open-vault/.github/workflows/test-go.yml

370 lines
16 KiB
YAML
Raw Normal View History

on:
workflow_call:
inputs:
go-arch:
description: The execution architecture (arm, amd64, etc.)
required: true
type: string
enterprise:
description: A flag indicating if this workflow is executing for the enterprise repository.
required: true
type: string
total-runners:
description: Number of runners to use for executing the tests on.
required: true
type: string
env-vars:
description: A map of environment variables as JSON.
required: false
type: string
default: '{}'
extra-flags:
description: A space-separated list of additional build flags.
required: false
type: string
default: ''
runs-on:
description: An expression indicating which kind of runners to use.
required: false
type: string
default: ubuntu-latest
go-tags:
description: A comma-separated list of additional build tags to consider satisfied during the build.
required: false
type: string
[QT-576] Optimize build workflow (#21486) (#21601) Improve our build workflow execution time by using custom runners, improved caching and conditional Web UI builds. Runners ------- We improve our build times[0] by using larger custom runners[1] when building the UI and Vault. Caching ------- We improve Vault caching by keeping a cache for each build job. This strategy has the following properties which should result in faster build times when `go.sum` hasn't been changed from prior builds, or when a pull request is retried or updated after a prior successful build: * Builds will restore cached Go modules and Go build cache according to the Go version, platform, architecture, go tags, and hash of `go.sum` that relates to each individual build workflow. This reduces the amount of time it will take to download the cache on hits and upload the cache on misses. * Parallel build workflows won't clobber each others build cache. This results in much faster compile times after cache hits because the Go compiler can reuse the platform, architecture, and tag specific build cache that it created on prior runs. * Older modules and build cache will not be uploaded when creating a new cache. This should result in lean cache sizes on an ongoing basis. * On cache misses we will have to upload our compressed module and build cache. This will slightly extend the build time for pull requests that modify `go.sum`. Web UI ------ We no longer build the web UI in every build workflow. Instead we separate the UI building into its own workflow and cache the resulting assets. The same UI assets are restored from cache during build worklows. This strategy has the following properties: * If the `ui` directory has not changed from prior builds we'll restore `http/web_ui` from cache and skip building the UI for no reason. * We continue to use the built-in `yarn` caching functionality in `action/setup-node`. The default mode saves the `yarn` global cache. to improve UI build times if the cache has not been modified. Changes ------- * Add per platform/archicture Go module and build caching * Move UI building into a separate job and cache the result * Restore UI cache during build * Pin workflows Notes ----- [0] https://hashicorp.atlassian.net/browse/QT-578 [1] https://github.com/hashicorp/vault/actions/runs/5415830307/jobs/9844829929 Signed-off-by: Ryan Cragun <me@ryan.ec>
2023-07-05 21:19:49 +00:00
name:
description: A suffix to append to archived test results
required: false
default: ''
type: string
go-test-parallelism:
description: The parallelism parameter for Go tests
required: false
default: 20
type: number
timeout-minutes:
description: The maximum number of minutes that this workflow should run
