Unable to build images on arm64
I have a task definition like
platform: linux
image_resource:
type: registry-image
source:
repository: concourse/oci-build-task
inputs:
- name: repo
outputs:
- name: my-image
run:
path: build
params:
CONTEXT: repo
When I run it on my M1 Mac, when it gets to the first RUN command in the Dockerfile I get the error:
#7 0.260 container_linux.go:380: starting container process caused: process_linux.go:393: copying bootstrap data to pipe caused: write init-p: broken pipe
When I run it in a linux/amd64 EC2 instance, the same task completes successfully. I'm using the docker-compose from the QuickStart, but with the concourse image changed to rdclda/concourse:7.8.2 since concourse/concourse won't run on my M1.
The full file is:
version: '3'
services:
concourse-db:
image: postgres
environment:
POSTGRES_DB: concourse
POSTGRES_PASSWORD: concourse_pass
POSTGRES_USER: concourse_user
PGDATA: /database
concourse:
image: rdclda/concourse:7.8.2
command: quickstart
privileged: true
depends_on: [ concourse-db ]
ports: [ "8080:8080" ]
environment:
CONCOURSE_POSTGRES_HOST: concourse-db
CONCOURSE_POSTGRES_USER: concourse_user
CONCOURSE_POSTGRES_PASSWORD: concourse_pass
CONCOURSE_POSTGRES_DATABASE: concourse
CONCOURSE_EXTERNAL_URL: http://localhost:8080
CONCOURSE_ADD_LOCAL_USER: test:test
CONCOURSE_MAIN_TEAM_LOCAL_USER: test
# instead of relying on the default "detect"
CONCOURSE_WORKER_BAGGAGECLAIM_DRIVER: overlay
CONCOURSE_CLIENT_SECRET: Y29uY291cnNlLXdlYgo=
CONCOURSE_TSA_CLIENT_SECRET: Y29uY291cnNlLXdvcmtlcgo=
CONCOURSE_X_FRAME_OPTIONS: allow
CONCOURSE_CONTENT_SECURITY_POLICY: "*"
CONCOURSE_CLUSTER_NAME: tutorial
CONCOURSE_WORKER_CONTAINERD_DNS_SERVER: "8.8.8.8"
CONCOURSE_WORKER_RUNTIME: "containerd"
Hi there @palmsey
I'm running into a similar issue, so I was wondering if you ever got this to work and kept a working config somewhere ?
Thanks in advance!
Sorry, I didn't find a solution for this and eventually switched to GH Actions.
thanks for the prompt reply @palmsey !
For What It's Worth (FWIW), I did eventually manage to get this going. More details on https://github.com/concourse/concourse/issues/8719#issuecomment-1676821087