I can't find a satisfying solution for my case.
I want to start a job manually only when a certain previous job has failed. The job in question dose a validation. I want to make the next job manual so that the user acknowledges that something wasn't good and make him investigate the problem and continue only if he deems that the fail can be ignored.
stages:
- test
- validate
- build
lint:
stage: test
allow_failure: true
script:
- npm run lint
check:reducer:
stage: test
allow_failure: true
script:
- chmod +x ./check-reducers.py
- ./check-reducers.py $CI_PROJECT_ID $CI_COMMIT_BRANCH
except:
- master
- development
fail:pause:
stage: validate
allow_failure: true
script:
- echo The 'validate:reducer' job has failed
- echo Check the job and decide if this should continue
when: manual
needs: ["check:reducer"]
build:
stage: build
script:
- cp --recursive _meta/ $BUILD_PATH
- npm run build
artifacts:
name: "build"
expire_in: 1 week
paths:
- $BUILD_PATH
needs: ["fail:pause"]
I would like that if check:reducer fails, fail:pause to wait for the user input. If check:reducer exits with 0, fail:pause should start automatically or buildshould start.
Unfortunately, this isn't possible as the when keyword is evaluated at the very start of the pipeline (I.e., before any job execution has run), so you cannot set the when condition based on the previous job status.
This is possible if you use a generated gitlab-ci.yml as a child workflow.
stages:
- test
- approve
- deploy
generate-config:
stage: test
script:
- ./bin/run-tests.sh
- ./bin/generate-workflows.sh $?
artifacts:
paths:
- deploy-gitlab-ci.yml
trigger-workflows:
stage: deploy
trigger:
include:
- artifact: deploy-gitlab-ci.yml
job: generate-config
The generate-workflows.sh script writes out a deploy-gitlab-ci.yml that either has the approval job or not based on the return code of the run-test.sh passed as the first argument to the script.
You can make it easier on yourself using includes, where you either include the approve step or not in the generated deploy-gitlab-ci.yml file, and make the steps in the deploy optionally need the approal.
approve-gitlab-ci.yml
approve:
stage: approve
when: manual
script:
- echo "Approved!"
deploy-gitlab-ci.yml
deploy:
stage: deploy
needs:
- job: approve
- optional: true
Then the deploy-gitlab-ci.yml is simply an includes with the jobs to run:
includes:
- approve-gitlab-ci.yml
- deploy-gitlab-ci.yml
Related
I am trying to setup CI in gitlab so
the second job (pushdev) will be available for running manually only after the devjob has run successfully.
the third job pushtostage will only run iff file has changed.
the way the jobs are setup, second and third jobs alway run. What is missing in the pipeline spec
devjob:
image: node:16
stage: publishdev
script:
- echo "running validation checks"
- npm run validate
rules:
- changes:
- ./src/myfile.txt
- when: manual
# - this jobs needs to run after "devjob" has run successfully
# and myfile.txt has changed
# - "needs" artifacts from the "lint" job
pushdev:
image: node:16
stage: publishdev
needs: [ "devjob", "lint"]
script:
- echo "Pushing changes after validation to dev"
- npm run pushdev
rules:
- changes:
- ./src/myfile.txt
when: on_success
- when: manual
pushtostage:
image: node:16
stage: pushstage
script:
- echo "Pushing changes to stage"
rules:
- changes:
- ./src/myfile.txt
- when: manual
I change your sample to look like this:
stages:
- publishdev
- pushstage
default:
image: ubuntu:20.04
lint:
stage: publishdev
script:
- echo "lint job"
devjob:
stage: publishdev
script:
- echo "running validation checks"
rules:
- changes:
- README.md
when: manual
allow_failure: false
pushdev:
stage: publishdev
needs: [ "devjob", "lint"]
script:
- echo "Pushing changes after validation to dev"
rules:
- changes:
- README.md
when: manual
allow_failure: false
pushtostage:
stage: pushstage
script:
- echo "Pushing changes to stage"
rules:
- changes:
- README.md
when: manual
allow_failure: false
I add allow_failure: false, because allow_failure when manual job default is true.
