Some checks failed
build-and-release / build (catering-services) (push) Successful in 17s
build-and-release / build (companies) (push) Successful in 30s
build-and-release / build (cron) (push) Successful in 37s
build-and-release / build (developers) (push) Successful in 27s
build-and-release / build (fitness-trainers) (push) Successful in 31s
build-and-release / build (graphic-designers) (push) Successful in 15s
build-and-release / build (job-seekers) (push) Successful in 11s
build-and-release / build (makeup-artists) (push) Successful in 9s
build-and-release / build (payments) (push) Successful in 10s
build-and-release / build (gateway) (push) Successful in 42s
build-and-release / build (photographers) (push) Successful in 14s
build-and-release / build (social-media-managers) (push) Successful in 13s
build-and-release / build (tutors) (push) Successful in 15s
build-and-release / build (ugc-content-creators) (push) Successful in 15s
build-and-release / build (customers) (push) Failing after 1m46s
build-and-release / build (employees) (push) Successful in 2m0s
backend-integration-tests / ai-credits (push) Failing after 58s
build-and-release / build (jobs) (push) Successful in 1m56s
build-and-release / build (video-editors) (push) Successful in 1m34s
build-and-release / build (users) (push) Successful in 2m53s
The bare bookworm runner image has no apt-get/package manager at all (the earlier build-essential install attempt failed with 'apt-get: command not found'), and a plain 'docker run -v $PWD:/workspace' wouldn't work either - DOCKER_HOST points at the sibling dind engine (same nested-container setup build.yaml's own comments describe), so a bind mount would look for the path on the wrong filesystem. Matches build.yaml's actual working pattern instead: docker buildx build with context transfer (not a volume mount), using a Dockerfile (Dockerfile.test) whose RUN step runs the test suite and fails the build on a non-zero exit. TEST_DATABASE_URL passed via buildx --secret (not --build-arg) so it never lands in image layer history. Isolated this by testing each piece of the original workflow independently (env/secrets, id+GITHUB_OUTPUT, if: conditionals all checked out fine one at a time) until apt-get was confirmed as the actual failure point. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
87 lines
3.3 KiB
YAML
87 lines
3.3 KiB
YAML
name: backend-integration-tests
|
|
|
|
on:
|
|
push:
|
|
branches:
|
|
- main
|
|
- high-performance
|
|
|
|
concurrency:
|
|
group: backend-tests-${{ github.ref }}
|
|
cancel-in-progress: true
|
|
|
|
jobs:
|
|
ai-credits:
|
|
runs-on: docker-ready
|
|
env:
|
|
TEST_DATABASE_URL: ${{ secrets.TEST_DATABASE_URL }}
|
|
steps:
|
|
- name: Checkout
|
|
uses: actions/checkout@v4
|
|
with:
|
|
fetch-depth: 0
|
|
|
|
- name: Check for relevant changes
|
|
id: check
|
|
run: |
|
|
set -euo pipefail
|
|
if git rev-parse --verify HEAD^ >/dev/null 2>&1; then
|
|
BASE="${{ github.event.before }}"
|
|
if [ -z "$BASE" ] || [ "$BASE" = "0000000000000000000000000000000000000000" ] \
|
|
|| ! git rev-parse --verify "${BASE}^{commit}" >/dev/null 2>&1; then
|
|
BASE="HEAD^"
|
|
fi
|
|
CHANGED_FILES="$(git diff --name-only "$BASE" HEAD)"
|
|
else
|
|
CHANGED_FILES="$(git ls-files)"
|
|
fi
|
|
|
|
run=false
|
|
if echo "$CHANGED_FILES" | grep -Eq '^(crates/db/|Cargo\.toml|Cargo\.lock|Dockerfile\.test|\.forgejo/workflows/test\.yaml)'; then
|
|
run=true
|
|
fi
|
|
echo "run=$run" >> "$GITHUB_OUTPUT"
|
|
if [ "$run" = "false" ]; then
|
|
echo "No changes relevant to crates/db - skipping integration tests."
|
|
fi
|
|
|
|
- name: Verify TEST_DATABASE_URL is configured
|
|
if: steps.check.outputs.run == 'true'
|
|
run: |
|
|
test -n "${TEST_DATABASE_URL:-}" || { echo "TEST_DATABASE_URL secret is not set - see docs/LIVE_SERVER_RUNBOOK.md step 6"; exit 1; }
|
|
|
|
- name: Point DOCKER_HOST at this container's own gateway
|
|
if: steps.check.outputs.run == 'true'
|
|
run: |
|
|
set -euo pipefail
|
|
# Same trick build.yaml uses: this job container is nested one level
|
|
# inside the runner pod's dind sidecar, so its own loopback isn't the
|
|
# sidecar's - read the container's actual default-route gateway
|
|
# directly from /proc/net/route instead of relying on `ip`/`route`.
|
|
GATEWAY="$(awk '$2 == "00000000" {print $3}' /proc/net/route | head -1 | \
|
|
sed -E 's/(..)(..)(..)(..)/0x\4 0x\3 0x\2 0x\1/' | \
|
|
{ read -r a b c d; printf '%d.%d.%d.%d' "$a" "$b" "$c" "$d"; })"
|
|
echo "Detected docker host gateway: $GATEWAY"
|
|
echo "DOCKER_HOST=tcp://$GATEWAY:2375" >> "$GITHUB_ENV"
|
|
|
|
- name: Set up Docker Buildx
|
|
if: steps.check.outputs.run == 'true'
|
|
run: |
|
|
set -euo pipefail
|
|
docker version
|
|
docker buildx create --use --name nxtgauge-builder || docker buildx use nxtgauge-builder
|
|
docker buildx inspect --bootstrap
|
|
|
|
- name: Run ai_credits integration tests
|
|
if: steps.check.outputs.run == 'true'
|
|
run: |
|
|
set -euo pipefail
|
|
# Build context transfer (not a bind mount) is what makes this work
|
|
# despite the nested dind setup - same reason build.yaml uses
|
|
# buildx rather than plain `docker run -v`. The Dockerfile's RUN
|
|
# step is the actual test; a non-zero exit fails this build (and
|
|
# this workflow step) the same way a failed `cargo test` would.
|
|
docker buildx build \
|
|
--secret id=test_db_url,env=TEST_DATABASE_URL \
|
|
-f Dockerfile.test \
|
|
.
|