1. Documentation
×

Persistent storage using Logical Volume Manager Storage

Logical Volume Manager Storage uses the TopoLVM CSI driver to dynamically provision local storage on the OKD clusters.

LVM Storage creates thin-provisioned volumes using Logical Volume Manager and provides dynamic provisioning of block storage on a limited resources cluster.

You can create volume groups, persistent volume claims (PVCs), volume snapshots, and volume clones by using LVM Storage.

Logical Volume Manager Storage installation

You can install Logical Volume Manager (LVM) Storage on an OKD cluster and configure it to dynamically provision storage for your workloads.

You can install LVM Storage by using the OKD CLI (oc), OKD web console, or Red Hat Advanced Cluster Management (RHACM).

When using LVM Storage on multi-node clusters, LVM Storage only supports provisioning local storage. LVM Storage does not support storage data replication mechanisms across nodes. You must ensure storage data replication through active or passive replication mechanisms to avoid a single point of failure.

Prerequisites to install LVM Storage

The prerequisites to install LVM Storage are as follows:

  • Ensure that you have a minimum of 10 milliCPU and 100 MiB of RAM.

  • Ensure that every managed cluster has dedicated disks that are used to provision storage. LVM Storage uses only those disks that are empty and do not contain file system signatures. To ensure that the disks are empty and do not contain file system signatures, wipe the disks before using them.

  • Before installing LVM Storage in a private CI environment where you can reuse the storage devices that you configured in the previous LVM Storage installation, ensure that you have wiped the disks that are not in use. If you do not wipe the disks before installing LVM Storage, you cannot reuse the disks without manual intervention.

    You cannot wipe the disks that are in use.

  • If you want to install LVM Storage by using Red Hat Advanced Cluster Management (RHACM), ensure that you have installed RHACM on an OKD cluster. See the "Installing LVM Storage using RHACM" section.

Additional resources

Installing LVM Storage by using the CLI

As a cluster administrator, you can install LVM Storage by using the OpenShift CLI.

Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to OKD as a user with cluster-admin and Operator installation permissions.

Procedure

  1. Create a YAML file with the configuration for creating a namespace:

    Example YAML configuration for creating a namespace
    apiVersion: v1
kind: Namespace
metadata:
  labels:
    openshift.io/cluster-monitoring: "true"
    pod-security.kubernetes.io/enforce: privileged
    pod-security.kubernetes.io/audit: privileged
    pod-security.kubernetes.io/warn: privileged
  name: openshift-storage

  2. Create the namespace by running the following command:

    $ oc create -f <file_name>

  3. Create an OperatorGroup CR YAML file:

    Example OperatorGroup CR
    apiVersion: operators.coreos.com/v1
kind: OperatorGroup
metadata:
  name: openshift-storage-operatorgroup
  namespace: openshift-storage
spec:
  targetNamespaces:
  - openshift-storage

  4. Create the OperatorGroup CR by running the following command:

    $ oc create -f <file_name>

  5. Create a Subscription CR YAML file:

    Example Subscription CR
    apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: lvms
  namespace: openshift-storage
spec:
  installPlanApproval: Automatic
  name: lvms-operator
  source: redhat-operators
  sourceNamespace: openshift-marketplace

  6. Create the Subscription CR by running the following command:

    $ oc create -f <file_name>
Verification

  1. To verify that LVM Storage is installed, run the following command:

    $ oc get csv -n openshift-storage -o custom-columns=Name:.metadata.name,Phase:.status.phase
    Example output
    Name                         Phase
4.13.0-202301261535          Succeeded

Installing LVM Storage by using the web console

You can install LVM Storage by using the OKD web console.

Prerequisites

  • You have access to the cluster.

  • You have access to OKD with cluster-admin and Operator installation permissions.

Procedure

  1. Log in to the OKD web console.

  2. Click Operators → OperatorHub.

  3. Click LVM Storage on the OperatorHub page.

  4. Set the following options on the Operator Installation page:

    1. Update Channel as stable-4.15.

    2. Installation Mode as A specific namespace on the cluster.

    3. Installed Namespace as Operator recommended namespace openshift-storage. If the openshift-storage namespace does not exist, it is created during the operator installation.

    4. Update approval as Automatic or Manual.

      If you select Automatic updates, the Operator Lifecycle Manager (OLM) automatically updates the running instance of LVM Storage without any intervention.

      If you select Manual updates, the OLM creates an update request. As a cluster administrator, you must manually approve the update request to update LVM Storage to a newer version.

  5. Optional: Select the Enable Operator recommended cluster monitoring on this Namespace checkbox.

  6. Click Install.

Verification steps

  • Verify that LVM Storage shows a green tick, indicating successful installation.

Installing LVM Storage in a disconnected environment

You can install LVM Storage on OKD in a disconnected environment. All sections referenced in this procedure are linked in the "Additional resources" section.

Prerequisites

  • You read the "About disconnected installation mirroring" section.

  • You have access to the OKD image repository.

  • You created a mirror registry.

Procedure

  1. Follow the steps in the "Creating the image set configuration" procedure. To create an ImageSetConfiguration custom resource (CR) for LVM Storage, you can use the following example ImageSetConfiguration CR configuration:

    Example ImageSetConfiguration CR for LVM Storage
    kind: ImageSetConfiguration
apiVersion: mirror.openshift.io/v1alpha2
archiveSize: 4 (1)
storageConfig: (2)
  registry:
    imageURL: example.com/mirror/oc-mirror-metadata (3)
    skipTLS: false
mirror:
  platform:
    channels:
    - name: stable-4.15 (4)
      type: ocp
    graph: true (5)
  operators:
  - catalog: registry.redhat.io/redhat/redhat-operator-index:v4.15 (6)
    packages:
    - name: lvms-operator (7)
      channels:
      - name: stable (8)
  additionalImages:
  - name: registry.redhat.io/ubi9/ubi:latest (9)
  helm: {}
    1 Set the maximum size (in GiB) of each file within the image set.
    2 Specify the location in which you want to save the image set. This location can be a registry or a local directory. You must configure the storageConfig field unless you are using the Technology Preview OCI feature.
    3 Specify the storage URL for the image stream when using a registry. For more information, see Why use imagestreams.
    4 Specify the channel from which you want to retrieve the OKD images.
    5 Set this field to true to generate the OpenShift Update Service (OSUS) graph image. For more information, see About the OpenShift Update Service.
    6 Specify the Operator catalog from which you want to retrieve the OKD images.
    7 Specify the Operator packages to include in the image set. If this field is empty, all packages in the catalog are retrieved.
    8 Specify the channels of the Operator packages to include in the image set. You must include the default channel for the Operator package even if you do not use the bundles in that channel. You can find the default channel by running the following command: $ oc mirror list operators --catalog=<catalog_name> --package=<package_name>.
    9 Specify any additional images to include in the image set.

  2. Follow the procedure in the "Mirroring an image set to a mirror registry" section.

  3. Follow the procedure in the "Configuring image registry repository mirroring" section.

Additional resources

Installing LVM Storage by using RHACM

To install LVM Storage on the clusters by using Red Hat Advanced Cluster Management (RHACM), you must create a Policy custom resource (CR). You can also configure the criteria to select the clusters on which you want to install LVM Storage.

The Policy CR that is created to install LVM Storage is also applied to the clusters that are imported or created after creating the Policy CR.
Prerequisites

  • You have access to the RHACM cluster using an account with cluster-admin and Operator installation permissions.

  • You have dedicated disks that LVM Storage can use on each cluster.

  • The cluster must be managed by RHACM.

Procedure

  1. Log in to the RHACM CLI using your OKD credentials.

  2. Create a namespace.

    $ oc create ns <namespace>

  3. Create a Policy CR YAML file:

    Example Policy CR to install and configure LVM Storage
    apiVersion: apps.open-cluster-management.io/v1
kind: PlacementRule
metadata:
  name: placement-install-lvms
spec:
  clusterConditions:
  - status: "True"
    type: ManagedClusterConditionAvailable
  clusterSelector: (1)
    matchExpressions:
    - key: mykey
      operator: In
      values:
      - myvalue
---
apiVersion: policy.open-cluster-management.io/v1
kind: PlacementBinding
metadata:
  name: binding-install-lvms
placementRef:
  apiGroup: apps.open-cluster-management.io
  kind: PlacementRule
  name: placement-install-lvms
subjects:
- apiGroup: policy.open-cluster-management.io
  kind: Policy
  name: install-lvms
---
apiVersion: policy.open-cluster-management.io/v1
kind: Policy
metadata:
  annotations:
    policy.open-cluster-management.io/categories: CM Configuration Management
    policy.open-cluster-management.io/controls: CM-2 Baseline Configuration
    policy.open-cluster-management.io/standards: NIST SP 800-53
  name: install-lvms
spec:
  disabled: false
  remediationAction: enforce
  policy-templates:
  - objectDefinition:
      apiVersion: policy.open-cluster-management.io/v1
      kind: ConfigurationPolicy
      metadata:
        name: install-lvms
      spec:
        object-templates:
        - complianceType: musthave
          objectDefinition: (2)
            apiVersion: v1
            kind: Namespace
            metadata:
              labels:
                openshift.io/cluster-monitoring: "true"
                pod-security.kubernetes.io/enforce: privileged
                pod-security.kubernetes.io/audit: privileged
                pod-security.kubernetes.io/warn: privileged
              name: openshift-storage
        - complianceType: musthave
          objectDefinition: (3)
            apiVersion: operators.coreos.com/v1
            kind: OperatorGroup
            metadata:
              name: openshift-storage-operatorgroup
              namespace: openshift-storage
            spec:
              targetNamespaces:
              - openshift-storage
        - complianceType: musthave
          objectDefinition: (4)
            apiVersion: operators.coreos.com/v1alpha1
            kind: Subscription
            metadata:
              name: lvms
              namespace: openshift-storage
            spec:
              installPlanApproval: Automatic
              name: lvms-operator
              source: redhat-operators
              sourceNamespace: openshift-marketplace
        remediationAction: enforce
        severity: low
    1 Set the key field and values field in PlacementRule.spec.clusterSelector to match the labels that are configured in the clusters on which you want to install LVM Storage.
    2 Namespace configuration.
    3 The OperatorGroup CR configuration.
    4 The Subscription CR configuration.

