Skip to content

fix: port EKS plugin cache fixes to v2 branch#116

Merged
Uchinda Padmaperuma (uchinda-sph) merged 1 commit intov2from
fix/eks-cache-v2
Apr 7, 2026
Merged

fix: port EKS plugin cache fixes to v2 branch#116
Uchinda Padmaperuma (uchinda-sph) merged 1 commit intov2from
fix/eks-cache-v2

Conversation

@uchinda-sph
Copy link
Copy Markdown
Contributor

This PR adds the TF_PLUGIN_CACHE_DIR: "" configuration to the precommit and tflint steps in .github/workflows/terraform.yaml. These changes were previously implemented in the eks branch to address module cache issues and are now being ported to the v2 branch to ensure consistency and reliable workflow execution.

Changes:

  • Added TF_PLUGIN_CACHE_DIR: "" to all precommit and tflint steps in .github/workflows/terraform.yaml.
  • Applied these changes across both standard and dual-stack runner jobs.

@uchinda-sph Uchinda Padmaperuma (uchinda-sph) merged commit bcc001b into v2 Apr 7, 2026
2 of 3 checks passed
@uchinda-sph Uchinda Padmaperuma (uchinda-sph) deleted the fix/eks-cache-v2 branch April 7, 2026 02:08
@niroz89
Copy link
Copy Markdown
Contributor

Uchinda Padmaperuma (@uchinda-sph) can share more on why TF_PLUGIN_CACHE_DIR: "" was necessary ? Is this change causing the issue here https://sphmediatrust.slack.com/archives/C0466PJNWH2/p1775571328239739

From the k8s logs, the node goes into disk pressure, likely due to mulitple tf init (tflint step, tf validate step etc) and provider downloads per each module. They are running pre_commit for all files (not only changed files) and likely provider gets downloaded for all modules.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants