Parameter "disklessOnRemaining" not honored

[bmalynovytch]

What happens:

Volume created with disklessOnRemaing set to true is created with 2 data nodes as expected but only 1 diskless node, which isn't suffisent for quorum in a 10 nodes cluster.

Version:

linstor-csi-plugin      quay.io/piraeusdatastore/piraeus-csi:v0.20.0

Storage pools:

╭────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╮
┊ StoragePool          ┊ Node               ┊ Driver   ┊ PoolName     ┊ FreeCapacity ┊ TotalCapacity ┊ CanSnapshots ┊ State ┊ SharedName ┊
╞════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════════╡
┊ DfltDisklessStorPool ┊ k8s-master         ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-1       ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-11      ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-12      ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-13      ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-3       ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-4       ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-5       ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-6       ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ DfltDisklessStorPool ┊ k8s-worker-7       ┊ DISKLESS ┊              ┊              ┊               ┊ False        ┊ Ok    ┊            ┊
┊ piraeus-data         ┊ k8s-worker-1       ┊ LVM      ┊ piraeus-data ┊   778.84 GiB ┊      1.10 TiB ┊ False        ┊ Ok    ┊            ┊
┊ piraeus-data         ┊ k8s-worker-11      ┊ LVM      ┊ piraeus-data ┊     2.30 TiB ┊      3.49 TiB ┊ False        ┊ Ok    ┊            ┊
┊ piraeus-data         ┊ k8s-worker-12      ┊ LVM      ┊ piraeus-data ┊     2.30 TiB ┊      3.49 TiB ┊ False        ┊ Ok    ┊            ┊
┊ piraeus-data         ┊ k8s-worker-13      ┊ LVM      ┊ piraeus-data ┊     2.49 TiB ┊      3.49 TiB ┊ False        ┊ Ok    ┊            ┊
┊ piraeus-data         ┊ k8s-worker-3       ┊ LVM      ┊ piraeus-data ┊   778.84 GiB ┊      1.10 TiB ┊ False        ┊ Ok    ┊            ┊
╰────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯

StorageClass:

allowVolumeExpansion: true
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: piraeus-piraeus-data
parameters:
  autoPlace: "2"
  csi.storage.k8s.io/fstype: ext4
  disklessOnRemaining: "true"
  linstor.csi.linbit.com/allowRemoteVolumeAccess: "true"
  linstor.csi.linbit.com/placementCount: "2"
  property.linstor.csi.linbit.com/DrbdOptions/Net/rr-conflict: retry-connect
  property.linstor.csi.linbit.com/DrbdOptions/Resource/on-no-data-accessible: suspend-io
  property.linstor.csi.linbit.com/DrbdOptions/Resource/on-suspended-primary-outdated: force-secondary
  property.linstor.csi.linbit.com/DrbdOptions/auto-quorum: suspend-io
  storagePool: piraeus-data
provisioner: linstor.csi.linbit.com
reclaimPolicy: Retain
volumeBindingMode: Immediate

This issue is related to piraeusdatastore/piraeus-ha-controller#23 which makes HA-controller think that nodes are in trouble while they aren't, and make them unschedulable in K8s.

The trick was to manually force quorum of the volume to 2 (number of data nodes), which is not what is expected.