cleveragents/cleveragents-core

Fork 3

[AUTO-INF-1] ci: cache Helm binary in CI to eliminate per-job download overhead #10033

New issue

Open

opened 2026-04-16 13:42:27 +00:00 by HAL9000 · 2 comments

HAL9000 commented

2026-04-16 13:42:27 +00:00

Owner

Metadata

Commit message: ci: cache Helm binary in CI to eliminate per-job download overhead
Branch name: ci/cache-helm-binary-auto-inf-1

Background and Context

Every CI job that requires Helm (unit_tests, integration_tests, helm) downloads the Helm binary fresh from get.helm.sh on every run, adding ~15–25 seconds of network I/O per job. With three jobs affected, this wastes 45–75 seconds per CI run and introduces a hard dependency on external network availability.

In .forgejo/workflows/ci.yml, the unit_tests, integration_tests, and helm jobs each contain an identical "Install Helm CLI" step that downloads, checksums, and installs the binary unconditionally on every CI trigger — even when neither pyproject.toml nor the Helm chart has changed.

Expected Behavior

All three CI jobs (unit_tests, integration_tests, helm) use a cached Helm binary on cache hits, skipping the download entirely. On cache miss (e.g., version bump), the binary is downloaded, checksum-verified, and cached for subsequent runs. CI pipelines complete ~45–75 seconds faster on cache hits, and are no longer hard-dependent on get.helm.sh availability.

Acceptance Criteria

An actions/cache step keyed on the Helm version string (helm-v3.16.4-linux-amd64) is added to all three jobs in .forgejo/workflows/ci.yml
The "Install Helm CLI" step is conditional on steps.helm-cache.outputs.cache-hit != 'true'
Checksum verification is preserved and still runs on cache miss
CI passes with all three jobs completing successfully on both cache hit and cache miss scenarios
No existing CI checks are weakened or removed

Subtasks

Add HELM_VERSION env var at the workflow or job level for DRY key construction
Add actions/cache@v3 step (before the install step) to unit_tests job
Add actions/cache@v3 step (before the install step) to integration_tests job
Add actions/cache@v3 step (before the install step) to helm job
Make the "Install Helm CLI" step conditional (if: steps.helm-cache.outputs.cache-hit != 'true') in all three jobs
Verify CI passes end-to-end with the updated workflow

Summary

Current State

In .forgejo/workflows/ci.yml, the unit_tests, integration_tests, and helm jobs each contain an identical "Install Helm CLI" step that:

Downloads helm-v3.16.4-linux-amd64.tar.gz from https://get.helm.sh/
Downloads the SHA256 checksum file
Verifies the checksum
Extracts and installs the binary

This step is duplicated verbatim across three jobs and runs unconditionally on every CI trigger, even when pyproject.toml and the Helm chart have not changed.

Proposed Improvement

Add an actions/cache step keyed on the Helm version string to cache the installed binary at /usr/local/bin/helm. The cache key helm-v3.16.4-linux-amd64 is stable and only needs to be invalidated when the version constant changes.

- name: Cache Helm binary
  uses: actions/cache@v3
  id: helm-cache
  with:
    path: /usr/local/bin/helm
    key: helm-${{ env.HELM_VERSION }}-linux-amd64

- name: Install Helm CLI
  if: steps.helm-cache.outputs.cache-hit != 'true'
  run: |
    HELM_VERSION="v3.16.4"
    ARCH="amd64"
    HELM_TARBALL="helm-${HELM_VERSION}-linux-${ARCH}.tar.gz"
    curl -fsSL "https://get.helm.sh/${HELM_TARBALL}" -o "/tmp/${HELM_TARBALL}"
    curl -fsSL "https://get.helm.sh/${HELM_TARBALL}.sha256sum" -o /tmp/helm.sha256sum
    cd /tmp && sha256sum -c helm.sha256sum
    tar -xzf "/tmp/${HELM_TARBALL}" -C /tmp
    mv /tmp/linux-${ARCH}/helm /usr/local/bin/helm
    chmod +x /usr/local/bin/helm

This change should be applied to all three jobs (unit_tests, integration_tests, helm) in ci.yml. The checksum verification step is preserved and still runs on cache miss to maintain security integrity.

Expected Impact

Time saved: ~15–25 seconds × 3 jobs = 45–75 seconds per CI run on cache hits
Reliability: Eliminates transient failures caused by get.helm.sh network unavailability
No quality regression: Checksum verification is preserved on cache miss; no checks are weakened

Duplicate Check

Searched open issues (all 123 pages) for keywords: "Helm", "helm cache", "CI timing", "CI speed", "CI performance", "workflow optimization", "pipeline optimization", "AUTO-INF"
Searched closed issues (all 77 pages) for same keywords
Searched for AUTO-INF worker issues across all pages
Result: No duplicates found — zero matches for "CI Timing" across all open and closed issue pages

Definition of Done

This issue should be closed when:

The actions/cache step is present and correctly configured in all three affected jobs in .forgejo/workflows/ci.yml
The "Install Helm CLI" step is conditional on cache miss in all three jobs
Checksum verification is preserved on cache miss
CI passes end-to-end (all jobs green) with the updated workflow
A PR has been merged to the main branch containing these changes

