staging -> prod (#477, #478, #480, #481, #482, #484, #486, #493, #494)

[rabernat]

fixes pangeo-data/pangeo-datastore#74
fixes pangeo-data/pangeo-datastore#73
fixes pangeo-data/pangeo#754

[jhamman]

It looks like I need to add the gateway secrets for the prod deployments. I'll try to do that now.

[jhamman]

@tjcrone - I added the dask-gateway secrets for the prod clusters on the google hubs. Can you make sure the ooi hub has the correct secrets? I don't have the git-crypt key for the ooi deployment.

[rabernat]

Anything I can do to anything I can do to help here?

Our entire data catalog is broken until we can deploy this.

[jhamman]

@rabernat - We need to hear from @tjcrone to make sure the OOI cluster can handle this merge. We also need to sort out why the builds are timing out.

[rabernat]

We also need to sort out why the builds are timing out.

To get around this for ocean, I added this:

pangeo-cloud-federation/.circleci/config.yml

Lines 55 to 58 in b52b559

	- run:
	name: Test building ocean.pangeo.io image if needed
	when: always
	no_output_timeout: 60m

The bigger question is, why does it now take 35 minutes to build these images? 🤦‍♂

[tjcrone]

I'm watching this and prepared to help. We are in a good spot in terms of a merge that might break our deployment. Shouldn't be a problem this week. Thank you very much for checking. Please go ahead and merge if you can, and if something goes wrong with our deployment, I will fix it.

[tjcrone]

I have updated our secrets on our fork, but I'm not sure how to add these to this PR. Let's go ahead and merge this, and then I will do a separate PR to merge our secrets files. Does this sound reasonable? We are fine on our side for any merge you decide to do here. Thank you again!