I merge your rules. because GitLab rules one - is one rule:
Rules are evaluated when the pipeline is created, and evaluated in order until the first match.
your .gitlab-ci.yml first job devjob is manual, so it is always a success, and your second job pushdev first rule changes and when: on_success always match, so it always run.
I change your .gitlab-ci.yml, first job devjob merge your rules when file change and set it is manual job and not allow_failure. and so on.
the sample code in Files · try-rules-stackoverflow-72594854-manual · GitLab PlayGround / Workshop / Tryci · GitLab
I got an error on GitLab Pipeline when I commit the next .gitlab-ci.yml for a repository.
Pipeline executed to build solution, deploy to Artifactory and trigger and API call
Deploy job have to be executed manually, and there are two different job options to execute.
stages:
- build
- deploy
- trigger
variables:
APP_PROJECT_ID: ${CUSTOMER_RELEASED}
build_job:
stage: build
tags:
- dotnet
script:
- echo "build"
only:
- tags
allow_failure: false
.deploy_job_base:
stage: deploy
needs: [build_job]
tags:
- dotnet
script:
- echo "deploy"
dependencies:
- build_job
only:
- tags
deploy_job_sport:
extends: .deploy_job_base
after_script:
- $APP_PROJECT_ID = "2096"
when: manual
allow_failure: false
deploy_job_others:
extends: .deploy_job_base
after_script:
- $APP_PROJECT_ID = "0"
when: manual
allow_failure: false
.trigger_base:
stage: trigger
script:
- echo "Customer Project ID '{$APP_PROJECT_ID}'"
- echo "Call API..."
trigger_sport:
extends: .trigger_base
needs: [deploy_job_sport]
trigger_others:
extends: .trigger_base
needs: [deploy_job_others]
Lint status is correct
but I get that error GitLab Pipeline when I commit changes:
Found errors in your .gitlab-ci.yml: 'trigger_sport' job needs
'deploy_job_sport' job but 'deploy_job_sport' is not in any previous
stage 'trigger_others' job needs 'deploy_job_others' job but
'deploy_job_others' is not in any previous stage
If I remove trigger_sport and trigger_others job and create only one job, it works fine but I don't know how I can target the two manual jobs (deploy_job_sport and deploy_job_others) to a single job.
Do you have any idea?
Thanks in advance.
I think this is related to the fact that you're using only: tags in your template for deploy jobs and the build job also is limited to run when commits contain a tag.
But the trigger template is missing this limitation which most likely causes this error when pushing a commit without a tag because the pipeline creation would add trigger_XY to the pipeline, which has dependencies to the previous deploy_XY jobs.
When updating your job template for trigger jobs to the following this error should be resolved:
.trigger_base:
stage: trigger
script:
- echo "Customer Project ID '{$APP_PROJECT_ID}'"
- echo "Call API..."
only:
- tags
I have like this gitlab ci cd configuration file:
image: docker:git
stages:
- develop
- production
default:
before_script:
- apk update && apk upgrade && apk add git curl
deploy:
stage: develop
script:
- echo "Hello World"
backup:
stage: develop
when:
- manual
- on_success
remove:
stage: develop
when:
- delayed
- on_success
start_in: 30 minutes
In my case job deploy runs automaticaly and job backup must runs manually only when successfully completed job deploy. But in my case this configuration doesn't works and I get error with message:
Found errors in your .gitlab-ci.yml:
jobs:backup when should be one of:
on_success
on_failure
always
manual
delayed
How I can use multiple when option arguments in my case?
