I have recently setup a new project that deploy a static blog to firebase. The CI pipeline were successful but suddenly starts to fail the last few times. When I check, there seems to be an extra job pages:deploy that is not defined on my .gitlab-ci.yml files. Hovering over the job says: pages:deploy no entries extracted and I cannot click on or restart that job.
Here is the content of my .gitlab-ci.yml:
# This file is a template, and might need editing before it works on your project.
image: node:lts-alpine
# This folder is cached between builds
# http://docs.gitlab.com/ce/ci/yaml/README.html#cache
cache:
paths:
- node_modules/
- .yarn
stages:
- build
- deploy
pages:
stage: build
script:
- yarn install
- yarn build
artifacts:
paths:
- dist
deploy:
stage: deploy
script:
- yarn deploy
environment:
name: production
only:
- master
Appreciate any help!
I realise it’s been a while since you posted this, and you never got a reply - but I just came across this exact issue with my own GitLab Pages CI job, and this unanswered question came up.
It seems like the artifacts: paths argument basically has to be public, like this example.
My .gitlab-ci.yml now looks like:
image: node:8.16
pages:
stage: deploy
script:
- npm install
- npm audit fix
- gulp
artifacts:
paths:
- public
Previously I used to have the path as site, and I got the same pages:deploy no entries extracted error. It doesn’t make any sense that is must be public, because otherwise what’s the point of being able to specify it, but it worked for me.
2 Likes
I just came across the issue for days and would like to confirm the solution.
The ghost job for gitlab page comes from nowhere.
The job name have to be pages.
The state can be any.
The image can also be anything.
The artifacts path have to be public, cannot be in any subdirectory either.
I think GitLab should show some warning about this incorrect GitLab Pages format.
The developer just guess the correct format and never know it works or not until the end.