. If this parameter is not set, all Cluster Role Binding configuration will be saved unless crbconfig=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

crbconfig=[0|1]

disables Cluster Role Binding configuration backups when crbconfig=0 is set. The default is to backup Cluster Role Binding object configurations.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

migration_policy=<name>

specifies a KubeVirt Migration Policy configuration to be backed up. Multiple migration_policy=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple migration policies using a single migration_policy parameter. If a Migration Policy with name can not be found, it will be silently ignored. If this parameter is not set, all Migration Policy configuration will be saved unless vmcf_config=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

mp_config=[0|1]

disables Migration Policy configuration backups when mp_config=0 is set. By default, MigrationPolicy object configurations are backed up.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

virtual_machine_cluster_instance_type=<name>

specifies a KubeVirt Virtual Machine Cluster Instance Type configuration to be backed up. Multiple virtual_machine_cluster_instance_type=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple virtual machine cluster instance types using a single virtual_machine_cluster_instance_type parameter. If a virtual machine cluster instance type with name cannot be found, it will be silently ignored. If this parameter is not set, all Virtual Machine Cluster Instance Type configuration will be saved unless vmcf_config=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

vmcf_config=[0|1]

disables Virtual Machine Cluster Instance Type configuration backups when vmcf_config=0 is set. By default, Virtual Machine Cluster Instance Type object configurations are backed up.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

virtual_machine_cluster_preference=<name>

specifies a KubeVirt Virtual Machine Cluster Preference configuration to be backed up. Multiple virtual_machine_cluster_preference=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple Virtual Machine Cluster Preferences using a single virtual_machine_cluster_preference parameter. If a virtual machine cluster Preference with name cannot be found, it will be silently ignored. If this parameter is not set, all Virtual Machine Cluster Preference configuration will be saved unless vmcp_config=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.1.

vmcp_config=[0|1]

disables Virtual Machine Cluster Preference configuration backups when vmcp_config=0 is set. By default, Virtual Machine Cluster Preference object configurations are backed up.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.2.

volume_snapshot_class=<name>

specifies a Kubernetes Volume Snapshot Class configuration to be backed up. Multiple volume_snapshot_class=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple Volume Snapshot Classes using a single volume_snapshot_class parameter. If a Volume Snapshot Class with name cannot be found, it will be silently ignored. If this parameter is not set, all Volume Snapshot Class configuration will be saved unless vsc_config=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.2.

vsclass_config=[0|1]

disables Volume Snapshot Content configuration backups when vsclass_config=0 is set. By default, VolumeSnapshotClass object configurations are backed up.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.2.

volume_snapshot_content=<name>

specifies a Kubernetes Volume Snapshot Content configuration to be backed up. Multiple volume_snapshot_content=<name> parameters are allowed if you want to backup multiple objects. Standard shell wildcard pattern matching can be used to select multiple Volume Snapshot Contents using a single volume_snapshot_content parameter. If a Volume Snapshot Content with name cannot be found, it will be silently ignored. If this parameter is not set, all Volume Snapshot Content configuration will be saved unless vsc_config=0 is set as described below.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.2.

vsc_config=[0|1]

disables Volume Snapshot Content configuration backups when vsc_config=0 is set. By default, VolumeSnapshotContent object configurations are backed up.

This parameter is optional.

Note

This parameter is available since Bacula Enterprise 18.2.

Namespaced Resources

backup_mode=[snapshot|clone|standard]

sets the mode to perform backup of the persistent volume data (pvcdata backup). When this parameter is specified, the selected mode must be compatible with the underlying storage. If the Persistent Volume is not compatible with the backup mode specified (snapshot or clone), the plugin will try another backup mode in this order: Clone and Standard.

Note

The bacula/backup.mode annotation in he Kubernetes resource takes precedence over this parameter for the PVCs defined in the pod.

