Multi NIC support & cluster ID fix for Slurm controller#5563
Conversation
…tial cluster ID mismatches in controller setup
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request enhances the Slurm controller configuration by introducing support for multiple network interfaces and improving the robustness of the cluster initialization process. These changes ensure better network flexibility and prevent state-related errors during controller recreation, leading to a more stable deployment experience. Highlights
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. Footnotes
|
There was a problem hiding this comment.
Code Review
This pull request adds support for additional network interfaces to the Slurm controller and includes a fix in the setup script to prevent cluster ID mismatches by removing the clustername file. It also updates bucket-related outputs for better consistency. Feedback was provided regarding a redundant ternary operator in the slurm_bucket output, which can be simplified since the module already evaluates to an empty list when not created.
kvenkatachala333
left a comment
There was a problem hiding this comment.
Implementation wise LGTM, please ensure all slurm tests are passed before merging
LAVEEN
left a comment
There was a problem hiding this comment.
LGTM. I see a3 high test is failing. Please make sure its not due to this change.
5bcef44
into
GoogleCloudPlatform:develop
This Pull Request introduces support for configuring additional network interfaces on the Slurm controller, resolves potential cluster ID mismatches during the controller setup, and fixes Terraform output errors encountered when using an existing Google Cloud Storage (GCS) bucket
Key Changes
Support for Additional Network Interfaces: Updates the controller configuration and variable definitions (in community/modules/scheduler/schedmd-slurm-gcp-v6-controller/controller.tf and variables_controller_instance.tf) to include a dynamic block for network_interface. This allows the controller to connect to multiple networks (such as a primary HPC network and a secondary VPN network) as specified in var.additional_networks.
Cluster ID Mismatch Resolution: Modifies the initialization logic in community/modules/scheduler/schedmd-slurm-gcp-v6-controller/modules/slurm_files/scripts/setup.py to automatically remove the cached cluster ID file (/var/spool/slurm/clustername) during setup. This prevents "CLUSTER ID MISMATCH" errors that occur in multi-cluster environments when a controller is recreated while the database retains the old registration.
Bucket Output Fix: Adjusts community/modules/scheduler/schedmd-slurm-gcp-v6-controller/outputs.tf to reference module.slurm_files.bucket_name instead of module.bucket[0]. This resolves "Unsupported attribute" errors that occur when create_bucket is set to false, ensuring Terraform can successfully output the bucket name even when using pre-existing infrastructure.
Submission Checklist
NOTE: Community submissions can take up to 2 weeks to be reviewed.
Please take the following actions before submitting this pull request.