Basically you can't because when does not expect an array. You can work around it though with needs. But this solution does only work if you run your jobs in different stages.
image: docker:git
stages:
- deploy
- backup
- remove
deploy:develop:
stage: deploy
script:
- exit 1
backup:develop:
stage: backup
script:
- echo "backup"
when: manual
needs: ["deploy:develop"]
remove:develop:
stage: remove
script:
- echo "remove"
when: delayed
needs: ["backup:develop"]
start_in: 30 minutes
I have a .gitlab-ci.yml that looks like this:
image: "python:3.7"
.python-tag:
tags:
- python
before_script:
- python --version
- pip install -r requirements.txt
- export PYTHONPATH=${PYTHONPATH}:./src
- python -c "import sys;print(sys.path)"
stages:
- Static Analysis
- Local Tests
- Integration Tests
- Deploy
mypy:
stage: Static Analysis
extends:
- .python-tag
script:
- mypy .
pytest-smoke:
stage: Local Tests
extends:
- .python-tag
script:
- pytest -m smoke
int-tests-1:
stage: Integration Tests
when: manual
allow_failure: false
trigger:
project: tests/gitlab-integration-testing-integration-tests
strategy: depend
int-tests-2:
stage: Integration Tests
when: manual
allow_failure: false
trigger:
project: tests/gitlab-integration-testing-integration-tests
strategy: depend
deploy:
stage: Deploy
extends:
- .python-tag
script:
- echo "Deployed!"
The Integrations stage has multiple jobs in it that take a decent chunk of time to run. It is unusual that all of the integration tests need to be run. This is why we stuck a manual flag on these, and the specific ones needed will be manually run.
How do I make it so that the Deploy stage requires that one or more of the jobs in Integration Tests has passed? I can either do all like I have now or I can do none by removing allow_failure: false from the integration test jobs.
I want to require that at least once has passed.
if each job generate an artifcact only when the job is successful
artifacts:
paths:
- success.txt
script:
# generate the success.txt file
you should be able to test if the file exist in the next stage
you need to add this (below) in the next stage to be able to read the file:
artifacts:
paths:
- success.txt
I was setting up a gitlab environment. After each push 1 am running 5 test cases. But if the any of the test cases is falling other testcase are skipped.
I want to run all the cases. Because they are independent to each other.
gitlab-ci.yml
stages:
- build
- unit_test_1
- unit_test_2
- unit_test_3
job1:
stage: build
script:
- bash build.sh
job2:
stage: unit_test_1
script:
- bash ./unit_test_1.sh
job3:
stage: unit_test_2
script:
- bash ./unit_test_2.sh
job4:
stage: unit_test_3
script:
- bash ./unit_test_3.sh
If uint_test_1.sh is failing. Other tests are skipped.
You can use the when property to make your jobs run every time, regardless of the status of jobs from prior stages of the build.
stages:
- build
- test
job1:
stage: build
script:
- bash build.sh
job2:
stage: test
when: always
script:
- bash ./unit_test_1.sh
job3:
stage: test
when: always
script:
- bash ./unit_test_2.sh
job4:
stage: test
when: always
script:
- bash ./unit_test_3.sh
Also, if you want to make sure you never have jobs running in parallel then you can configure your runners with concurrency limits.
Configuring it globally will limit all your runners to only run 1 job concurrently between all runners.
Configuring it per runner will limit that runner to only run 1 job concurrently per build token.
You can try like this:
gitlab-ci.yml
stages:
- build
- test
job1:
stage: build
script:
- bash build.sh
job2:
stage: test
script:
- bash ./unit_test_1.sh
job3:
stage: test
script:
- bash ./unit_test_2.sh
job4:
stage: test
script:
- bash ./unit_test_3.sh
The documentation say:
The ordering of elements in stages defines the ordering of builds' execution:
Builds of the same stage are run in parallel.
Builds of the next stage are run after the jobs from the previous stage complete successfully.
https://docs.gitlab.com/ce/ci/yaml/README.html#stages
To run in parallel you have to put the same stage name
https://docs.gitlab.com/ce/ci/pipelines.html#pipelines