The csi-netapp module: configuration

How much of a volume is reserved for the superuser?

Nothing, by default: the module tells Trident to format with -m0, so the whole volume is available to the workload.

Where the classic 5% ext4 reserve is wanted — for example to keep a filesystem writable for a privileged process after a workload has filled it — annotate the NetappStorageConnection with the percentage to reserve:

d8 k annotate netappstorageconnection <name> storage.deckhouse.io/ext4-reserved-percent=5

Note the resource: unlike the other storage modules, the percentage is set per connection, not per StorageClass. Trident keeps the mkfs options on the backend, so the value applies to every StorageClass served by that connection.

The value is a whole number of percent between 0 and 50; an invalid one leaves the NetappStorageConnection with Ready=False and the reason in its status, instead of breaking volume creation later.

The reserve applies only to volumes created after the annotation was set — filesystems that already exist keep the reserve they were created with.

What happens if a node’s repositories cannot provide open-iscsi and multipath-tools?

The module installs them itself, from an image of its own.

On every node it serves — those carrying storage.deckhouse.io/csi-netapp-node — the NodeGroupConfiguration first asks the node’s package manager for open-iscsi and multipath-tools (iscsi-initiator-utils and device-mapper-multipath on RPM distributions). That stays the preferred path: a host’s own client and daemon are what the rest of that host is built around, and nothing about them is the module’s business while they are there.

When the install fails — a closed environment with no route to the repositories, or a distribution that does not ship the packages — the module falls back to its own package image, iscsi-tools. bashible pulls it from the module’s registry, unpacks it on the node and runs its install script. The payload lands under /var/lib/deckhouse/sds/csi-netapp, every binary is started through its own dynamic loader with its own library path, and two units come up for the daemons:

systemctl is-active d8-csi-netapp-iscsid.service d8-csi-netapp-multipathd.service

Everything the NodeGroupConfiguration does after the install runs on such a node exactly as it does on any other: the find_multipaths no override, the NetApp ONTAP device block, the kernel modules, and the storage.deckhouse.io/csi-netapp-node-ready label the CSI node waits for.

Two things the install script arranges that are worth knowing about, because both are invisible until something does not mount.

Trident reaches iscsiadm, multipath and multipathd on the host by name, through nsenter, so each name is resolved against the container’s PATH inside the host’s mount namespace. The script leaves a wrapper for all three in /usr/local/sbin, and each wrapper prefers a distribution binary the moment one appears.

And multipathd from the package reads its configuration under the prefix it was built with, not under /etc. The script links multipath.conf and multipath from that prefix onto the node’s own. Without that link Trident’s own pre-flight checks see find_multipaths: yes and refuse to stage a volume at all.

The two paths never mix. A host that already has iscsiadm of its own keeps everything of its own, and the install script does nothing there: the client and the daemon must always come from the same source, because iscsiadm of one version does not speak to iscsid of another.

To tell which path a node took, look at what is running on it:

# the distribution's stack
systemctl is-active iscsid multipathd
# the module's own
ls /var/lib/deckhouse/sds/csi-netapp/bin
multipathd show config | grep find_multipaths

Two things the fallback deliberately does not change. It never merges anything into the host’s /lib64 or /usr/lib — the payload is self-contained. And it does not touch /etc/iscsi/initiatorname.iscsi when the node already has one: the IQN is the node’s identity on the array, registered there in a host object, and a node that comes back under a different name is a node the array has never heard of.

Requirements

To the Deckhouse version: 1.72 and above.

Parameters

Schema version: 1

• settings

  object
  • settings.logLevel

    string

    The verbosity level of the module’s log output.

    Default: INFO

    Allowed values: ERROR, WARN, INFO, DEBUG, TRACE

  • settings.nodeSelector

    object

    Specifies where shares can be mounted. Works the same way as the spec.nodeSelector parameter in Kubernetes Pods. If this parameter is omitted, NetApp shares can be mounted on any node in the cluster running the Linux OS.

    Default: { "kubernetes.io/os": "linux" }

  • settings.storageClassLabelIgnoredPrefixes

    array of strings

    List of label-key prefixes that MUST NOT be propagated from a NetappStorageClass to the managed Kubernetes StorageClass.

    The controller treats this list as a complement to a built-in (system) list and drops any NetappStorageClass label whose key starts with any prefix from either list.

    The defaults cover labels typically added by GitOps tooling (Argo CD, Flux, Rancher Fleet) so that their reconcilers do not fight the storage controller over labels on the managed StorageClass.

    Default: [ "argocd.argoproj.io/", "kustomize.toolkit.fluxcd.io/", "helm.toolkit.fluxcd.io/", "fleet.cattle.op/" ]