I have longhorn running as my default storage class for multiple clusters. The original k3s StorageClass of

local-path

was patched to unset it as a "default", but I have noticed that over time the

local-path

comes back as a default storage class. I've unset it several times now but I'd like it to stop reverting back to that state. Does anyone know how I can accomplish this or have you seen this yourself?

start k3s with --disable=local-storage. If you don’t do that, the local-storage manifest will get re-applied every time you restart a server.

where do I put that for a systemd service? In the service file or is there another/better place

yes, the way I just said

Are the manifests files located

/var/lib/rancher/k3s/server/manifests/

included in ETCD snapshots? I'm not very familiar with ETCD and how it works. However you helped answer a question I had yesterday related to taking backups. I'm wondering if I update

local-storage.yaml

in the manifests directory if that will be included in my snapshots? Otherwise I'll need to update my backup script to include

/etc/systemd/system/k3s.service

,

/etc/rancher/k3s/config.yaml

, or

/var/lib/rancher/k3s/server/manifests/local-storage.yaml

, depending on what I change to disable the local-storage.

All of the packaged manifests in /var/lib/rancher/k3s/server/manifests come with the product, and are extracted out on startup and applied to the cluster, unless you --disable them. There’s no point in backing them up since they come back on their own. If you add any of your own manifests to that directory then you would probably want to back those up as well, although ideally they would be managed along with the rest of your system configuration. Their contents are also applied (as in

kubectl apply

) to the cluster though, so they would (in a sense) be covered by the etcd snapshots.