Cluster (re-)creation documentation perhaps a little disjointed when it comes to creating node pools?

[gmerritt]

In Step 1. of "Create a new cluster," this line links to our Kubernetes Cluster config page:

datahub/docs/admins/howto/clusterswitch.md

Line 9 in a6008bb

https://docs.datahub.berkeley.edu/en/latest/admins/cluster-config.html

On that linked page is included gcloud container node-pools create for creating node pools:

datahub/docs/admins/cluster-config.rst

Line 47 in a6008bb

gcloud container node-pools create \

Coming back to the present page, one then (again?) encounters node pool creation in Step 3. of "Create a new cluster":

datahub/docs/admins/howto/clusterswitch.md

Line 14 in a6008bb

3. Re-create all existing node pools for hubs, support and prometheus deployments in the new cluster. If the old cluster is still up and running, you will probably run out of CPU quota, as the new node pools will immediately default to three nodes. Wait ~15m for the new pools to wind down to zero, and then continue.

After setting kubectl context, the reader then (a third time?) encounters...node pool creation, this time under "Recreate node pools" --

datahub/docs/admins/howto/clusterswitch.md

Line 21 in a6008bb

## Recreate node pools

I think I get it overall, but I feel like I'm "reconstructing" the flow rather than following a smooth path. Or am I missing some aspect of this how-to?

Thanks!