Merge pull request #69065 from sr1kar99/add-new-param-to-deviceSelector

Author: sheriff-rh

modules/lvms-creating-logical-volume-manager-cluster.adoc

@@ -57,11 +57,12 @@ spec:
           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 <5>
         thinPoolConfig:
           name: thin-pool-1
           sizePercent: 90
           overprovisionRatio: 10
-        nodeSelector: <5>
+        nodeSelector: <6>
           nodeSelectorTerms:
             - matchExpressions:
               - key: app
@@ -76,7 +77,8 @@ When configuring multiple device classes, you must specify the device path for e
 <3> Mandatory: The `LVMCluster` resource must contain a single default storage class. Set `default: false` for secondary device storage classes.
 If you are updating the `LVMCluster` resource from a previous version, you must specify a single default device class.
 <4> Optional. To control or restrict the volume group to your preferred devices, you can manually specify the local paths of the devices in the `deviceSelector` section of the `LVMCluster` YAML. The `paths` section refers to devices the `LVMCluster` adds, which means those paths must exist. The `optionalPaths` section refers to devices the `LVMCluster` might add. You must specify at least one of `paths` or `optionalPaths` when specifying the `deviceSelector` section. If you specify `paths`, it is not mandatory to specify `optionalPaths`. If you specify `optionalPaths`, it is not mandatory to specify `paths` but at least one optional path must be present on the node. If you do not specify any paths, then the `LVMCluster` adds the unused devices on the node.
-<5> Optional: To control what worker nodes the `LVMCluster` CR is applied to, specify a set of node selector labels.
+<5> Optional: To force wipe the selected disks, set `forceWipeDevicesAndDestroyAllData` to true. This parameter is set to `false` by default.
+<6> Optional: To control what worker nodes the `LVMCluster` CR is applied to, specify a set of node selector labels.
 The specified labels must be present on the node in order for the `LVMCluster` to be scheduled on that node.
 
 .. Create the `LVMCluster` CR:
@@ -97,6 +99,17 @@ The `LVMCluster` resource creates the following system-managed CRs:
 `LVMVolumeGroup`:: Tracks individual volume groups across multiple nodes.
 `LVMVolumeGroupNodeStatus`:: Tracks the status of the volume groups on a node.
 
+[NOTE]
+====
+Wiping the disk can lead to inconsistencies in data integrity if any of the following conditions is met:
+
+* The disk is being used as swap space.
+* The disk is part of the Redundant Array of Independent Disks (RAID).
+* The disk is mounted.
+
+In this case, you must manually wipe the disk.
+====
+
 .Verification
 
 Verify that the `LVMCluster` resource has created the `StorageClass`, `LVMVolumeGroup`, and `LVMVolumeGroupNodeStatus` CRs.

modules/lvms-reference-file.adoc

@@ -39,36 +39,38 @@ spec:
         optionalPaths:
         - /dev/disk/by-path/pci-0000:89:00.0-nvme-1
         - /dev/disk/by-path/pci-0000:90:00.0-nvme-1
-      thinPoolConfig: <6>
-        name: thin-pool-1 <7>
-        sizePercent: 90 <8>
-        overprovisionRatio: 10 <9>
+        forceWipeDevicesAndDestroyAllData: true <6>
+      thinPoolConfig: <7>
+        name: thin-pool-1 <8>
+        sizePercent: 90 <9>
+        overprovisionRatio: 10 <10>
 status:
-    deviceClassStatuses: <10>
+    deviceClassStatuses: <11>
     - name: vg1
-      nodeStatus: <11>
-      - devices: <12>
+      nodeStatus: <12>
+      - devices: <13>
         - /dev/nvme0n1
         - /dev/nvme1n1
         - /dev/nvme2n1
-        node: my-node.example.com <13>
-        status: Ready <14>
-    ready: true <15>
-    state: Ready <16>
+        node: my-node.example.com <14>
+        status: Ready <15>
+    ready: true <16>
+    state: Ready <17>
 ----
 <1> The LVM volume groups to be created on the cluster. Currently, only a single `deviceClass` is supported.
 <2> The name of the LVM volume group to be created on the nodes.
 <3> The nodes on which to create the LVM volume group. If the field is empty, all nodes are considered.
 <4> A list of node selector requirements.
 <5> A list of device paths which is used to create the LVM volume group. If this field is empty, all unused disks on the node will be used.
-<6> The LVM thin pool configuration.
-<7> The name of the thin pool to be created in the LVM volume group.
-<8> The percentage of remaining space in the LVM volume group that should be used for creating the thin pool.
-<9> The factor by which additional storage can be provisioned compared to the available storage in the thin pool.
-<10> The status of the `deviceClass`.
-<11> The status of the LVM volume group on each node.
-<12> The list of devices used to create the LVM volume group.
-<13> The node on which the `deviceClass` was created.
-<14> The status of the LVM volume group on the node.
-<15> This field is deprecated.
-<16> The status of the `LVMCluster`.
+<6> To force wipe the selected disks, set this parameter to `true`. This parameter is set to `false` by default.
+<7> The LVM thin pool configuration.
+<8> The name of the thin pool to be created in the LVM volume group.
+<9> The percentage of remaining space in the LVM volume group that should be used for creating the thin pool.
+<10> The factor by which additional storage can be provisioned compared to the available storage in the thin pool.
+<11> The status of the `deviceClass`.
+<12> The status of the LVM volume group on each node.
+<13> The list of devices used to create the LVM volume group.
+<14> The node on which the `deviceClass` was created.
+<15> The status of the LVM volume group on the node.
+<16> This field is deprecated.
+<17> The status of the `LVMCluster`.