You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In your documentation example about rook,
1,2,3 have rook:true but 3 doesn't have data disk available. 4 does have data disk available but rook:false is set.
How would this work? Shouldn't rook be needed to run on at least 3 nodes with data disks available? Although all nodes regardless whether it is also serving storage should be able run pods with rook storage claim?
The text was updated successfully, but these errors were encountered:
In the time of writing, the idea was to show how to exclude some nodes from the Rook cluster deployment (on excluded nodes, the additional volumes should remain intact and no Rook component should be deployed on them).
In addition, it was intended to show that Kubitect detects which nodes contain valid storage volumes for a Rook cluster. If a new volume is attached to the node later it will be automatically consumed by Rook.
However, with your questions in place, I see that this guide is not clear enough and needs to be updated.
In your documentation example about rook,
1,2,3 have rook:true but 3 doesn't have data disk available. 4 does have data disk available but rook:false is set.
How would this work? Shouldn't rook be needed to run on at least 3 nodes with data disks available? Although all nodes regardless whether it is also serving storage should be able run pods with rook storage claim?
The text was updated successfully, but these errors were encountered: