Improvements to Cloud SQL & cluster creation #328
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
imreddy13
commented
Mar 11, 2024
•
imreddy13
commented
- Pass already constructed cloud SQL instance connection name to Ray workers so users don't have to set it in the notebook
- Set Cloud SQL region in frontend
- Fix bug with cluster creation where random zones without GPU availability were being selected
- Reduce SA name length (gets truncated when IM deployment prefix is added for Marketplace)
blackzlq
reviewed
Mar 11, 2024
blackzlq
reviewed
Mar 11, 2024
bjornsen
approved these changes
Mar 11, 2024
imreddy13
force-pushed
the
fix-instance-name
branch
from
1fabd8e
to
bc016a0
Compare
imreddy13
changed the title
imreddy13
force-pushed
the
fix-instance-name
branch
from
bc016a0
to
3d56c3e
Compare
|
/gcbrun |
imreddy13
force-pushed
the
fix-instance-name
branch
from
3d56c3e
to
00faf09
Compare
|
/gcbrun |
andrewsykim
approved these changes
Mar 13, 2024
infrastructure/main.tf
Outdated
| zone = length(split("-", var.cluster_location)) > 2 ? split(",", var.cluster_location) : [] | ||
| # zone needs to be set even for regional clusters, otherwise this module picks random zones that don't have GPU availability: | ||
| # https://github.com/terraform-google-modules/terraform-google-kubernetes-engine/blob/af354afdf13b336014cefbfe8f848e52c17d4415/main.tf#L46 | ||
| zone = length(split("-", var.cluster_location)) > 2 ? split(",", var.cluster_location) : split(",", local.gpu_l4_t4_location[local.region]) |
There was a problem hiding this comment.
Wow, good catch. It's kind of unexpected that it does this
imreddy13
force-pushed
the
fix-instance-name
branch
from
00faf09
to
227cdfc
Compare
|
bypassing pr-review-trigger after updating 2 files with tf fmt passed on previous iteration: https://github.com/GoogleCloudPlatform/ai-on-gke/pull/328/checks?check_run_id=22591229809 |
blackzlq
reviewed
Mar 13, 2024
blackzlq
reviewed
Mar 13, 2024
| @@ -62,7 +62,9 @@ locals { | |||
| subnetwork_name = var.create_network ? module.custom-network[0].subnets_names[0] : var.subnetwork_name | |||
| region = length(split("-", var.cluster_location)) == 2 ? var.cluster_location : "" | |||
| regional = local.region != "" ? true : false | |||
There was a problem hiding this comment.
In this case regional should be false?
There was a problem hiding this comment.
this is so complicated.
I think region = var.cluster_location if it's a regional cluster.
so local.region != "" is true for regional clusters. so this is correct.
@umeshkumhar this naming is super confusing
There was a problem hiding this comment.
Just to clarify.. we take cluster_location input only.
Then we check if user provided region or zone by splitting on hypen. Once done, we check if region is set that means cluster will be regional, so make it true, else false.
blackzlq
reviewed
Mar 13, 2024
blackzlq
reviewed
Mar 13, 2024
blackzlq
reviewed
Mar 13, 2024
1) Pass already constructed cloud SQL instance connection name to Ray workers so users don't have to set it in the notebook 2) Set Cloud SQL region in frontend 3) Fix bug with cluster creation where random zones without GPU availability were being selected 4) Reduce SA name length (gets truncated when IM deployment prefix is added for Marketplace)
imreddy13
force-pushed
the
fix-instance-name
branch
from
227cdfc
to
b556c85
Compare
annapendleton
pushed a commit
to annapendleton/ai-on-gke
that referenced
this pull request
Mar 26, 2024
Improvements to Cloud SQL & cluster creation: 1) Pass already constructed cloud SQL instance connection name to Ray workers so users don't have to set it in the notebook 2) Set Cloud SQL region in frontend 3) Fix bug with cluster creation where random zones without GPU availability were being selected 4) Reduce SA name length (gets truncated when IM deployment prefix is added for Marketplace)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.