https://rancher.com/ logo
Join Slack
Powered by
This message was deleted.
# longhorn-storage
a
This message was deleted.
b
When you say node is down, what do you mean?
Like it's offline? You've got the disk mounted somewhere else and are editing it?
Or the cluster is down?
You can manage the replicas "manually" through the UI and delete them from specific nodes, if that's what you're aiming to do. Most of the time longhorn will choose another node to balance it out, if there's enough of them available. You'd probably want to put it in maint mode? You shouldn't really have to do this though.
f
Sorry @curved-piano-98970, please provide more context when you get the chance. Like @bland-article-62755, I am a bit confused what you are trying to do (and why).
c
The longhorn node was down, from the ui in the node pages you saw "down"
Therefore any delete input provided via UI wouldn't work and i wanted to free up the corresponding volume replica in the underlying node
I had to solve by expanding the underlying VM storage and then restarting the pod related to the node manager and it was correctly scheduled on the node and it brought the nodes backup
But now that my nodes are back online they still have the deleted volumes somewhere in my VM because the disk usage doesn't match. and I can't find the corresponding file to delete to that PV replica.
@faint-sunset-36608 @bland-article-62755
f
The replica data is usually stored at 
/var/lib/longhorn/replicas
on the node. From your first comment, it sounds like there is nothing of significant size there.
But now that my nodes are back online they still have the deleted volumes somewhere in my VM because the disk usage doesn't match. and I can't find the corresponding file to delete to that PV replica.
I don't know what you mean by this. Maybe you have orphaned replicas? https://longhorn.io/docs/1.6.0/advanced-resources/data-cleanup/orphaned-data-cleanup/#orphaned-replica-directories
3 Views
Open in Slack
PreviousNext
Powered by