  4. Create the Policy CR by running the following command:

    $ oc create -f <file_name> -n <namespace>

    Upon creating the Policy CR, the following custom resources are created on the clusters that match the selection criteria configured in the PlacementRule CR:

    • Namespace

    • OperatorGroup

    • Subscription

About the LVMCluster custom resource

You can configure the LVMCluster CR to perform the following actions:

  • Create LVM volume groups that you can use to provision persistent volume claims (PVCs).

  • Configure a list of devices that you want to add to the LVM volume groups.

  • Configure the requirements to select the nodes on which you want to create an LVM volume group, and the thin pool configuration for the volume group.

  • Force wipe the selected devices.

After you have installed LVM Storage, you must create an LVMCluster custom resource (CR).

Example LVMCluster CR YAML file
apiVersion: lvm.topolvm.io/v1alpha1
kind: LVMCluster
metadata:
  name: my-lvmcluster
  namespace: openshift-storage
spec:
  tolerations:
  - effect: NoSchedule
    key: xyz
    operator: Equal
    value: "true"
  storage:
    deviceClasses:
    - name: vg1
      fstype: ext4 (1)
      default: true
      nodeSelector: (1)
        nodeSelectorTerms:
        - matchExpressions:
          - key: mykey
            operator: In
            values:
            - ssd
      deviceSelector: (1)
        paths:
        - /dev/disk/by-path/pci-0000:87:00.0-nvme-1
        - /dev/disk/by-path/pci-0000:88:00.0-nvme-1
        optionalPaths:
        - /dev/disk/by-path/pci-0000:89:00.0-nvme-1
        - /dev/disk/by-path/pci-0000:90:00.0-nvme-1
        forceWipeDevicesAndDestroyAllData: true
      thinPoolConfig:
        name: thin-pool-1
        sizePercent: 90 (1)
        overprovisionRatio: 10
1 Optional field

Explanation of fields in the LVMCluster CR

The LVMCluster CR fields are described in the following table:

Table 1. LVMCluster CR fields
Field Type Description

spec.storage.deviceClasses

array

Contains the configuration to assign the local storage devices to the LVM volume groups.

LVM Storage creates a storage class and volume snapshot class for each device class that you create.

After creating an LVMCluster CR, if you add or remove a device class, the update reflects in the LVMCluster CR only after deleting and recreating the topolvm-node pod.

deviceClasses.name

string

Specify a name for the LVM volume group (VG).

You can also configure this field to reuse a volume group that you created in the previous installation. For more information, see "Reusing a volume group from the previous LVM Storage installation".

deviceClasses.fstype

string

Set this field to ext4 or xfs. By default, this field is set to xfs.

deviceClasses.default

boolean

Set this field to true to indicate that a device class is the default. Otherwise, you can set it to false. You can only configure a single default device class.

deviceClasses.nodeSelector

object

Contains the configuration to choose the nodes on which you want to create the LVM volume group. If this field is empty, all nodes without no-schedule taints are considered.

On the control-plane node, LVM Storage detects and uses the additional worker nodes when the new nodes become active in the cluster.

nodeSelector.nodeSelectorTerms

array

Configure the requirements that are used to select the node.

deviceClasses.deviceSelector

object

Contains the configuration to perform the following actions:

  • Specify the paths to the devices that you want to add to the LVM volume group.

  • Force wipe the devices that are added to the LVM volume group.

For more information, see "About adding devices to a volume group".

deviceSelector.paths

array

Specify the device paths.

If the device path specified in this field does not exist, or the device is not supported by LVM Storage, the LVMCluster CR moves to the Failed state.

deviceSelector.optionalPaths

array

Specify the optional device paths.

If the device path specified in this field does not exist, or the device is not supported by LVM Storage, LVM Storage ignores the device without causing an error.

deviceSelector. forceWipeDevicesAndDestroyAllData

boolean

LVM Storage uses only those disks that are empty and do not contain file system signatures. To ensure that the disks are empty and do not contain file system signatures, wipe the disks before using them.

To force wipe the selected devices, set this field to true. By default, this field is set to false.

If this field is set to true, LVM Storage wipes all previous data on the devices. Use this feature with caution.

Wiping the device can lead to inconsistencies in data integrity if any of the following conditions are met:

  • The device is being used as swap space.

  • The device is part of a RAID array.

  • The device is mounted.

If any of these conditions are true, do not force wipe the disk. Instead, you must manually wipe the disk.

deviceClasses.thinPoolConfig

object

Contains the configuration to create a thin pool in the LVM volume group.

thinPoolConfig.name

string

Specify a name for the thin pool.

thinPoolConfig.sizePercent

integer

Specify the percentage of space in the LVM volume group for creating the thin pool.

By default, this field is set to 90. The minimum value that you can set is 10, and the maximum value is 90.

thinPoolConfig.overprovisionRatio

integer

Specify a factor by which you can provision additional storage based on the available storage in the thin pool.

For example, if this field is set to 10, you can provision up to 10 times the amount of available storage in the thin pool.

To disable over-provisioning, set this field to 1.
Additional resources

Limitations to configure the size of the devices used in LVM Storage

The limitations to configure the size of the devices that you can use to provision storage using LVM Storage are as follows:

  • The total storage size that you can provision is limited by the size of the underlying Logical Volume Manager (LVM) thin pool and the over-provisioning factor.

  • The size of the logical volume depends on the size of the Physical Extent (PE) and the Logical Extent (LE).

    • You can define the size of PE and LE during the physical and logical device creation.

    • The default PE and LE size is 4 MB.

    • If the size of the PE is increased, the maximum size of the LVM is determined by the kernel limits and your disk space.

Table 2. Size limits for different architectures using the default PE and LE size
Architecture RHEL 6 RHEL 7 RHEL 8 RHEL 9

32-bit

16 TB

-

-

-

64-bit

8 EB [1]

100 TB [2]

8 EB [1]

500 TB [2]

8 EB

8 EB

  1. Theoretical size.

  2. Tested size.

About adding devices to a volume group

The deviceSelector field in the LVMCluster CR contains the configuration to specify the paths to the devices that you want to add to the Logical Volume Manager (LVM) volume group.

You can specify the device paths in the deviceSelector.paths field, the deviceSelector.optionalPaths field, or both. If you do not specify the device paths in both the deviceSelector.paths field and the deviceSelector.optionalPaths field, LVM Storage adds the supported unused devices to the volume group (VG).

It is recommended to avoid referencing disks using symbolic naming, such as /dev/sdX, as these names may change across reboots within RHCOS. Instead, you must use stable naming schemes, such as /dev/disk/by-path/ or /dev/disk/by-id/, to ensure consistent disk identification.

With this change, you might need to adjust existing automation workflows in the cases where monitoring collects information about the install device for each node.

For more information, see the Fedora documentation.

You can add the path to the Redundant Array of Independent Disks (RAID) arrays in the deviceSelector field to integrate the RAID arrays with LVM Storage. You can create the RAID array by using the mdadm utility. LVM Storage does not support creating a software RAID.

You can create a RAID array only during an OKD installation. For information on creating a RAID array, see the following sections:

If you do not add the deviceSelector field in the LVMCluster CR, LVM Storage automatically adds the new devices when the devices are available.

LVM Storage adds the devices to the LVM volume group only if the following conditions are met:

  • The device path exists.

  • The device is supported by LVM Storage.

You can also add the path to the RAID arrays to integrate the RAID arrays with LVM Storage. For more information, see "Integrating RAID arrays with LVM Storage" in the "Additional resources" section.

After a device is added to the LVM volume group, it cannot be removed.
Additional resources

Devices not supported by LVM Storage

When you are adding the device paths in the deviceSelector field of the LVMCluster custom resource (CR), ensure that the devices are supported by LVM Storage. If you add paths to the unsupported devices, LVM Storage excludes the devices to avoid complexity in managing logical volumes.

If you do not specify any device path in the deviceSelector field, LVM Storage adds only the unused devices that it supports.

To get information about the devices, run the following command:

$ lsblk --paths --json -o \
NAME,ROTA,TYPE,SIZE,MODEL,VENDOR,RO,STATE,KNAME,SERIAL,PARTLABEL,FSTYPE

LVM Storage does not support the following devices:

Read-only devices

Devices with the ro parameter set to true.

Suspended devices

Devices with the state parameter set to suspended.

ROM devices

Devices with the type parameter set to rom.

LVM partition devices

Devices with the type parameter set to lvm.

Devices with invalid partition labels

Devices with the partlabel parameter set to bios, boot, or reserved.

Devices with an invalid filesystem

Devices with the fstype parameter set to any value other than null or LVM2_member.

LVM Storage supports devices with fstype parameter set to LVM2_member only if the devices do not contain children devices.
Devices that are part of another volume group

To get the information about the volume groups of the device, run the following command:

$ pvs <device-name> (1)
1 Replace <device-name> with the device name.
Devices with bind mounts

To get the mount points of a device, run the following command:

$ cat /proc/1/mountinfo | grep <device-name> (1)
1 Replace <device-name> with the device name.
Devices that contain children devices

It is recommended to wipe the device before using it in LVM Storage to prevent unexpected behavior.

Ways to create an LVMCluster custom resource

You can create an LVMCluster custom resource (CR) by using the OpenShift CLI (oc) or the OKD web console. If you have installed LVM Storage by using Red Hat Advanced Cluster Management (RHACM), you can also create an LVMCluster CR by using RHACM.

Upon creating the LVMCluster CR, LVM Storage creates the following system-managed CRs:

  • A storageClass and volumeSnapshotClass for each device class.

    LVM Storage configures the name of the storage class and volume snapshot class in the format lvms-<device_class_name>, where, <device_class_name> is the value of the deviceClasses.name field in the LVMCluster CR. For example, if the deviceClasses.name field is set to vg1, the name of the storage class and volume snapshot class is lvms-vg1.

  • LVMVolumeGroup: This CR is a specific type of persistent volume (PV) that is backed by an LVM volume group. It tracks the individual volume groups across multiple nodes.

  • LVMVolumeGroupNodeStatus: This CR tracks the status of the volume groups on a node.

Reusing a volume group from the previous LVM Storage installation

You can reuse an existing volume group (VG) from the previous LVM Storage installation instead of creating a new VG.

You can only reuse a VG but not the logical volume associated with the VG.

You can perform this procedure only while creating an LVMCluster custom resource (CR).
Prerequisites

  • The VG that you want to reuse must not be corrupted.

  • The VG that you want to reuse must have the lvms tag. For more information on adding tags to LVM objects, see Grouping LVM objects with tags.

Procedure

  1. Open the LVMCluster CR YAML file.

  2. Configure the LVMCluster CR parameters as described in the following example:

    Example LVMCluster CR YAML file
    apiVersion: lvm.topolvm.io/v1alpha1
kind: LVMCluster
metadata:
  name: my-lvmcluster
  namespace: openshift-storage
spec:
# ...
  storage:
    deviceClasses:
    - name: vg1  (1)
      fstype: ext4 (2)
      default: true
      deviceSelector: (3)
# ...
        forceWipeDevicesAndDestroyAllData: false (4)
      thinPoolConfig: (5)
# ...
      nodeSelector: (6)
# ...
    1 Set this field to the name of a VG from the previous LVM Storage installation.
    2 Set this field to ext4 or xfs. By default, this field is set to xfs.
    3 You can add new devices to the VG that you want to reuse by specifying the new device paths in the deviceSelector field. If you do not want to add new devices to the VG, ensure that the deviceSelector configuration in the current LVM Storage installation is same as that of the previous LVM Storage installation.
    4 If this field is set to true, LVM Storage wipes all the data on the devices that are added to the VG.
    5 To retain the thinPoolConfig configuration of the VG that you want to reuse, ensure that the thinPoolConfig configuration in the current LVM Storage installation is same as that of the previous LVM Storage installation. Otherwise, you can configure the thinPoolConfig field as required.
    6 Configure the requirements to choose the nodes on which you want to create the LVM volume group. If this field is empty, all nodes without no-schedule taints are considered.

  3. Save the LVMCluster CR YAML file.

To view the devices that are part a volume group, run the following command:

$ pvs -S vgname=<vg_name> (1)
1 Replace <vg_name> with the name of the volume group.

Creating an LVMCluster CR by using the CLI

You can create an LVMCluster custom resource (CR) on a worker node using the OpenShift CLI (oc).

You can only create a single instance of the LVMCluster custom resource (CR) on an OKD cluster.
Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to OKD as a user with cluster-admin privileges.

  • You have installed LVM Storage.

  • You have installed a worker node in the cluster.

  • You read the "About the LVMCluster custom resource" section.

Procedure

  1. Create an LVMCluster custom resource (CR) YAML file:

    Example LVMCluster CR YAML file
    apiVersion: lvm.topolvm.io/v1alpha1
kind: LVMCluster
metadata:
  name: my-lvmcluster
  namespace: openshift-storage
spec:
# ...
  storage:
    deviceClasses: (1)
# ...
      nodeSelector: (2)
# ...
      deviceSelector: (3)
# ...
      thinPoolConfig: (4)
# ...
    1 Contains the configuration to assign the local storage devices to the LVM volume groups.
    2 Contains the configuration to choose the nodes on which you want to create the LVM volume group. If this field is empty, all nodes without no-schedule taints are considered.
    3 Contains the configuration to specify the paths to the devices that you want to add to the LVM volume group, and force wipe the devices that are added to the LVM volume group.
    4 Contains the configuration to create a thin pool in the LVM volume group.

  2. Create the LVMCluster CR by running the following command:

    $ oc create -f <file_name>
    Example output
    lvmcluster/lvmcluster created
Verification

  1. Check that the LVMCluster CR is in the Ready state:

    $ oc get lvmclusters.lvm.topolvm.io -o jsonpath='{.items[*].status}' -n <namespace>
    Example output
    {"deviceClassStatuses": (1)
[
  {
    "name": "vg1",
    "nodeStatus": [ (2)
        {
            "devices": [ (3)
                "/dev/nvme0n1",
                "/dev/nvme1n1",
                "/dev/nvme2n1"
            ],
            "node": "kube-node", (4)
            "status": "Ready" (5)
        }
    ]
  }
]
"state":"Ready"} (6)
    1 The status of the device class.
    2 The status of the LVM volume group on each node.
    3 The list of devices used to create the LVM volume group.
    4 The node on which the device class is created.
    5 The status of the LVM volume group on the node.
    6 The status of the LVMCluster CR.

    If the LVMCluster CR is in the Failed state, you can view the reason for failure in the status field.

    Example of status field with the reason for failue:

    status:
  deviceClassStatuses:
    - name: vg1
      nodeStatus:
        - node: my-node-1.example.com
          reason: no available devices found for volume group
          status: Failed
  state: Failed

  2. Optional: To view the storage classes created by LVM Storage for each device class, run the following command:

    $ oc get storageclass
    Example output
    NAME          PROVISIONER          RECLAIMPOLICY   VOLUMEBINDINGMODE      ALLOWVOLUMEEXPANSION   AGE
lvms-vg1      topolvm.io           Delete          WaitForFirstConsumer   true                   31m

  3. Optional: To view the volume snapshot classes created by LVM Storage for each device class, run the following command:

    $ oc get volumesnapshotclass
    Example output
    NAME          DRIVER               DELETIONPOLICY   AGE
lvms-vg1      topolvm.io           Delete           24h
Additional resources

Creating an LVMCluster CR by using the web console

You can create an LVMCluster CR on a worker node using the OKD web console.

You can only create a single instance of the LVMCluster custom resource (CR) on an OKD cluster.
Prerequisites

  • You have access to the OKD cluster with cluster-admin privileges.

  • You have installed LVM Storage.

  • You have installed a worker node in the cluster.

  • You read the "About the LVMCluster custom resource" section.

Procedure

  1. Log in to the OKD web console.

  2. Click Operators → Installed Operators.

  3. In the openshift-storage namespace, click LVM Storage.

  4. Click Create LVMCluster and select either Form view or YAML view.

  5. Configure the required LVMCluster CR parameters.

  6. Click Create.

  7. Optional: If you want to edit the LVMCLuster CR, perform the following actions:

    1. Click the LVMCluster tab.

    2. From the Actions menu, select Edit LVMCluster.

    3. Click YAML and edit the required LVMCLuster CR parameters.

    4. Click Save.

Verification

  1. On the LVMCLuster page, check that the LVMCluster CR is in the Ready state.

  2. Optional: To view the available storage classes created by LVM Storage for each device class, click Storage → StorageClasses.

  3. Optional: To view the available volume snapshot classes created by LVM Storage for each device class, click Storage → VolumeSnapshotClasses.

Additional resources

Creating an LVMCluster CR by using RHACM

After you have installed LVM Storage by using RHACM, you must create an LVMCluster custom resource (CR).

Prerequisites

  • You have installed LVM Storage by using RHACM.

  • You have access to the RHACM cluster using an account with cluster-admin permissions.

  • You read the "About the LVMCluster custom resource" section.

Procedure

  1. Log in to the RHACM CLI using your OKD credentials.

  2. Create a ConfigurationPolicy CR YAML file with the configuration to create an LVMCluster CR:

    Example ConfigurationPolicy CR YAML file to create an LVMCluster CR
    apiVersion: policy.open-cluster-management.io/v1
kind: ConfigurationPolicy
metadata:
  name: lvms
  namespace: openshift-storage
spec:
  object-templates:
  - complianceType: musthave
    objectDefinition:
      apiVersion: lvm.topolvm.io/v1alpha1
      kind: LVMCluster
      metadata:
        name: my-lvmcluster
        namespace: openshift-storage
      spec:
        storage:
          deviceClasses: (1)
# ...
            deviceSelector: (2)
# ...
            thinPoolConfig: (3)
# ...
            nodeSelector: (4)
# ...
  remediationAction: enforce
  severity: low
    1 Contains the configuration to assign the local storage devices to the LVM volume groups.
    2 Contains the configuration to specify the paths to the devices that you want to add to the LVM volume group, and force wipe the devices that are added to the LVM volume group.
    3 Contains the LVM thin pool configuration.
    4 Contains the configuration to choose the nodes on which you want to create the LVM volume groups. If this field is empty, then all nodes without no-schedule taints are considered.

  3. Create the ConfigurationPolicy CR by running the following command:

    $ oc create -f <file_name> -n <cluster_namespace> (1)
    1 Namespace of the OKD cluster on which LVM Storage is installed.
Additional resources

Ways to delete an LVMCluster custom resource

You can delete an LVMCluster custom resource (CR) by using the OpenShift CLI (oc) or the OKD web console. If you have installed LVM Storage by using Red Hat Advanced Cluster Management (RHACM), you can also delete an LVMCluster CR by using RHACM.

Upon deleting the LVMCluster CR, LVM Storage deletes the following CRs:

  • storageClass

  • volumeSnapshotClass

  • LVMVolumeGroup

  • LVMVolumeGroupNodeStatus

Deleting an LVMCluster CR by using the CLI

You can delete the LVMCluster custom resource (CR) using the OpenShift CLI (oc).

Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

  • You have deleted the persistent volume claims (PVCs), volume snapshots, and volume clones provisioned by LVM Storage. You have also deleted the applications that are using these resources.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Delete the LVMCluster CR by running the following command:

    $ oc delete lvmcluster <lvmclustername> -n openshift-storage
Verification

  • To verify that the LVMCluster CR has been deleted, run the following command:

    $ oc get lvmcluster -n <namespace>
    Example output
    No resources found in openshift-storage namespace.

Deleting an LVMCluster CR by using the web console

You can delete the LVMCluster custom resource (CR) using the OKD web console.

Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

  • You have deleted the persistent volume claims (PVCs), volume snapshots, and volume clones provisioned by LVM Storage. You have also deleted the applications that are using these resources.

Procedure

  1. Log in to the OKD web console.

  2. Click Operators → Installed Operators to view all the installed Operators.

  3. Click LVM Storage in the openshift-storage namespace.

  4. Click the LVMCluster tab.

  5. From the Actions, select Delete LVMCluster.

  6. Click Delete.

Verification

  • On the LVMCLuster page, check that the LVMCluster CR has been deleted.

Deleting an LVMCluster CR by using RHACM

If you have installed LVM Storage by using Red Hat Advanced Cluster Management (RHACM), you can delete an LVMCluster CR by using RHACM.

Prerequisites

  • You have access to the RHACM cluster as a user with cluster-admin permissions.

  • You have deleted the persistent volume claims (PVCs), volume snapshots, and volume clones provisioned by LVM Storage. You have also deleted the applications that are using these resources.

Procedure

  1. Log in to the RHACM CLI using your OKD credentials.

  2. Delete the ConfigurationPolicy CR YAML file that was created for the LVMCluster CR:

    $ oc delete -f <file_name> -n <cluster_namespace> (1)
    1 Namespace of the OKD cluster on which LVM Storage is installed.

  3. Create a Policy CR YAML file to delete the LVMCluster CR:

    Example Policy CR to delete the LVMCluster CR
    apiVersion: policy.open-cluster-management.io/v1
kind: Policy
metadata:
  name: policy-lvmcluster-delete
  annotations:
    policy.open-cluster-management.io/standards: NIST SP 800-53
    policy.open-cluster-management.io/categories: CM Configuration Management
    policy.open-cluster-management.io/controls: CM-2 Baseline Configuration
spec:
  remediationAction: enforce
  disabled: false
  policy-templates:
    - objectDefinition:
        apiVersion: policy.open-cluster-management.io/v1
        kind: ConfigurationPolicy
        metadata:
          name: policy-lvmcluster-removal
        spec:
          remediationAction: enforce (1)
          severity: low
          object-templates:
            - complianceType: mustnothave
              objectDefinition:
                kind: LVMCluster
                apiVersion: lvm.topolvm.io/v1alpha1
                metadata:
                  name: my-lvmcluster
                  namespace: openshift-storage (2)
---
apiVersion: policy.open-cluster-management.io/v1
kind: PlacementBinding
metadata:
  name: binding-policy-lvmcluster-delete
placementRef:
  apiGroup: apps.open-cluster-management.io
  kind: PlacementRule
  name: placement-policy-lvmcluster-delete
subjects:
  - apiGroup: policy.open-cluster-management.io
    kind: Policy
    name: policy-lvmcluster-delete
---
apiVersion: apps.open-cluster-management.io/v1
kind: PlacementRule
metadata:
  name: placement-policy-lvmcluster-delete
spec:
  clusterConditions:
    - status: "True"
      type: ManagedClusterConditionAvailable
  clusterSelector: (3)
    matchExpressions:
      - key: mykey
        operator: In
        values:
          - myvalue
    1 The spec.remediationAction in policy-template is overridden by the preceding parameter value for spec.remediationAction.
    2 This namespace field must have the openshift-storage value.
    3 Configure the requirements to select the clusters. LVM Storage is uninstalled on the clusters that match the selection criteria.

  4. Create the Policy CR by running the following command:

    $ oc create -f <file_name> -n <namespace>

  5. Create a Policy CR YAML file to check if the LVMCluster CR has been deleted:

    Example Policy CR to check if the LVMCluster CR has been deleted
    apiVersion: policy.open-cluster-management.io/v1
kind: Policy
metadata:
  name: policy-lvmcluster-inform
  annotations:
    policy.open-cluster-management.io/standards: NIST SP 800-53
    policy.open-cluster-management.io/categories: CM Configuration Management
    policy.open-cluster-management.io/controls: CM-2 Baseline Configuration
spec:
  remediationAction: inform
  disabled: false
  policy-templates:
    - objectDefinition:
        apiVersion: policy.open-cluster-management.io/v1
        kind: ConfigurationPolicy
        metadata:
          name: policy-lvmcluster-removal-inform
        spec:
          remediationAction: inform (1)
          severity: low
          object-templates:
            - complianceType: mustnothave
              objectDefinition:
                kind: LVMCluster
                apiVersion: lvm.topolvm.io/v1alpha1
                metadata:
                  name: my-lvmcluster
                  namespace: openshift-storage (2)
---
apiVersion: policy.open-cluster-management.io/v1
kind: PlacementBinding
metadata:
  name: binding-policy-lvmcluster-check
placementRef:
  apiGroup: apps.open-cluster-management.io
  kind: PlacementRule
  name: placement-policy-lvmcluster-check
subjects:
  - apiGroup: policy.open-cluster-management.io
    kind: Policy
    name: policy-lvmcluster-inform
---
apiVersion: apps.open-cluster-management.io/v1
kind: PlacementRule
metadata:
  name: placement-policy-lvmcluster-check
spec:
  clusterConditions:
    - status: "True"
      type: ManagedClusterConditionAvailable
  clusterSelector:
    matchExpressions:
      - key: mykey
        operator: In
        values:
          - myvalue
    1 The policy-template spec.remediationAction is overridden by the preceding parameter value for spec.remediationAction.
    2 The namespace field must have the openshift-storage value.

  6. Create the Policy CR by running the following command:

    $ oc create -f <file_name> -n <namespace>
Verification

  • Check the status of the Policy CRs by running the following command:

    $ oc get policy -n <namespace>
    Example output
    NAME                       REMEDIATION ACTION   COMPLIANCE STATE   AGE
policy-lvmcluster-delete   enforce              Compliant          15m
policy-lvmcluster-inform   inform               Compliant          15m

    The Policy CRs must be in Compliant state.

Provisioning storage

After you have created the LVM volume groups using the LVMCluster custom resource (CR), you can provision the storage by creating persistent volume claims (PVCs).

The following are the minimum storage sizes that you can request for each file system type:

  • block: 8 MiB

  • xfs: 300 MiB

  • ext4: 32 MiB

To create a PVC, you must create a PersistentVolumeClaim object.

