Caching artifacts in GitHub actions using runner controller - github

Question

I want to set up self-hosted runners on a k8s cluster using actions-runner-controller.
My question is, given that as per the official docs, persistent runners are not recommended
Although not generally recommended, it’s possible to disable the
passing of the --ephemeral flag by explicitly setting ephemeral: false
in the RunnerDeployment or RunnerSet spec. When disabled, your runner
becomes “persistent”.
how can one leverage artifact caching when using this controller?
Where will the cache content will be stored in the k8s cluster, given that containers are ephemeral?

Answer 1

If you are not using the enterprise version, the caches will be handled by Github itself. I came across some similar problems at my self-hosted runner to create a cache for nodeJs, VueJs, and Java. Here's what I did:
VueJs (moving dist folder) (note the actions/upload-artifact#v3)
name: CI
on:
push:
branches: [ "master" ]
pull_request:
branches: [ "master" ]
workflow_dispatch:
jobs:
build-web:
runs-on: self-hosted
container:
image: node:14
steps:
- uses: actions/checkout#v3
- name: Build shc-web
run: |
yarn config set cache-folder .yarn
yarn
yarn run build
- uses: actions/upload-artifact#v3
with:
name: dist-folder
path: dist/
registry-web:
runs-on: self-hosted
needs: ['build-web']
steps:
- uses: actions/checkout#v3
- uses: actions/download-artifact#v3
with:
name: dist-folder
path: dist/
- name: Configure AWS
uses: aws-actions/configure-aws-credentials#v1
with:
aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
aws-region: us-east-1
- name: Login to Amazon ECR
id: login-ecr
uses: aws-actions/amazon-ecr-login#v1
- name: Registry on AWS repository
env:
REGISTRY: ${{ steps.login-ecr.outputs.registry }}
REPOSITORY: shccp
run: |
docker build -t $REGISTRY/$REPOSITORY:3.1.x-$GITHUB_RUN_ID .
docker push $REGISTRY/$REPOSITORY:3.1.x-$GITHUB_RUN_ID
Also, I used two different jobs to handle the build. It could be done in only one so there was no need to upload/download the dist. Actually, that was precisely what I had to do in the NodeJs action. The node_modules is just too big to be uploaded.
NodeJS:
name: CI
on:
push:
branches: [ "stage" ]
pull_request:
branches: [ "stage" ]
workflow_dispatch:
jobs:
ci-api:
runs-on: self-hosted
steps:
- uses: actions/checkout#v3
- uses: actions/setup-node#v3
with:
node-version: 14
- name: Build api
run: npm install
- name: Configure AWS
uses: aws-actions/configure-aws-credentials#v1
with:
aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
aws-region: us-east-1
- name: Login to Amazon ECR
id: login-ecr
uses: aws-actions/amazon-ecr-login#v1
- name: Registry on AWS repository
id: registry-aws
env:
REGISTRY: ${{ steps.login-ecr.outputs.registry }}
REPOSITORY: shcapi
run: |
docker build -t $REGISTRY/$REPOSITORY:3.1.x-$GITHUB_RUN_ID .
docker push $REGISTRY/$REPOSITORY:3.1.x-$GITHUB_RUN_ID
echo "::set-output name=image-tag::$REGISTRY/$REPOSITORY:3.1.x-$GITHUB_RUN_ID"
No cache is needed once it is done in a single job. That is a pretty nice feature of Github actions btw.
The Java cache, on the other hand, is handled by the following action:
name: CI
on:
push:
branches: [ "master" ]
pull_request:
branches: [ "master" ]
workflow_dispatch:
jobs:
ci-etlv4:
runs-on: self-hosted
steps:
- uses: actions/checkout#v3
- uses: actions/setup-java#v3
with:
distribution: adopt-openj9
java-version: 8
cache: 'maven'
- uses: stCarolas/setup-maven#v4.4
with:
maven-version: 3.8.2
- name: Build ETLv4
run: |
echo ${{ secrets.SETTINGS_BASE64 }} | base64 -d > settings.xml
mvn --settings settings.xml --global-settings settings.xml clean package -DskipTests=true
- uses: docker/login-action#v2
with:
registry: "iad.ocir.io"
username: ${{ secrets.OCI_REGISTRY_USER }}
password: ${{ secrets.OCI_REGISTRY_PASSWORD }}
- uses: docker/setup-qemu-action#v2
- uses: docker/setup-buildx-action#v2
with:
driver: docker
- uses: docker/build-push-action#v3
with:
context: .
push: true
tags: XXXXX
The actions/setup-java#v3 can deal with the maven/gradle caches.
Hope it helps.