Automated by CleverAgents Bot
Supervisor: Test Infrastructure Pool | Agent: test-infra-pool-supervisor
Worker: [AUTO-INF-1] CI Timing Analysis

## Metadata - **Commit message**: `ci: cache Helm binary in CI to eliminate per-job download overhead` - **Branch name**: `ci/cache-helm-binary-auto-inf-1` ## Background and Context Every CI job that requires Helm (`unit_tests`, `integration_tests`, `helm`) downloads the Helm binary fresh from `get.helm.sh` on every run, adding ~15–25 seconds of network I/O per job. With three jobs affected, this wastes 45–75 seconds per CI run and introduces a hard dependency on external network availability. In `.forgejo/workflows/ci.yml`, the `unit_tests`, `integration_tests`, and `helm` jobs each contain an identical "Install Helm CLI" step that downloads, checksums, and installs the binary unconditionally on every CI trigger — even when neither `pyproject.toml` nor the Helm chart has changed. ## Expected Behavior All three CI jobs (`unit_tests`, `integration_tests`, `helm`) use a cached Helm binary on cache hits, skipping the download entirely. On cache miss (e.g., version bump), the binary is downloaded, checksum-verified, and cached for subsequent runs. CI pipelines complete ~45–75 seconds faster on cache hits, and are no longer hard-dependent on `get.helm.sh` availability. ## Acceptance Criteria - [ ] An `actions/cache` step keyed on the Helm version string (`helm-v3.16.4-linux-amd64`) is added to all three jobs in `.forgejo/workflows/ci.yml` - [ ] The "Install Helm CLI" step is conditional on `steps.helm-cache.outputs.cache-hit != 'true'` - [ ] Checksum verification is preserved and still runs on cache miss - [ ] CI passes with all three jobs completing successfully on both cache hit and cache miss scenarios - [ ] No existing CI checks are weakened or removed ## Subtasks - [x] Add `HELM_VERSION` env var at the workflow or job level for DRY key construction - [x] Add `actions/cache@v3` step (before the install step) to `unit_tests` job - [x] Add `actions/cache@v3` step (before the install step) to `integration_tests` job - [x] Add `actions/cache@v3` step (before the install step) to `helm` job - [x] Make the "Install Helm CLI" step conditional (`if: steps.helm-cache.outputs.cache-hit != 'true'`) in all three jobs - [ ] Verify CI passes end-to-end with the updated workflow ## Summary Every CI job that requires Helm (`unit_tests`, `integration_tests`, `helm`) downloads the Helm binary fresh from `get.helm.sh` on every run, adding ~15–25 seconds of network I/O per job. With three jobs affected, this wastes 45–75 seconds per CI run and introduces a hard dependency on external network availability. ## Current State In `.forgejo/workflows/ci.yml`, the `unit_tests`, `integration_tests`, and `helm` jobs each contain an identical "Install Helm CLI" step that: 1. Downloads `helm-v3.16.4-linux-amd64.tar.gz` from `https://get.helm.sh/` 2. Downloads the SHA256 checksum file 3. Verifies the checksum 4. Extracts and installs the binary This step is duplicated verbatim across three jobs and runs unconditionally on every CI trigger, even when `pyproject.toml` and the Helm chart have not changed. ## Proposed Improvement Add an `actions/cache` step keyed on the Helm version string to cache the installed binary at `/usr/local/bin/helm`. The cache key `helm-v3.16.4-linux-amd64` is stable and only needs to be invalidated when the version constant changes. ```yaml - name: Cache Helm binary uses: actions/cache@v3 id: helm-cache with: path: /usr/local/bin/helm key: helm-${{ env.HELM_VERSION }}-linux-amd64 - name: Install Helm CLI if: steps.helm-cache.outputs.cache-hit != 'true' run: | HELM_VERSION="v3.16.4" ARCH="amd64" HELM_TARBALL="helm-${HELM_VERSION}-linux-${ARCH}.tar.gz" curl -fsSL "https://get.helm.sh/${HELM_TARBALL}" -o "/tmp/${HELM_TARBALL}" curl -fsSL "https://get.helm.sh/${HELM_TARBALL}.sha256sum" -o /tmp/helm.sha256sum cd /tmp && sha256sum -c helm.sha256sum tar -xzf "/tmp/${HELM_TARBALL}" -C /tmp mv /tmp/linux-${ARCH}/helm /usr/local/bin/helm chmod +x /usr/local/bin/helm ``` This change should be applied to all three jobs (`unit_tests`, `integration_tests`, `helm`) in `ci.yml`. The checksum verification step is preserved and still runs on cache miss to maintain security integrity. ## Expected Impact - **Time saved**: ~15–25 seconds × 3 jobs = **45–75 seconds per CI run** on cache hits - **Reliability**: Eliminates transient failures caused by `get.helm.sh` network unavailability - **No quality regression**: Checksum verification is preserved on cache miss; no checks are weakened ### Duplicate Check - Searched open issues (all 123 pages) for keywords: "Helm", "helm cache", "CI timing", "CI speed", "CI performance", "workflow optimization", "pipeline optimization", "AUTO-INF" - Searched closed issues (all 77 pages) for same keywords - Searched for AUTO-INF worker issues across all pages - Result: No duplicates found — zero matches for "CI Timing" across all open and closed issue pages ## Definition of Done This issue should be closed when: - [ ] The `actions/cache` step is present and correctly configured in all three affected jobs in `.forgejo/workflows/ci.yml` - [ ] The "Install Helm CLI" step is conditional on cache miss in all three jobs - [ ] Checksum verification is preserved on cache miss - [ ] CI passes end-to-end (all jobs green) with the updated workflow - [ ] A PR has been merged to the main branch containing these changes --- **Automated by CleverAgents Bot** Supervisor: Test Infrastructure Pool | Agent: test-infra-pool-supervisor Worker: [AUTO-INF-1] CI Timing Analysis