Prerequisites

  • You have created an LVMCluster CR.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Create a PersistentVolumeClaim object:

    Example PersistentVolumeClaim object
    apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: lvm-block-1 (1)
  namespace: default
spec:
  accessModes:
    - ReadWriteOnce
  volumeMode: Block (2)
  resources:
    requests:
      storage: 10Gi (3)
    limits:
      storage: 20Gi (4)
  storageClassName: lvms-vg1 (5)
    1 Specify a name for the PVC.
    2 To create a block PVC, set this field to Block. To create a file PVC, set this field to Filesystem.
    3 Specify the storage size. If the value is less than the minimum storage size, the requested storage size is rounded to the minimum storage size. The total storage size you can provision is limited by the size of the Logical Volume Manager (LVM) thin pool and the over-provisioning factor.
    4 Optional: Specify the storage limit. Set this field to a value that is greater than or equal to the minimum storage size. Otherwise, PVC creation fails with an error.
    5 The value of the storageClassName field must be in the format lvms-<device_class_name> where <device_class_name> is the value of the deviceClasses.name field in the LVMCluster CR. For example, if the deviceClasses.name field is set to vg1, you must set the storageClassName field to lvms-vg1.

    The volumeBindingMode field of the storage class is set to WaitForFirstConsumer.

  3. Create the PVC by running the following command:

    # oc create -f <file_name> -n <application_namespace>

    The created PVCs remain in Pending state until you deploy the pods that use them.
Verification

  • To verify that the PVC is created, run the following command:

    $ oc get pvc -n <namespace>
    Example output
    NAME          STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
lvm-block-1   Bound    pvc-e90169a8-fd71-4eea-93b8-817155f60e47   1Gi        RWO            lvms-vg1       5s

Ways to scale up the storage of clusters

OKD supports additional worker nodes for clusters on bare metal user-provisioned infrastructure. You can scale up the storage of clusters either by adding new worker nodes with available storage or by adding new devices to the existing worker nodes.

Logical Volume Manager (LVM) Storage detects and uses additional worker nodes when the nodes become active.

To add a new device to the existing worker nodes on a cluster, you must add the path to the new device in the deviceSelector field of the LVMCluster custom resource (CR).

You can add the deviceSelector field in the LVMCluster CR only while creating the LVMCluster CR. If you have not added the deviceSelector field while creating the LVMCluster CR, you must delete the LVMCluster CR and create a new LVMCluster CR containing the deviceSelector field.

If you do not add the deviceSelector field in the LVMCluster CR, LVM Storage automatically adds the new devices when the devices are available.

LVM Storage adds only the supported devices. For information about unsupported devices, see "Devices not supported by LVM Storage".
Additional resources

Scaling up the storage of clusters by using the CLI

You can scale up the storage capacity of the worker nodes on a cluster by using the OpenShift CLI (oc).

Prerequisites

  • You have additional unused devices on each cluster to be used by Logical Volume Manager (LVM) Storage.

  • You have installed the OpenShift CLI (oc).

  • You have created an LVMCluster custom resource (CR).

Procedure

  1. Edit the LVMCluster CR by running the following command:

    $ oc edit <lvmcluster_file_name> -n <namespace>

  2. Add the path to the new device in the deviceSelector field.

    Example LVMCluster CR
    apiVersion: lvm.topolvm.io/v1alpha1
kind: LVMCluster
metadata:
  name: my-lvmcluster
spec:
  storage:
    deviceClasses:
# ...
      deviceSelector: (1)
        paths: (2)
        - /dev/disk/by-path/pci-0000:87:00.0-nvme-1
        - /dev/disk/by-path/pci-0000:88:00.0-nvme-1
        optionalPaths: (3)
        - /dev/disk/by-path/pci-0000:89:00.0-nvme-1
        - /dev/disk/by-path/pci-0000:90:00.0-nvme-1
# ...
    1 Contains the configuration to specify the paths to the devices that you want to add to the LVM volume group. You can specify the device paths in the paths field, the optionalPaths field, or both. If you do not specify the device paths in both paths and optionalPaths, Logical Volume Manager (LVM) Storage adds the supported unused devices to the LVM volume group. LVM Storage adds the devices to the LVM volume group only if the following conditions are met:

    • The device path exists.

    • The device is supported by LVM Storage. For information about unsupported devices, see "Devices not supported by LVM Storage".
    2 Specify the device paths. If the device path specified in this field does not exist, or the device is not supported by LVM Storage, the LVMCluster CR moves to the Failed state.
    3 Specify the optional device paths. If the device path specified in this field does not exist, or the device is not supported by LVM Storage, LVM Storage ignores the device without causing an error.

    After a device is added to the LVM volume group, it cannot be removed.

  3. Save the LVMCluster CR.

Additional resources

Scaling up the storage of clusters by using the web console

You can scale up the storage capacity of the worker nodes on a cluster by using the OKD web console.

Prerequisites

  • You have additional unused devices on each cluster to be used by Logical Volume Manager (LVM) Storage.

  • You have created an LVMCluster custom resource (CR).

Procedure

  1. Log in to the OKD web console.

  2. Click Operators → Installed Operators.

  3. Click LVM Storage in the openshift-storage namespace.

  4. Click the LVMCluster tab to view the LVMCluster CR created on the cluster.

  5. From the Actions menu, select Edit LVMCluster.

  6. Click the YAML tab.

  7. Edit the LVMCluster CR to add the new device path in the deviceSelector field:

    Example LVMCluster CR
    apiVersion: lvm.topolvm.io/v1alpha1
kind: LVMCluster
metadata:
  name: my-lvmcluster
spec:
  storage:
    deviceClasses:
# ...
      deviceSelector: (1)
        paths: (2)
        - /dev/disk/by-path/pci-0000:87:00.0-nvme-1
        - /dev/disk/by-path/pci-0000:88:00.0-nvme-1
        optionalPaths: (3)
        - /dev/disk/by-path/pci-0000:89:00.0-nvme-1
        - /dev/disk/by-path/pci-0000:90:00.0-nvme-1
# ...
    1 Contains the configuration to specify the paths to the devices that you want to add to the LVM volume group. You can specify the device paths in the paths field, the optionalPaths field, or both. If you do not specify the device paths in both paths and optionalPaths, Logical Volume Manager (LVM) Storage adds the supported unused devices to the LVM volume group. LVM Storage adds the devices to the LVM volume group only if the following conditions are met:

    • The device path exists.

    • The device is supported by LVM Storage. For information about unsupported devices, see "Devices not supported by LVM Storage".
    2 Specify the device paths. If the device path specified in this field does not exist, or the device is not supported by LVM Storage, the LVMCluster CR moves to the Failed state.
    3 Specify the optional device paths. If the device path specified in this field does not exist, or the device is not supported by LVM Storage, LVM Storage ignores the device without causing an error.

    After a device is added to the LVM volume group, it cannot be removed.

  8. Click Save.

Additional resources

Scaling up the storage of clusters by using RHACM

You can scale up the storage capacity of worker nodes on the clusters by using RHACM.

Prerequisites

  • You have access to the RHACM cluster using an account with cluster-admin privileges.

  • You have created an LVMCluster custom resource (CR) by using RHACM.

  • You have additional unused devices on each cluster to be used by Logical Volume Manager (LVM) Storage.

Procedure

  1. Log in to the RHACM CLI using your OKD credentials.

  2. Edit the LVMCluster CR that you created using RHACM by running the following command:

    $ oc edit -f <file_name> -ns <namespace> (1)
    1 Replace <file_name> with the name of the LVMCluster CR.

  3. In the LVMCluster CR, add the path to the new device in the deviceSelector field.

    Example LVMCluster CR:
    apiVersion: policy.open-cluster-management.io/v1
      kind: ConfigurationPolicy
      metadata:
        name: lvms
      spec:
        object-templates:
           - complianceType: musthave
             objectDefinition:
               apiVersion: lvm.topolvm.io/v1alpha1
               kind: LVMCluster
               metadata:
                 name: my-lvmcluster
                 namespace: openshift-storage
               spec:
                 storage:
                   deviceClasses:
# ...
                     deviceSelector: (1)
                       paths: (2)
                       - /dev/disk/by-path/pci-0000:87:00.0-nvme-1
                       optionalPaths: (3)
                       - /dev/disk/by-path/pci-0000:89:00.0-nvme-1
# ...
    1 Contains the configuration to specify the paths to the devices that you want to add to the LVM volume group. You can specify the device paths in the paths field, the optionalPaths field, or both. If you do not specify the device paths in both paths and optionalPaths, Logical Volume Manager (LVM) Storage adds the supported unused devices to the LVM volume group. LVM Storage adds the devices to the LVM volume group only if the following conditions are met:

    • The device path exists.

    • The device is supported by LVM Storage. For information about unsupported devices, see "Devices not supported by LVM Storage".
    2 Specify the device paths. If the device path specified in this field does not exist, or the device is not supported by LVM Storage, the LVMCluster CR moves to the Failed state.
    3 Specify the optional device paths. If the device path specified in this field does not exist, or the device is not supported by LVM Storage, LVM Storage ignores the device without causing an error.

    After a device is added to the LVM volume group, it cannot be removed.

  4. Save the LVMCluster CR.

Additional resources

Expanding a persistent volume claim

After scaling up the storage of a cluster, you can expand the existing persistent volume claims (PVCs).

To expand a PVC, you must update the storage field in the PVC.

Prerequisites

  • Dynamic provisioning is used.

  • The StorageClass object associated with the PVC has the allowVolumeExpansion field set to true.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Update the value of the spec.resources.requests.storage field to a value that is greater than the current value by running the following command:

    $ oc patch <pvc_name> -n <application_namespace> -p \ (1)
'{ "spec": { "resources": { "requests": { "storage": "<desired_size>" }}}} --type=merge' (2)
    1 Replace <pvc_name> with the name of the PVC that you want to expand.
    2 Replace <desired_size> with the new size to expand the PVC.
Verification

  • To verify that resizing is completed, run the following command:

    $ oc get pvc <pvc_name> -n <application_namespace> -o=jsonpath={.status.capacity.storage}

    LVM Storage adds the Resizing condition to the PVC during expansion. It deletes the Resizing condition after the PVC expansion.

Additional resources

Deleting a persistent volume claim

You can delete a persistent volume claim (PVC) by using the OpenShift CLI (oc).

Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Delete the PVC by running the following command:

    $ oc delete pvc <pvc_name> -n <namespace>
Verification

  • To verify that the PVC is deleted, run the following command:

    $ oc get pvc -n <namespace>

    The deleted PVC must not be present in the output of this command.

About volume snapshots

You can create snapshots of persistent volume claims (PVCs) that are provisioned by LVM Storage.

You can perform the following actions using the volume snapshots:

  • Back up your application data.

    Volume snapshots are located on the same devices as the original data. To use the volume snapshots as backups, you must move the snapshots to a secure location. You can use OpenShift API for Data Protection (OADP) backup and restore solutions. For information about OADP, see "OADP features".

  • Revert to a state at which the volume snapshot was taken.

You can also create volume snapshots of the volume clones.

Limitations for creating volume snapshots in multi-node topology

LVM Storage has the following limitations for creating volume snapshots in multi-node topology:

  • Creating volume snapshots is based on the LVM thin pool capabilities.

  • After creating a volume snapshot, the node must have additional storage space for further updating the original data source.

  • You can create volume snapshots only on the node where you have deployed the original data source.

  • Pods relying on the PVC that uses the snapshot data can be scheduled only on the node where you have deployed the original data source.

Additional resources

Creating volume snapshots

You can create volume snapshots based on the available capacity of the thin pool and the over-provisioning limits. To create a volume snapshot, you must create a VolumeSnapshotClass object.

Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

  • You ensured that the persistent volume claim (PVC) is in Bound state. This is required for a consistent snapshot.

  • You stopped all the I/O to the PVC.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Create a VolumeSnapshot object:

    Example VolumeSnapshot object
    apiVersion: snapshot.storage.k8s.io/v1
kind: VolumeSnapshot
metadata:
  name: lvm-block-1-snap (1)
spec:
  source:
    persistentVolumeClaimName: lvm-block-1 (2)
  volumeSnapshotClassName: lvms-vg1 (3)
    1 Specify a name for the volume snapshot.
    2 Specify the name of the source PVC. LVM Storage creates a snapshot of this PVC.
    3 Set this field to the name of a volume snapshot class.

    To get the list of available volume snapshot classes, run the following command:

    $ oc get volumesnapshotclass

  3. Create the volume snapshot in the namespace where you created the source PVC by running the following command:

    $ oc create -f <file_name> -n <namespace>

    LVM Storage creates a read-only copy of the PVC as a volume snapshot.

Verification

  • To verify that the volume snapshot is created, run the following command:

    $ oc get volumesnapshot -n <namespace>
    Example output
    NAME               READYTOUSE   SOURCEPVC     SOURCESNAPSHOTCONTENT   RESTORESIZE   SNAPSHOTCLASS   SNAPSHOTCONTENT                                    CREATIONTIME   AGE
lvm-block-1-snap   true         lvms-test-1                           1Gi           lvms-vg1        snapcontent-af409f97-55fc-40cf-975f-71e44fa2ca91   19s            19s

    The value of the READYTOUSE field for the volume snapshot that you created must be true.

Restoring volume snapshots

To restore a volume snapshot, you must create a persistent volume claim (PVC) with the dataSource.name field set to the name of the volume snapshot.

The restored PVC is independent of the volume snapshot and the source PVC.

Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

  • You have created a volume snapshot.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Create a PersistentVolumeClaim object with the configuration to restore the volume snapshot:

    Example PersistentVolumeClaim object to restore a volume snapshot
    kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: lvm-block-1-restore
spec:
  accessModes:
  - ReadWriteOnce
  volumeMode: Block
  Resources:
    Requests:
      storage: 2Gi (1)
  storageClassName: lvms-vg1 (2)
  dataSource:
    name: lvm-block-1-snap (3)
    kind: VolumeSnapshot
    apiGroup: snapshot.storage.k8s.io
    1 Specify the storage size of the restored PVC. The storage size of the requested PVC must be greater than or equal to the stoage size of the volume snapshot that you want to restore. If a larger PVC is required, you can also resize the PVC after restoring the volume snapshot.
    2 Set this field to the value of the storageClassName field in the source PVC of the volume snapshot that you want to restore.
    3 Set this field to the name of the volume snapshot that you want to restore.

  3. Create the PVC in the namespace where you created the volume snapshot by running the following command:

    $ oc create -f <file_name> -n <namespace>
Verification

  • To verify that the volume snapshot is restored, run the following command:

    $ oc get pvc -n <namespace>
    Example output
    NAME                  STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
lvm-block-1-restore   Bound    pvc-e90169a8-fd71-4eea-93b8-817155f60e47   1Gi        RWO            lvms-vg1       5s

Deleting volume snapshots

You can delete the volume snapshots of the persistent volume claims (PVCs).

When you delete a persistent volume claim (PVC), LVM Storage deletes only the PVC, but not the snapshots of the PVC.
Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

  • You have ensured that the volume snpashot that you want to delete is not in use.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Delete the volume snapshot by running the following command:

    $ oc delete volumesnapshot <volume_snapshot_name> -n <namespace>
Verification

  • To verify that the volume snapshot is deleted, run the following command:

    $ oc get volumesnapshot -n <namespace>

    The deleted volume snapshot must not be present in the output of this command.

About volume clones

A volume clone is a duplicate of an existing persistent volume claim (PVC). You can create a volume clone to make a point-in-time copy of the data.

Limitations for creating volume clones in multi-node topology

LVM Storage has the following limitations for creating volume clones in multi-node topology:

  • Creating volume clones is based on the LVM thin pool capabilities.

  • The node must have additional storage after creating a volume clone for further updating the original data source.

  • You can create volume clones only on the node where you have deployed the original data source.

  • Pods relying on the PVC that uses the clone data can be scheduled only on the node where you have deployed the original data source.

Creating volume clones

To create a clone of a persistent volume claim (PVC), you must create a PersistentVolumeClaim object in the namespace where you created the source PVC.

The cloned PVC has write access.
Prerequisites

  • You ensured that the source PVC is in Bound state. This is required for a consistent clone.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Create a PersistentVolumeClaim object:

    Example PersistentVolumeClaim object to create a volume clone
    kind: PersistentVolumeClaim
apiVersion: v1
metadata:
  name: lvm-pvc-clone
spec:
  accessModes:
  - ReadWriteOnce
  storageClassName: lvms-vg1 (1)
  volumeMode: Filesystem (2)
  dataSource:
    kind: PersistentVolumeClaim
    name: lvm-pvc (3)
  resources:
    requests:
      storage: 1Gi (4)
    1 Set this field to the value of the storageClassName field in the source PVC.
    2 Set this field to the volumeMode field in the source PVC.
    3 Specify the name of the source PVC.
    4 Specify the storage size for the cloned PVC. The storage size of the cloned PVC must be greater than or equal to the storage size of the source PVC.

  3. Create the PVC in the namespace where you created the source PVC by running the following command:

    $ oc create -f <file_name> -n <namespace>
Verification

  • To verify that the volume clone is created, run the following command:

    $ oc get pvc -n <namespace>
    Example output
    NAME                STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
lvm-block-1-clone   Bound    pvc-e90169a8-fd71-4eea-93b8-817155f60e47   1Gi        RWO            lvms-vg1       5s

Deleting volume clones

You can delete volume clones.

When you delete a persistent volume claim (PVC), LVM Storage deletes only the source persistent volume claim (PVC) but not the clones of the PVC.
Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Delete the cloned PVC by running the following command:

    # oc delete pvc <clone_pvc_name> -n <namespace>
Verification

  • To verify that the volume clone is deleted, run the following command:

    $ oc get pvc -n <namespace>

    The deleted volume clone must not be present in the output of this command.

Updating LVM Storage

You can update LVM Storage to ensure compatibility with the OKD version.

Prerequisites

  • You have updated your OKD cluster.

  • You have installed a previous version of LVM Storage.

  • You have installed the OpenShift CLI (oc).