Note

The plugin automatically verifies whether the selected backup mode is compatible before starting the job. However, this does not mean that it will automatically switch to another mode if a failure occurs. If a backup fails while running in snapshot, clone, or standard mode, the plugin will not retry the operation using a different mode.

pvcdata[=<pvcname1,pvcname2,…>]

specifies Kubernetes Persistent Volume Claim names for which a PVC Data archive will be created. Since PVCs are namespaced objects, you must specify the required namespace(s) with the namespace=<name> parameter. If the pvcdata parameter is defined without an equals sign (“=”) and subsequent value, all detected PVCs in the specified namespace(s) will be selected for the PVC Data archive backup. You can select several Persistent Volume Claim in this parameter.

This parameter is optional.

Filtering Resources

For example, if we apply the following filters:

filter_pvcs_by_storage_class=longhorn and filter_pvcs_by_volume_mode=block, only the PVCs that meet both conditions will be backed up.

However, if we apply: filter_pvcs_by_storage_class=longhorn,ceph, then PVCs that belong to storageClass longhorn as well as those that belong to ceph will be backed up.

filter_pvcs_by_storage_class[=<storage_class_1,storage_class_2,…>]

when set, only PVCs whose StorageClass name matches the provided filter are included in the backup; all other PVCs are excluded.

This parameter is optional.

filter_pvcs_by_volume_mode[=block|filesystem]

when set, only PVCs whose Volume mode matches the provided filter are included in the backup; all other PVCs are excluded.

This parameter is optional.

Note

• Different filter types behave as a logical AND

• Multiple values within the same filter type behave as a logical OR.

exclude_namespaced_resources[=pods|secrets|configmaps|…]

excludes namespaced resources by the resource category. The resource category is the name used by Kubernetes.

Note

Persistent Volume Claims cannot be excluded.

Etcd Backup

etcd_backup=[snapshot]

enables etcd to be backed up. If this parameter is omitted, etcd is not backed up.

This parameter is optional.

etcd_namespace=<name>

specifies the kubernetes namespace name assigned to etcd. If this parameter is not defined, the default kube-system is used.

This parameter is optional and applies when etcd_backup is enabled.

etcd_api=<number>

specifies the API version that the etcdctl binary is required to operate with. If this parameter is not defined, the default 3 will be used.

This parameter is optional and applies when etcd_backup is enabled.

etcd_endpoints=<http://static-ip:etcd-port,https://domain-name:etcd-port>

defines the etcd endpoints that are utilized for operation. If the Kubernetes cluster has multiple etcd nodes, you must concatenate all etcd nodes.

This parameter depends on the activation of etcd_backup.

etcd_secret=<name>

defines the Kubernetes Secret which stores the etcd certificates and key. This Kubernetes Secret must be created in the defined etcd_namespace (by default kube-system).

This parameter depends on the activation of etcd_backup. This parameter is optional if you use the parameters etcd_ca_cert, etcd_tls_cert and etcd_tls_key.

etcd_ca_cert=</path/to/file>

is the CA certificate file location used to access to etcd.

This parameter depends on the activation of etcd_backup. This parameter is optional if you use the parameter etcd_secret.

etcd_tls_cert=</path/to/file>

is the TLS certificate file location used to access to etcd.

This parameter depends on the activation of etcd_backup. This parameter is optional if you use the parameter etcd_secret.

etcd_tls_key=</path/to/file>

is the TLS key file location used to access to etcd.

This parameter depends on the activation of etcd_backup. This parameter is optional if you use the parameter etcd_secret.

If none of the parameters above are specified, then all available Namespaces and Persistent Volume Configurations will be backed up. However, the plugin does not force a PVC Data archive backup neither etcd backup in this case.

longhorn_system=<longhorn-namespace>

defines the namespace where Longhorn is deployed. Longhorn documentation recommends using longhorn_system. The default value of this parameter is longhorn_system.

This parameter is optional and required only when performing a CSI snapshot backup with Longhorn as the provider.

longhorn_url=<longhorn-endpoint>

defines the Longhorn UI endpoint to manage Longhorn storages. For more information, refer to the Longhorn Snapshot Installation section.

This parameter is optional and required only when performing a snapshot backup with Longhorn without the CSI driver.

See also

Previous articles:

• Generic Plugin Parameters

Next articles:

• Backup and Restore Plugin Parameters

• Restore Plugin Parameters

• Tuning Plugin Parameters

• Fileset Examples

Go back to: Configuration.