[QT-576] Optimize build workflow (#21486) (#21601) Improve our build workflow execution time by using custom runners, improved caching and conditional Web UI builds. Runners ------- We improve our build times[0] by using larger custom runners[1] when building the UI and Vault. Caching ------- We improve Vault caching by keeping a cache for each build job. This strategy has the following properties which should result in faster build times when `go.sum` hasn't been changed from prior builds, or when a pull request is retried or updated after a prior successful build: * Builds will restore cached Go modules and Go build cache according to the Go version, platform, architecture, go tags, and hash of `go.sum` that relates to each individual build workflow. This reduces the amount of time it will take to download the cache on hits and upload the cache on misses. * Parallel build workflows won't clobber each others build cache. This results in much faster compile times after cache hits because the Go compiler can reuse the platform, architecture, and tag specific build cache that it created on prior runs. * Older modules and build cache will not be uploaded when creating a new cache. This should result in lean cache sizes on an ongoing basis. * On cache misses we will have to upload our compressed module and build cache. This will slightly extend the build time for pull requests that modify `go.sum`. Web UI ------ We no longer build the web UI in every build workflow. Instead we separate the UI building into its own workflow and cache the resulting assets. The same UI assets are restored from cache during build worklows. This strategy has the following properties: * If the `ui` directory has not changed from prior builds we'll restore `http/web_ui` from cache and skip building the UI for no reason. * We continue to use the built-in `yarn` caching functionality in `action/setup-node`. The default mode saves the `yarn` global cache. to improve UI build times if the cache has not been modified. Changes ------- * Add per platform/archicture Go module and build caching * Move UI building into a separate job and cache the result * Restore UI cache during build * Pin workflows Notes ----- [0] https://hashicorp.atlassian.net/browse/QT-578 [1] https://github.com/hashicorp/vault/actions/runs/5415830307/jobs/9844829929 Signed-off-by: Ryan Cragun <me@ryan.ec>
2023-07-05 21:19:49 +00:00
required: false
default: 60
[QT-576] Optimize build workflow (#21486) (#21601) Improve our build workflow execution time by using custom runners, improved caching and conditional Web UI builds. Runners ------- We improve our build times[0] by using larger custom runners[1] when building the UI and Vault. Caching ------- We improve Vault caching by keeping a cache for each build job. This strategy has the following properties which should result in faster build times when `go.sum` hasn't been changed from prior builds, or when a pull request is retried or updated after a prior successful build: * Builds will restore cached Go modules and Go build cache according to the Go version, platform, architecture, go tags, and hash of `go.sum` that relates to each individual build workflow. This reduces the amount of time it will take to download the cache on hits and upload the cache on misses. * Parallel build workflows won't clobber each others build cache. This results in much faster compile times after cache hits because the Go compiler can reuse the platform, architecture, and tag specific build cache that it created on prior runs. * Older modules and build cache will not be uploaded when creating a new cache. This should result in lean cache sizes on an ongoing basis. * On cache misses we will have to upload our compressed module and build cache. This will slightly extend the build time for pull requests that modify `go.sum`. Web UI ------ We no longer build the web UI in every build workflow. Instead we separate the UI building into its own workflow and cache the resulting assets. The same UI assets are restored from cache during build worklows. This strategy has the following properties: * If the `ui` directory has not changed from prior builds we'll restore `http/web_ui` from cache and skip building the UI for no reason. * We continue to use the built-in `yarn` caching functionality in `action/setup-node`. The default mode saves the `yarn` global cache. to improve UI build times if the cache has not been modified. Changes ------- * Add per platform/archicture Go module and build caching * Move UI building into a separate job and cache the result * Restore UI cache during build * Pin workflows Notes ----- [0] https://hashicorp.atlassian.net/browse/QT-578 [1] https://github.com/hashicorp/vault/actions/runs/5415830307/jobs/9844829929 Signed-off-by: Ryan Cragun <me@ryan.ec>