  • You have access to the cluster using an account with cluster-admin permissions.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Update the Subscription custom resource (CR) that you created while installing LVM Storage by running the following command:

    $ oc patch subscription lvms-operator -n openshift-storage --type merge --patch '{"spec":{"channel":"<update_channel>"}}' (1)
    1 Replace <update_channel> with the version of LVM Storage that you want to install. For example, stable-4.15.

  3. View the update events to check that the installation is complete by running the following command:

    $ oc get events -n openshift-storage
    Example output
    ...
8m13s       Normal    RequirementsUnknown   clusterserviceversion/lvms-operator.v4.15   requirements not yet checked
8m11s       Normal    RequirementsNotMet    clusterserviceversion/lvms-operator.v4.15   one or more requirements couldn't be found
7m50s       Normal    AllRequirementsMet    clusterserviceversion/lvms-operator.v4.15   all requirements found, attempting install
7m50s       Normal    InstallSucceeded      clusterserviceversion/lvms-operator.v4.15   waiting for install components to report healthy
7m49s       Normal    InstallWaiting        clusterserviceversion/lvms-operator.v4.15   installing: waiting for deployment lvms-operator to become ready: deployment "lvms-operator" waiting for 1 outdated replica(s) to be terminated
7m39s       Normal    InstallSucceeded      clusterserviceversion/lvms-operator.v4.15   install strategy completed with no errors
...
Verification

  • Verify the LVM Storage version by running the following command:

    $ oc get subscription lvms-operator -n openshift-storage -o jsonpath='{.status.installedCSV}'
    Example output
    lvms-operator.v4.15

Monitoring LVM Storage

To enable cluster monitoring, you must add the following label in the namespace where you have installed LVM Storage:

openshift.io/cluster-monitoring=true

For information about enabling cluster monitoring in RHACM, see Observability and Adding custom metrics.

Metrics

You can monitor LVM Storage by viewing the metrics.

The following table describes the topolvm metrics:

Table 3. topolvm metrics
Alert Description

topolvm_thinpool_data_percent

Indicates the percentage of data space used in the LVM thinpool.

topolvm_thinpool_metadata_percent

Indicates the percentage of metadata space used in the LVM thinpool.

topolvm_thinpool_size_bytes

Indicates the size of the LVM thin pool in bytes.

topolvm_volumegroup_available_bytes

Indicates the available space in the LVM volume group in bytes.

topolvm_volumegroup_size_bytes

Indicates the size of the LVM volume group in bytes.

topolvm_thinpool_overprovisioned_available

Indicates the available over-provisioned size of the LVM thin pool in bytes.

Metrics are updated every 10 minutes or when there is a change, such as a new logical volume creation, in the thin pool.

Alerts

When the thin pool and volume group reach maximum storage capacity, further operations fail. This can lead to data loss.

LVM Storage sends the following alerts when the usage of the thin pool and volume group exceeds a certain value:

Table 4. LVM Storage alerts
Alert Description

VolumeGroupUsageAtThresholdNearFull

This alert is triggered when both the volume group and thin pool usage exceeds 75% on nodes. Data deletion or volume group expansion is required.

VolumeGroupUsageAtThresholdCritical

This alert is triggered when both the volume group and thin pool usage exceeds 85% on nodes. In this case, the volume group is critically full. Data deletion or volume group expansion is required.

ThinPoolDataUsageAtThresholdNearFull

This alert is triggered when the thin pool data uusage in the volume group exceeds 75% on nodes. Data deletion or thin pool expansion is required.

ThinPoolDataUsageAtThresholdCritical

This alert is triggered when the thin pool data usage in the volume group exceeds 85% on nodes. Data deletion or thin pool expansion is required.

ThinPoolMetaDataUsageAtThresholdNearFull

This alert is triggered when the thin pool metadata usage in the volume group exceeds 75% on nodes. Data deletion or thin pool expansion is required.

ThinPoolMetaDataUsageAtThresholdCritical

This alert is triggered when the thin pool metadata usage in the volume group exceeds 85% on nodes. Data deletion or thin pool expansion is required.

Uninstalling LVM Storage by using the CLI

You can uninstall LVM Storage by using the OpenShift CLI (oc).

Prerequisites

  • You have logged in to oc as a user with cluster-admin permissions.

  • You deleted the persistent volume claims (PVCs), volume snapshots, and volume clones provisioned by LVM Storage. You have also deleted the applications that are using these resources.

  • You deleted the LVMCluster custom resource (CR).

Procedure

  1. Get the currentCSV value for the LVM Storage Operator by running the following command:

    $ oc get subscription.operators.coreos.com lvms-operator -n <namespace> -o yaml | grep currentCSV
    Example output
    currentCSV: lvms-operator.v4.15.3

  2. Delete the subscription by running the following command:

    $ oc delete subscription.operators.coreos.com lvms-operator -n <namespace>
    Example output
    subscription.operators.coreos.com "lvms-operator" deleted

  3. Delete the CSV for the LVM Storage Operator in the target namespace by running the following command:

    $ oc delete clusterserviceversion <currentCSV> -n <namespace> (1)
    1 Replace <currentCSV> with the currentCSV value for the LVM Storage Operator.
    Example output
    clusterserviceversion.operators.coreos.com "lvms-operator.v4.15.3" deleted
Verification

  • To verify that the LVM Storage Operator is uninstalled, run the following command:

    $ oc get csv -n <namespace>

    If the LVM Storage Operator was successfully uninstalled, it does not appear in the output of this command.

Uninstalling LVM Storage by using the web console

You can uninstall LVM Storage using the OKD web console.

Prerequisites

  • You have access to OKD as a user with cluster-admin permissions.

  • You have deleted the persistent volume claims (PVCs), volume snapshots, and volume clones provisioned by LVM Storage. You have also deleted the applications that are using these resources.

  • You have deleted the LVMCluster custom resource (CR).

Procedure

  1. Log in to the OKD web console.

  2. Click Operators → Installed Operators.

  3. Click LVM Storage in the openshift-storage namespace.

  4. Click the Details tab.

  5. From the Actions menu, select Uninstall Operator.

  6. Optional: When prompted, select the Delete all operand instances for this operator checkbox to delete the operand instances for LVM Storage.

  7. Click Uninstall.

Uninstalling LVM Storage installed using RHACM

To uninstall LVM Storage that you installed using RHACM, you must delete the RHACM Policy custom resource (CR) that you created for installing and configuring LVM Storage.

Prerequisites

  • You have access to the RHACM cluster as a user with cluster-admin permissions.

  • You have deleted the persistent volume claims (PVCs), volume snapshots, and volume clones provisioned by LVM Storage. You have also deleted the applications that are using these resources.

  • You have deleted the LVMCluster CR that you created using RHACM.

Procedure

  1. Log in to the OpenShift CLI (oc).

  2. Delete the RHACM Policy CR that you created for installing and configuring LVM Storage by using the following command:

    $ oc delete -f <policy> -n <namespace> (1)
    1 Replace <policy> with the name of the Policy CR YAML file.

  3. Create a Policy CR YAML file with the configuration to uninstall LVM Storage:

    Example Policy CR to uninstall LVM Storage
    apiVersion: apps.open-cluster-management.io/v1
kind: PlacementRule
metadata:
  name: placement-uninstall-lvms
spec:
  clusterConditions:
  - status: "True"
    type: ManagedClusterConditionAvailable
  clusterSelector:
    matchExpressions:
    - key: mykey
      operator: In
      values:
      - myvalue
---
apiVersion: policy.open-cluster-management.io/v1
kind: PlacementBinding
metadata:
  name: binding-uninstall-lvms
placementRef:
  apiGroup: apps.open-cluster-management.io
  kind: PlacementRule
  name: placement-uninstall-lvms
subjects:
- apiGroup: policy.open-cluster-management.io
  kind: Policy
  name: uninstall-lvms
---
apiVersion: policy.open-cluster-management.io/v1
kind: Policy
metadata:
  annotations:
    policy.open-cluster-management.io/categories: CM Configuration Management
    policy.open-cluster-management.io/controls: CM-2 Baseline Configuration
    policy.open-cluster-management.io/standards: NIST SP 800-53
  name: uninstall-lvms
spec:
  disabled: false
  policy-templates:
  - objectDefinition:
      apiVersion: policy.open-cluster-management.io/v1
      kind: ConfigurationPolicy
      metadata:
        name: uninstall-lvms
      spec:
        object-templates:
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: v1
            kind: Namespace
            metadata:
              name: openshift-storage
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: operators.coreos.com/v1
            kind: OperatorGroup
            metadata:
              name: openshift-storage-operatorgroup
              namespace: openshift-storage
            spec:
              targetNamespaces:
              - openshift-storage
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: operators.coreos.com/v1alpha1
            kind: Subscription
            metadata:
              name: lvms-operator
              namespace: openshift-storage
        remediationAction: enforce
        severity: low
  - objectDefinition:
      apiVersion: policy.open-cluster-management.io/v1
      kind: ConfigurationPolicy
      metadata:
        name: policy-remove-lvms-crds
      spec:
        object-templates:
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: apiextensions.k8s.io/v1
            kind: CustomResourceDefinition
            metadata:
              name: logicalvolumes.topolvm.io
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: apiextensions.k8s.io/v1
            kind: CustomResourceDefinition
            metadata:
              name: lvmclusters.lvm.topolvm.io
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: apiextensions.k8s.io/v1
            kind: CustomResourceDefinition
            metadata:
              name: lvmvolumegroupnodestatuses.lvm.topolvm.io
        - complianceType: mustnothave
          objectDefinition:
            apiVersion: apiextensions.k8s.io/v1
            kind: CustomResourceDefinition
            metadata:
              name: lvmvolumegroups.lvm.topolvm.io
        remediationAction: enforce
        severity: high

  4. Create the Policy CR by running the following command:

    $ oc create -f <policy> -ns <namespace>

Downloading log files and diagnostic information using must-gather

When LVM Storage is unable to automatically resolve a problem, use the must-gather tool to collect the log files and diagnostic information so that you or the Red Hat Support can review the problem and determine a solution.

