-
Notifications
You must be signed in to change notification settings - Fork 65
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
2i2c-aws-us: ncar-cisl added #2584
Conversation
consideRatio
commented
May 29, 2023
•
edited
Loading
edited
- for [Request deployment] New Hub: NCAR-CISL for UCAR #2523
Merging this PR will trigger the following deployment actions. Support and Staging deployments
Production deployments
|
tls: | ||
- hosts: [ncar-cisl.2i2c.cloud] | ||
secretName: https-auto-tls | ||
proxy: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is actually not required, as we do not use proxy
anymore for HTTPS. The proxy
stanza can be removed
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hmmmm, I recall it was required for some reason still, but I recall thinking the same thing.
If it isn't, it should be addressed for all our hubs, because its consistently setup for all hubs after #2349.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah, I figure it may be because of this, which is something we probably can relax:
infrastructure/helm-charts/basehub/values.schema.yaml
Lines 258 to 268 in 3a6b47f
proxy: | |
type: object | |
additionalProperties: true | |
required: | |
- https | |
properties: | |
hosts: | |
type: array | |
minItems: 1 | |
items: | |
type: string |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Aaah yes we should! Not sure why that's listed as required. Am ok if you wanna do that as a separate PR, or open an issue and deal with it later as well. Not required to block this PR
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One extreneous config stanza but otherwise lgtm
@yuvipanda I updated the PR title and PR description to reflect the PR better as it was a bit more than just add the hub for ncar-cisl. Do you think the decision to remove the image choice for researchdelight in the GPU section was acceptable as well? I didn't think it was important enough to put the profileList config in individual hubs, but at the same time didn't want image choices to be shared among hubs in a shared cluster for the gpu option. |
@consideRatio I do think whenever we offer GPUs, we should offer those two choices. The default image doesn't have any GPU stuff, and these two images are different enough that I think we should do this each time. |
It probably makes sense to put these in the hub specific config though. |
1c1a2f2
to
e35dcd1
Compare
Done in 710bc23!
Okay lets do it (4aed868, e35dcd1), but I feel uneasy about this because I see how we provide blurry lines about image management responsibility by doing this. I'll open an issue about discussing this in a meeting, which is my preference on how to discuss this topic. To quickly clarity why without entering a discussion about it here, consider the note I wrote in the config for singleuser.image below. image:
# image choice preliminary and is expected to be setup via
# https://ncar-cisl.2i2c.cloud/services/configurator/ by the community
#
# pangeo/pangeo-notebook is maintained at: https://github.com/pangeo-data/pangeo-docker-images
name: pangeo/pangeo-notebook
tag: "2023.05.18" The gist is that I don't think we have a clear and simple enough policy for this image management if we specify a kubespawner_override for the GPU profile's image. By specifying it, we discard the community's ability to self-manage the GPU profile's image via the default image in the configurator, and perhaps also make the community adopt expectations that it needs to be a dedicated image. |
@consideRatio yes i agree with your concerns re: management, hence opened 2i2c-org/features#26. However, if we don't provide them the option to use the GPU image, they can not use any of the GPU features. I appreciate you sticking to the status quo here so users can benefit despite your uneasiness. |
🎉🎉🎉🎉 Monitor the deployment of the hubs here 👉 https://github.com/2i2c-org/infrastructure/actions/runs/5110517276 |
I read your excellent writeup in 2i2c-org/features#26 and I figure we'll skip a discussion about handling this interim situation where this is a problem to save our time and focus on the long term things instead. |