2023-07-05 21:19:49 +00:00
type: number
env: ${{ fromJSON(inputs.env-vars) }}
jobs:
test-matrix:
permissions:
id-token: write # Note: this permission is explicitly required for Vault auth
contents: read
runs-on: ${{ fromJSON(inputs.runs-on) }}
steps:
- uses: actions/checkout@c85c95e3d7251135ab7dc9ce3241c5835cc595a9 # v3.5.3
- uses: ./.github/actions/set-up-go
with:
github-token: ${{ secrets.ELEVATED_GITHUB_TOKEN }}
no-restore: true # We don't need the vault Go modules when generating indices
- name: Authenticate to Vault
id: vault-auth
if: github.repository == 'hashicorp/vault-enterprise'
run: vault-auth
- name: Fetch Secrets
id: secrets
if: github.repository == 'hashicorp/vault-enterprise'
uses: hashicorp/vault-action@130d1f5f4fe645bb6c83e4225c04d64cfb62de6e
with:
url: ${{ steps.vault-auth.outputs.addr }}
caCertificate: ${{ steps.vault-auth.outputs.ca_certificate }}
token: ${{ steps.vault-auth.outputs.token }}
secrets: |
kv/data/github/${{ github.repository }}/datadog-ci DATADOG_API_KEY;
kv/data/github/${{ github.repository }}/github-token username-and-token | github-token;
kv/data/github/${{ github.repository }}/license license_1 | VAULT_LICENSE_CI;
kv/data/github/${{ github.repository }}/license license_2 | VAULT_LICENSE_2;
kv/data/github/${{ github.repository }}/hcp-link HCP_API_ADDRESS;
kv/data/github/${{ github.repository }}/hcp-link HCP_AUTH_URL;
kv/data/github/${{ github.repository }}/hcp-link HCP_CLIENT_ID;
kv/data/github/${{ github.repository }}/hcp-link HCP_CLIENT_SECRET;
kv/data/github/${{ github.repository }}/hcp-link HCP_RESOURCE_ID;
- id: setup-git-private
name: Setup Git configuration (private)
if: github.repository == 'hashicorp/vault-enterprise'
run: |
git config --global url."https://${{ steps.secrets.outputs.github-token }}@github.com".insteadOf https://github.com
- id: setup-git-public
name: Setup Git configuration (public)
if: github.repository != 'hashicorp/vault-enterprise'
run: |
git config --global url."https://${{ secrets.ELEVATED_GITHUB_TOKEN}}@github.com".insteadOf https://github.com
- run: go install gotest.tools/gotestsum@v1.9.0
- run: mkdir -p test-results/go-test
# We use a unique "read-" prefix to guarantee that we're not scribbling on
# the aggregated test data in the event of test failure. This key is
# unique for every test run and just used to restore the previous
# aggregated data. We persist all test data after a successful run and
# store that in the go-test-reports- cache.
- id: restore-from-cache
uses: actions/cache@88522ab9f39a2ea568f7027eddc7d8d8bc9d59c8 # v3.3.1
with:
path: test-results/go-test
key: read-go-test-reports-${{ github.run_number }}
restore-keys: go-test-reports-
- name: List cached results
id: list-cached-results
run: ls -lhR test-results/go-test
- name: Build matrix excluding binary and integration tests
id: build-non-binary
env:
GOPRIVATE: github.com/hashicorp/*
run: |
(
go list ./... | grep -v "_binary" | grep -v "vault/integ" | gotestsum tool ci-matrix --debug \
--partitions 16 \
--timing-files 'test-results/go-test/*.json' > matrix.json
)
- name: Capture list of binary tests
id: list-binary-tests
run: |
LIST="$(go list ./... | grep "_binary" | xargs)"
echo "list=$LIST" >> "$GITHUB_OUTPUT"
- name: Build complete matrix
id: build
run: |
set -exo pipefail
export BINARY_TESTS="${{ steps.list-binary-tests.outputs.list }}"
(
echo -n "matrix="
jq -c --arg BINARY "${BINARY_TESTS}" \
'.include += [{
"id": 16,
"estimatedRuntime": "N/A",
"packages": $BINARY,
"description": "partition 16 - binary test packages"
}]' matrix.json
) >> "$GITHUB_OUTPUT"
outputs:
matrix: ${{ steps.build.outputs.matrix }}
test-go:
needs: test-matrix
permissions:
actions: read
contents: read
id-token: write # Note: this permission is explicitly required for Vault auth
runs-on: ${{ fromJSON(inputs.runs-on) }}
strategy:
fail-fast: false
matrix: ${{ fromJSON(needs.test-matrix.outputs.matrix) }}
env:
GOPRIVATE: github.com/hashicorp/*
TIMEOUT_IN_MINUTES: ${{ inputs.timeout-minutes }}
steps:
[QT-576] Optimize build workflow (#21486) (#21601) Improve our build workflow execution time by using custom runners, improved caching and conditional Web UI builds. Runners ------- We improve our build times[0] by using larger custom runners[1] when building the UI and Vault. Caching ------- We improve Vault caching by keeping a cache for each build job. This strategy has the following properties which should result in faster build times when `go.sum` hasn't been changed from prior builds, or when a pull request is retried or updated after a prior successful build: * Builds will restore cached Go modules and Go build cache according to the Go version, platform, architecture, go tags, and hash of `go.sum` that relates to each individual build workflow. This reduces the amount of time it will take to download the cache on hits and upload the cache on misses. * Parallel build workflows won't clobber each others build cache. This results in much faster compile times after cache hits because the Go compiler can reuse the platform, architecture, and tag specific build cache that it created on prior runs. * Older modules and build cache will not be uploaded when creating a new cache. This should result in lean cache sizes on an ongoing basis. * On cache misses we will have to upload our compressed module and build cache. This will slightly extend the build time for pull requests that modify `go.sum`. Web UI ------ We no longer build the web UI in every build workflow. Instead we separate the UI building into its own workflow and cache the resulting assets. The same UI assets are restored from cache during build worklows. This strategy has the following properties: * If the `ui` directory has not changed from prior builds we'll restore `http/web_ui` from cache and skip building the UI for no reason. * We continue to use the built-in `yarn` caching functionality in `action/setup-node`. The default mode saves the `yarn` global cache. to improve UI build times if the cache has not been modified. Changes ------- * Add per platform/archicture Go module and build caching * Move UI building into a separate job and cache the result * Restore UI cache during build * Pin workflows Notes ----- [0] https://hashicorp.atlassian.net/browse/QT-578 [1] https://github.com/hashicorp/vault/actions/runs/5415830307/jobs/9844829929 Signed-off-by: Ryan Cragun <me@ryan.ec>
2023-07-05 21:19:49 +00:00
- uses: actions/checkout@c85c95e3d7251135ab7dc9ce3241c5835cc595a9 # v3.5.3
- uses: ./.github/actions/set-up-go
with:
github-token: ${{ secrets.ELEVATED_GITHUB_TOKEN }}
- name: Authenticate to Vault
id: vault-auth
if: github.repository == 'hashicorp/vault-enterprise'
run: vault-auth
- name: Fetch Secrets
id: secrets
if: github.repository == 'hashicorp/vault-enterprise'
uses: hashicorp/vault-action@130d1f5f4fe645bb6c83e4225c04d64cfb62de6e
with:
url: ${{ steps.vault-auth.outputs.addr }}
caCertificate: ${{ steps.vault-auth.outputs.ca_certificate }}
token: ${{ steps.vault-auth.outputs.token }}
secrets: |
kv/data/github/${{ github.repository }}/datadog-ci DATADOG_API_KEY;
kv/data/github/${{ github.repository }}/github-token username-and-token | github-token;
kv/data/github/${{ github.repository }}/license license_1 | VAULT_LICENSE_CI;
kv/data/github/${{ github.repository }}/license license_2 | VAULT_LICENSE_2;
kv/data/github/${{ github.repository }}/hcp-link HCP_API_ADDRESS;
kv/data/github/${{ github.repository }}/hcp-link HCP_AUTH_URL;
kv/data/github/${{ github.repository }}/hcp-link HCP_CLIENT_ID;
kv/data/github/${{ github.repository }}/hcp-link HCP_CLIENT_SECRET;
kv/data/github/${{ github.repository }}/hcp-link HCP_RESOURCE_ID;
- id: setup-git-private
name: Setup Git configuration (private)
if: github.repository == 'hashicorp/vault-enterprise'
run: |
git config --global url."https://${{ steps.secrets.outputs.github-token }}@github.com".insteadOf https://github.com
- id: setup-git-public
name: Setup Git configuration (public)
if: github.repository != 'hashicorp/vault-enterprise'
run: |
git config --global url."https://${{ secrets.ELEVATED_GITHUB_TOKEN}}@github.com".insteadOf https://github.com
- id: build
if: contains(matrix.packages, '_binary')
env:
GOPRIVATE: github.com/hashicorp/*
run: time make ci-bootstrap dev
- id: run-go-tests
name: Run Go tests
timeout-minutes: ${{ fromJSON(env.TIMEOUT_IN_MINUTES) }}
env:
COMMIT_SHA: ${{ github.sha }}
run: |
set -exo pipefail
# Build the dynamically generated source files.
make prep
# We don't want VAULT_LICENSE set when running Go tests, because that's
# not what developers have in their environments and it could break some
# tests; it would be like setting VAULT_TOKEN. However some non-Go
# CI commands, like the UI tests, shouldn't have to worry about licensing.
# So we provide the tests which want an externally supplied license with licenses
# via the VAULT_LICENSE_CI and VAULT_LICENSE_2 environment variables, and here we unset it.
# shellcheck disable=SC2034
VAULT_LICENSE=
# Assign test licenses to relevant variables if they aren't already
if [[ ${{ github.repository }} == 'hashicorp/vault' ]]; then
export VAULT_LICENSE_CI=${{ secrets.ci_license }}
export VAULT_LICENSE_2=${{ secrets.ci_license_2 }}
export HCP_API_ADDRESS=${{ secrets.HCP_API_ADDRESS }}
export HCP_AUTH_URL=${{ secrets.HCP_AUTH_URL }}
export HCP_CLIENT_ID=${{ secrets.HCP_CLIENT_ID }}
export HCP_CLIENT_SECRET=${{ secrets.HCP_CLIENT_SECRET }}
export HCP_RESOURCE_ID=${{ secrets.HCP_RESOURCE_ID }}
# Temporarily removing this variable to cause HCP Link tests
# to be skipped.
#export HCP_SCADA_ADDRESS=${{ secrets.HCP_SCADA_ADDRESS }}
fi
if [ -f bin/vault ]; then
VAULT_BINARY="$(pwd)/bin/vault"
export VAULT_BINARY
fi
# On a release branch, add a flag to rerun failed tests
if [[ "${{ github.ref_name }}" = "release/*" ]]; then
RERUN_FAILS="--rerun-fails"
fi
# shellcheck disable=SC2086 # can't quote RERUN_FAILS
GOARCH=${{ inputs.go-arch }} \
go run gotest.tools/gotestsum --format=short-verbose \
--junitfile test-results/go-test/results-${{ matrix.id }}.xml \
--jsonfile test-results/go-test/results-${{ matrix.id }}.json \
--jsonfile-timing-events failure-summary-${{ matrix.id }}${{ inputs.name != '' && '-' || '' }}${{ inputs.name }}.json \
$RERUN_FAILS \
--packages "${{ matrix.packages }}" \
-- \
-tags "${{ inputs.go-tags }}" \
-timeout=${{ env.TIMEOUT_IN_MINUTES }}m \
[QT-576] Optimize build workflow (#21486) (#21601) Improve our build workflow execution time by using custom runners, improved caching and conditional Web UI builds. Runners ------- We improve our build times[0] by using larger custom runners[1] when building the UI and Vault. Caching ------- We improve Vault caching by keeping a cache for each build job. This strategy has the following properties which should result in faster build times when `go.sum` hasn't been changed from prior builds, or when a pull request is retried or updated after a prior successful build: * Builds will restore cached Go modules and Go build cache according to the Go version, platform, architecture, go tags, and hash of `go.sum` that relates to each individual build workflow. This reduces the amount of time it will take to download the cache on hits and upload the cache on misses. * Parallel build workflows won't clobber each others build cache. This results in much faster compile times after cache hits because the Go compiler can reuse the platform, architecture, and tag specific build cache that it created on prior runs. * Older modules and build cache will not be uploaded when creating a new cache. This should result in lean cache sizes on an ongoing basis. * On cache misses we will have to upload our compressed module and build cache. This will slightly extend the build time for pull requests that modify `go.sum`. Web UI ------ We no longer build the web UI in every build workflow. Instead we separate the UI building into its own workflow and cache the resulting assets. The same UI assets are restored from cache during build worklows. This strategy has the following properties: * If the `ui` directory has not changed from prior builds we'll restore `http/web_ui` from cache and skip building the UI for no reason. * We continue to use the built-in `yarn` caching functionality in `action/setup-node`. The default mode saves the `yarn` global cache. to improve UI build times if the cache has not been modified. Changes ------- * Add per platform/archicture Go module and build caching * Move UI building into a separate job and cache the result * Restore UI cache during build * Pin workflows Notes ----- [0] https://hashicorp.atlassian.net/browse/QT-578 [1] https://github.com/hashicorp/vault/actions/runs/5415830307/jobs/9844829929 Signed-off-by: Ryan Cragun <me@ryan.ec>
2023-07-05 21:19:49 +00:00
-parallel=${{ inputs.go-test-parallelism }} \
${{ inputs.extra-flags }} \
- name: Prepare datadog-ci
if: github.repository == 'hashicorp/vault' && (success() || failure())
continue-on-error: true
run: |
curl -L --fail "https://github.com/DataDog/datadog-ci/releases/latest/download/datadog-ci_linux-x64" --output "/usr/local/bin/datadog-ci"
chmod +x /usr/local/bin/datadog-ci
- name: Upload test results to DataDog
continue-on-error: true
env:
DD_ENV: ci
run: |
if [[ ${{ github.repository }} == 'hashicorp/vault' ]]; then
export DATADOG_API_KEY=${{ secrets.DATADOG_API_KEY }}
fi
datadog-ci junit upload --service "$GITHUB_REPOSITORY" test-results/go-test/results-${{ matrix.id }}.xml
if: success() || failure()
- name: Archive test results
uses: actions/upload-artifact@0b7f8abb1508181956e8e162db84b466c27e18ce # v3.1.2
with:
name: test-results${{ inputs.name != '' && '-' || '' }}${{ inputs.name }}
path: test-results/go-test
if: success() || failure()
# GitHub Actions doesn't expose the job ID or the URL to the job execution,
# so we have to fetch it from the API
- name: Fetch job logs URL
uses: actions/github-script@d7906e4ad0b1822421a7e6a35d5ca353c962f410 # v6.4.1
if: success() || failure()
continue-on-error: true
with:
retries: 3
script: |
// We surround the whole script with a try-catch block, to avoid each of the matrix jobs
// displaying an error in the GHA workflow run annotations, which gets very noisy.
// If an error occurs, it will be logged so that we don't lose any information about the reason for failure.
try {
const fs = require("fs");
const result = await github.rest.actions.listJobsForWorkflowRun({
owner: context.repo.owner,
per_page: 100,
repo: context.repo.repo,
run_id: context.runId,
});
// Determine what job name to use for the query. These values are hardcoded, because GHA doesn't
// expose them in any of the contexts available within a workflow run.
let prefixToSearchFor;
switch ("${{ inputs.name }}") {
case "race":
prefixToSearchFor = 'Run Go tests with data race detection / test-go (${{ matrix.id }},'
break
case "fips":
prefixToSearchFor = 'Run Go tests with FIPS configuration / test-go (${{ matrix.id }},'
break
default:
prefixToSearchFor = 'Run Go tests / test-go (${{ matrix.id }},'
}
const jobData = result.data.jobs.filter(
(job) => job.name.startsWith(prefixToSearchFor)
);
const url = jobData[0].html_url;
const envVarName = "GH_JOB_URL";
const envVar = envVarName + "=" + url;
const envFile = process.env.GITHUB_ENV;
fs.appendFile(envFile, envVar, (err) => {
if (err) throw err;
console.log("Successfully set " + envVarName + " to: " + url);
});
} catch (error) {
console.log("Error: " + error);
return
}
- name: Prepare failure summary
if: success() || failure()
continue-on-error: true
run: |
# This jq query filters out successful tests, leaving only the failures.
# Then, it formats the results into rows of a Markdown table.
# An example row will resemble this:
# | github.com/hashicorp/vault/package | TestName | fips | 0 | 2 | [view results](github.com/link-to-logs) |
jq -r -n 'inputs
| select(.Action == "fail")
| "| ${{inputs.name}} | \(.Package) | \(.Test // "-") | \(.Elapsed) | ${{ matrix.id }} | [view test results :scroll:](${{ env.GH_JOB_URL }}) |"' \
failure-summary-${{ matrix.id }}${{ inputs.name != '' && '-' || '' }}${{inputs.name}}.json \
>> failure-summary-${{ matrix.id }}${{ inputs.name != '' && '-' || '' }}${{inputs.name}}.md
- name: Upload failure summary
uses: actions/upload-artifact@0b7f8abb1508181956e8e162db84b466c27e18ce # v3.1.2
if: success() || failure()
with:
name: failure-summary
path: failure-summary-${{ matrix.id }}${{ inputs.name != '' && '-' || '' }}${{inputs.name}}.md
test-collect-reports:
needs: test-go
runs-on: ${{ fromJSON(inputs.runs-on) }}
steps:
- uses: actions/cache@88522ab9f39a2ea568f7027eddc7d8d8bc9d59c8 # v3.3.1
with:
path: test-results/go-test
key: go-test-reports-${{ github.run_number }}
restore-keys: go-test-reports-
- uses: actions/download-artifact@9bc31d5ccc31df68ecc42ccf4149144866c47d8a # v3.0.2
with:
name: test-results
path: test-results/go-test
- run: |
ls -lhR test-results/go-test
find test-results/go-test -mindepth 1 -mtime +3 -delete
ls -lhR test-results/go-test