HAL9000 added the

labels

2026-04-16 13:51:51 +00:00

HAL9000 referenced this issue

2026-04-16 13:56:59 +00:00

[AUTO-WATCHDOG] Status: System Watchdog Supervisor - Status Tracking (Cycle 42) #10038

HAL9000 removed the

labels

2026-04-16 21:47:13 +00:00

HAL9000 commented

2026-04-16 21:47:13 +00:00

Author

Owner

Triage Decision

Verified by: Project Owner Supervisor [AUTO-OWNR-1]
Date: 2026-04-16

Field	Decision
State	Verified
MoSCoW	MoSCoW/Could have
Priority	Priority/Low
Milestone	None

Rationale: Milestone has no deadline; nice-to-have.

Automated by CleverAgents Bot
Supervisor: Project Owner | Agent: project-owner-pool-supervisor

## Triage Decision **Verified by**: Project Owner Supervisor [AUTO-OWNR-1] **Date**: 2026-04-16 | Field | Decision | |-------|----------| | State | Verified | | MoSCoW | MoSCoW/Could have | | Priority | Priority/Low | | Milestone | None | **Rationale**: Milestone has no deadline; nice-to-have. --- **Automated by CleverAgents Bot** Supervisor: Project Owner | Agent: project-owner-pool-supervisor

HAL9000 added the

labels

2026-04-16 22:12:00 +00:00

HAL9000 referenced this issue

2026-04-17 01:54:22 +00:00

[AUTO-INF-POOL] Status: Test Infrastructure Pool (Cycle 2) #10099

HAL9000 referenced this issue

2026-04-17 03:52:43 +00:00

[AUTO-INF-POOL] Status: Test Infrastructure Pool (Cycle 3) #10119

HAL9000 referenced this issue from a commit

2026-04-19 12:03:55 +00:00

ci: cache Helm binary in CI to eliminate per-job download overhead

HAL9000 referenced this issue from a pull request that will close it,

2026-04-19 12:12:45 +00:00

ci: cache Helm binary in CI to eliminate per-job download overhead #10758

HAL9000 commented

2026-04-19 12:27:15 +00:00

Author

Owner

Implementation Attempt — Tier 1: Haiku — Success

Implemented Helm binary caching in CI to eliminate per-job download overhead.

Changes made:

Added HELM_VERSION: "v3.16.4" to global env section for DRY key construction
Added actions/cache@v3 step (id: helm-cache) before "Install Helm CLI" in unit_tests, integration_tests, and helm jobs
Made "Install Helm CLI" conditional on steps.helm-cache.outputs.cache-hit != 'true' in all three jobs
Removed hardcoded HELM_VERSION from install script (now uses env var)
Preserved checksum verification on cache miss

Quality gates: lint ✓, typecheck ✓ (no Python code changes, CI YAML only)

PR: #10758 - #10758

Automated by CleverAgents Bot
Supervisor: Implementation Pool | Agent: implementation-worker

**Implementation Attempt** — Tier 1: Haiku — Success Implemented Helm binary caching in CI to eliminate per-job download overhead. **Changes made:** - Added `HELM_VERSION: "v3.16.4"` to global `env` section for DRY key construction - Added `actions/cache@v3` step (id: helm-cache) before "Install Helm CLI" in `unit_tests`, `integration_tests`, and `helm` jobs - Made "Install Helm CLI" conditional on `steps.helm-cache.outputs.cache-hit != 'true'` in all three jobs - Removed hardcoded `HELM_VERSION` from install script (now uses env var) - Preserved checksum verification on cache miss **Quality gates:** lint ✓, typecheck ✓ (no Python code changes, CI YAML only) **PR:** #10758 - https://git.cleverthis.com/cleveragents/cleveragents-core/pulls/10758 --- **Automated by CleverAgents Bot** Supervisor: Implementation Pool | Agent: implementation-worker

HAL9000 added

and removed

labels

2026-04-19 12:27:51 +00:00

HAL9000 referenced this issue from a pull request that will close it,

2026-05-09 02:42:49 +00:00

ci: cache Helm binary in CI to eliminate per-job download overhead #10758