I’ve been fiddling with this for an embarrassingly long time. There has to be something I’m missing, but I just can’t see it. I have a staged build. The deploy
stage is firing as expected, as are most of its phases, but not the deploy phase. Any idea why?
stages:
- name: build
- name: publish
if: (type == push && branch == rob-release-and-deploy) || tag IS present
- name: deploy
if: (type == push && branch == rob-release-and-deploy) || tag IS present
- name: clean
# ... Other bits until we hit the deploy stage of jobs: include: ...
- stage: deploy
name: "Deploy to dev|aut|stg"
install:
- curl -LO https://storage.googleapis.com/kubernetes-release/release/v1.23.6/bin/linux/amd64/kubectl
- chmod +x ./kubectl
- mv ./kubectl ${HOME}/.local/bin
script:
- echo "Placeholder?"
before_deploy:
- aws ecr get-login-password --region "${AWS_REGION}" | docker login --username AWS --password-stdin "${AWS_ECR_REGISTRY_URL}/tmp"
deploy:
- provider: script
script: "bash ./bin/deploy dev"
skip_cleanup: true
on:
branch: rob-release-and-deploy
- provider: script
script: "bash ./bin/deploy aut"
skip_cleanup: true
on:
condition: tag IS present && (tag =~ /^\d{8}\.rc\d+$/)
At this point, I’m committing code to the rob-release-and-deploy
branch (a PR open on that branch). There’s no indication that the deploy: phase is being recognized at all. It’s not being skipped with the message I might normally see if I were pushing to a different branch or something…it’s simply not doing anything at all. The build succeeds after the script
phase.
The last bit of the build stage output looks like this:
0.00s$ echo "Placeholder?"
189Placeholder?
190The command "echo "Placeholder?"" exited with 0.
191
192travis_run_after_success: command not found
193travis_run_after_failure: command not found
194travis_run_after_script: command not found
195travis_run_finish: command not found
196
197Done. Your build exited with 0.
I could really use another set of eyes on this.