Procedure

  • Run the must-gather command from the client connected to the LVM Storage cluster:

    $ oc adm must-gather --image=registry.redhat.io/lvms4/lvms-must-gather-rhel9:v4.15 --dest-dir=<directory_name>
Additional resources

Troubleshooting persistent storage

While configuring persistent storage using Logical Volume Manager (LVM) Storage, you can encounter several issues that require troubleshooting.

Investigating a PVC stuck in the Pending state

A persistent volume claim (PVC) can get stuck in the Pending state for the following reasons:

  • Insufficient computing resources.

  • Network problems.

  • Mismatched storage class or node selector.

  • No available persistent volumes (PVs).

  • The node with the PV is in the Not Ready state.

Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to the OpenShift CLI (oc) as a user with cluster-admin permissions.

Procedure

  1. Retrieve the list of PVCs by running the following command:

    $ oc get pvc
    Example output
    NAME        STATUS    VOLUME   CAPACITY   ACCESS MODES   STORAGECLASS   AGE
lvms-test   Pending                                      lvms-vg1       11s

  2. Inspect the events associated with a PVC stuck in the Pending state by running the following command:

    $ oc describe pvc <pvc_name> (1)
    1 Replace <pvc_name> with the name of the PVC. For example, lvms-vg1.
    Example output
    Type     Reason              Age               From                         Message
----     ------              ----              ----                         -------
Warning  ProvisioningFailed  4s (x2 over 17s)  persistentvolume-controller  storageclass.storage.k8s.io "lvms-vg1" not found

Recovering from a missing storage class

If you encounter the storage class not found error, check the LVMCluster custom resource (CR) and ensure that all the Logical Volume Manager (LVM) Storage pods are in the Running state.

Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to the OpenShift CLI (oc) as a user with cluster-admin permissions.

Procedure

  1. Verify that the LVMCluster CR is present by running the following command:

    $ oc get lvmcluster -n openshift-storage
    Example output
    NAME            AGE
my-lvmcluster   65m

  2. If the LVMCluster CR is not present, create an LVMCluster CR. For more information, see "Ways to create an LVMCluster custom resource".

  3. In the openshift-storage namespace, check that all the LVM Storage pods are in the Running state by running the following command:

    $ oc get pods -n openshift-storage
    Example output
    NAME                                  READY   STATUS    RESTARTS      AGE
lvms-operator-7b9fb858cb-6nsml        3/3     Running   0             70m
topolvm-controller-5dd9cf78b5-7wwr2   5/5     Running   0             66m
topolvm-node-dr26h                    4/4     Running   0             66m
vg-manager-r6zdv                      1/1     Running   0             66m

    The output of this command must contain a running instance of the following pods:

    • lvms-operator

    • vg-manager

    • topolvm-controller

    • topolvm-node

      If the topolvm-node pod is stuck in the Init state, it is due to a failure to locate an available disk for LVM Storage to use. To retrieve the necessary information to troubleshoot this issue, review the logs of the vg-manager pod by running the following command:

      $ oc logs -l app.kubernetes.io/component=vg-manager -n openshift-storage
Additional resources

Recovering from node failure

A persistent volume claim (PVC) can be stuck in the Pending state due to a node failure in the cluster.

To identify the failed node, you can examine the restart count of the topolvm-node pod. An increased restart count indicates potential problems with the underlying node, which might require further investigation and troubleshooting.

Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to the OpenShift CLI (oc) as a user with cluster-admin permissions.

Procedure

  • Examine the restart count of the topolvm-node pod instances by running the following command:

    $ oc get pods -n openshift-storage
    Example output
    NAME                                  READY   STATUS    RESTARTS      AGE
lvms-operator-7b9fb858cb-6nsml        3/3     Running   0             70m
topolvm-controller-5dd9cf78b5-7wwr2   5/5     Running   0             66m
topolvm-node-dr26h                    4/4     Running   0             66m
topolvm-node-54as8                    4/4     Running   0             66m
topolvm-node-78fft                    4/4     Running   17 (8s ago)   66m
vg-manager-r6zdv                      1/1     Running   0             66m
vg-manager-990ut                      1/1     Running   0             66m
vg-manager-an118                      1/1     Running   0             66m
Next steps

  • If the PVC is stuck in the Pending state even after you have resolved any issues with the node, you must perform a forced clean-up. For more information, see "Performing a forced clean-up".

Additional resources

Recovering from disk failure

If you see a failure message while inspecting the events associated with the persistent volume claim (PVC), there can be a problem with the underlying volume or disk.

Disk and volume provisioning issues result with a generic error message such as Failed to provision volume with storage class <storage_class_name>. The generic error message is followed by a specific volume failure error message.

The following table describes the volume failure error messages:

Table 5. Volume failure error messages
Error message Description

Failed to check volume existence

Indicates a problem in verifying whether the volume already exists. Volume verification failure can be caused by network connectivity problems or other failures.

Failed to bind volume

Failure to bind a volume can happen if the persistent volume (PV) that is available does not match the requirements of the PVC.

FailedMount or FailedAttachVolume

This error indicates problems when trying to mount the volume to a node. If the disk has failed, this error can appear when a pod tries to use the PVC.

FailedUnMount

This error indicates problems when trying to unmount a volume from a node. If the disk has failed, this error can appear when a pod tries to use the PVC.

Volume is already exclusively attached to one node and cannot be attached to another

This error can appear with storage solutions that do not support ReadWriteMany access modes.
Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to the OpenShift CLI (oc) as a user with cluster-admin permissions.

Procedure

  1. Inspect the events associated with a PVC by running the following command:

    $ oc describe pvc <pvc_name> (1)
    1 Replace <pvc_name> with the name of the PVC.

  2. Establish a direct connection to the host where the problem is occurring.

  3. Resolve the disk issue.

Next steps

  • If the volume failure messages persist or recur even after you have resolved the issue with the disk, you must perform a forced clean-up. For more information, see "Performing a forced clean-up".

Additional resources

Performing a forced clean-up

If the disk or node-related problems persist even after you have completed the troubleshooting procedures, you must perform a forced clean-up. A forced clean-up is used to address persistent issues and ensure the proper functioning of Logical Volume Manager (LVM) Storage.

Prerequisites

  • You have installed the OpenShift CLI (oc).

  • You have logged in to the OpenShift CLI (oc) as a user with cluster-admin permissions.

  • You have deleted all the persistent volume claims (PVCs) that were created by using LVM Storage.

  • You have stopped the pods that are using the PVCs that were created by using LVM Storage.

Procedure

  1. Switch to the openshift-storage namespace by running the following command:

    $ oc project openshift-storage

  2. Check if the LogicalVolume custom resources (CRs) are present by running the following command:

    $ oc get logicalvolume

    1. If the LogicalVolume CRs are present, delete them by running the following command:

      $ oc delete logicalvolume <name> (1)
      1 Replace <name> with the name of the LogicalVolume CR.

    2. After deleting the LogicalVolume CRs, remove their finalizers by running the following command:

      $ oc patch logicalvolume <name> -p '{"metadata":{"finalizers":[]}}' --type=merge (1)
      1 Replace <name> with the name of the LogicalVolume CR.

  3. Check if the LVMVolumeGroup CRs are present by running the following command:

    $ oc get lvmvolumegroup

    1. If the LVMVolumeGroup CRs are present, delete them by running the following command:

      $ oc delete lvmvolumegroup <name> (1)
      1 Replace <name> with the name of the LVMVolumeGroup CR.

    2. After deleting the LVMVolumeGroup CRs, remove their finalizers by running the following command:

      $ oc patch lvmvolumegroup <name> -p '{"metadata":{"finalizers":[]}}' --type=merge (1)
      1 Replace <name> with the name of the LVMVolumeGroup CR.

  4. Delete any LVMVolumeGroupNodeStatus CRs by running the following command:

    $ oc delete lvmvolumegroupnodestatus --all

  5. Delete the LVMCluster CR by running the following command:

    $ oc delete lvmcluster --all

    1. After deleting the LVMCluster CR, remove its finalizer by running the following command:

      $ oc patch lvmcluster <name> -p '{"metadata":{"finalizers":[]}}' --type=merge (1)
      1 Replace <name> with the name of the LVMCluster CR.