I'm developing a Github actions workflow. This workflow runs on Linux, Mac, and Windows.
As part of the workflow, I have to check whether 2 environment variables are equal. If they don't - fail the job.
As described here, Github Actions support if: condition:
steps:
- run: # How can I make a cross-platform failure here?
if: ${{ envA }} != ${{ envB }}
How can I make the job fail if the above condition is true?
In the beginning, I thought of a script, but there must be a more elegant way to fail a job.
I'd do run: exit 1. That will simply exit with an exit code of 1, on all three platforms.
Proof that it's cross-platform: https://github.com/rmunn/Testing/runs/220188838 which runs the following workflow:
name: Test exiting on failure
on: [push]
jobs:
build:
strategy:
matrix:
os: [ubuntu-latest, windows-latest, macOS-latest]
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout#v1
- name: Try to fail
run: exit 1
- name: Print message if we don't fail
run: echo Should not get here
(An earlier version of this answer recommended "/bin/false", but that would only work on Linux and macOS).
In 2021, there is perhaps a more graceful way to do this:
- name: A/B Check
if: ${{ envA }} != ${{ envB }}
uses: actions/github-script#v3
with:
script: |
core.setFailed('envA and envB are not equivalent!')
Here, we use the github-script action to provide a one liner script that will fail the job. The "A/B Check" step will only run if the condition in the if line is true, so the script will only run in that case, which is what we want.
The nice thing about this approach is that you will get nicely formatted output in the Actions UI in your repo, showing that the "A/B Check" step caused the failure, and why (i.e. "envA and envB are not equivalent").
Note that if you have additional steps in the job after this, and you do NOT want them to run if the A/B check fails, you'll want to use if: success() on them to prevent them from running in that case.
The Github workflow commands docs gives a hint on this.
Toolkit function
Equivalent workflow command
core.setFailed
Used as a shortcut for ::error and exit 1
Given that, you can do the following without using any external workflows.
steps:
- name: A/B Check
if: ${{ envA }} != ${{ envB }}
run: |
echo "::error file={name},line={line},endLine={endLine},title={title}::{message}"
exit 1
Related
I have configured my github actions which runs tests in parallel for different platforms. At the end of my tests I want the status to be saved to the outputs. Once all jobs complete I have another job that runs to send the results to a slack webhook.
I am having difficulty determining a method to save the output for multiple jobs and assuring there is no issues when they are running in parallel.
For example this is my code snippet
name: Test Notify
on:
push:
jobs:
build:
strategy:
matrix:
config:
- name: 'Ubuntu 18.04'
runner: 'ubuntu-18.04'
id: 'u18'
- name: 'Ubuntu 20.04'
runner: 'ubuntu-20.04'
fail-fast: false
runs-on: ${{ matrix.config.runner }}
outputs:
# Prefer to have one general output I can append to
global: ${{ steps.status.outputs.global }}
# I can output to separate outputs but I rather have a single one as shown above
u18: ${{ steps.status.outputs.u18 }}
u20: ${{ steps.status.outputs.u20 }}
steps:
- name: Test Failure u18
id: step1
if: ${{ matrix.config.id == 'u18' }}
run: |
exit 1
- name: Doing Step 2
id: step2
run: |
echo "DO NOTHING"
- name: Output Status
id: status
if: always()
env:
JOB_STATUS: "${{ job.status }}"
run: |
# This works, but is it safe since I have u18 and u20 running in parallel ?
echo "${{ matrix.config.id }}=$JOB_STATUS" >> $GITHUB_OUTPUT
# Is there a safe way to have a single status string that I add to, for example;
# echo "global=${{ github_output.global}}$JOB_STATUS" >> $GITHUB_OUTPUT
webhook:
needs: build
runs-on: 'ubuntu-20.04'
if: always()
steps:
- name: Send webhook update for all jobs
env:
JSON_RESULTS: "${{ toJSON(needs.build-and-test) }}"
run: |
# Will add code to properly send the information
echo $JSON_RESULTS
Currently, there is no easy way to reference all outputs of matrix jobs. Moreover combining it into a single output.
The issue is that only a single value is available for future jobs that need the strategy.matrix job’s output because even if the output is set by multiple matrix variations of the job, only one is retained.
For more detail, see the Community discussion.
TL;DR:
There are several workarounds:
defining separate outputs for a job with strategy.matrix by letting the job variations set different outputs
then process these outputs in a separate step that can provide single output for further steps
use artifacts to store the matrix jobs' outputs and then post-process it (discussioncomment-3814009)
If we were to use minimal resources, doing these steps would look something like:
setup --- checkout, install fixed versions via cache
then trigger in parallel 1 command to run:
lint
test
e2e
all in parallel.
Unfortunately, with the current options I've found out there, the best scenario I can find is to have a separate workflow file for lint, test, e2e that are called by a main workflow and upon success build an artifact.
The issue with this is.
checkout will run 3 times.
install command will run 3 times.
Which can be mitigated with caching, but you're still looking at 10-15 seconds extra per job for those steps.
Is there a way to optimise this and still keep the reusable steps?
Even if I would keep everything in 1 giant file, I'd still define the lint, test, e2e steps as separate jobs and not as 1 single step part of a job.
Can we call a 'workflow' and run it in the same environment on the spot somehow?
This example calls checkout and caches the repo using the $ {{ github.run_id }} to prevent the cache conflicting with other runs. The cache is then used by the child workflows in place of checking out. e2e, test and lint will wait for the cache to be created and then all run in parallel.
Depending on your use case it may be excessive to create a new cache for each run_id, you can adjust this as you need. If speed is what you're after I'd also suggest only caching the files that you need for your child workflows, and not take everything like I've done in the example below.
name: main_caller
on: workflow_dispatch
jobs:
cache_checkout:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout#v3
- uses: actions/upload-artifact#v3
with:
name: my-checkout-${{ github.run_id }}
path: ${{ github.workspace }}/**/*
run_lint:
needs: cache_checkout
uses: ./.github/workflows/lint.yml
run_test:
needs: cache_checkout
uses: ./.github/workflows/test.yml
./.github/workflows/lint.yml example: (e2e and test will follow this pattern)
name: lint
on: workflow_call
jobs:
lint:
runs-on: ubuntu-latest
steps:
- name: Get Cached Checkout
uses: actions/download-artifact#v3
with:
name: my-checkout-${{ github.run_id }}
- name: Do linting things
run: echo "I'm linting!"
Edited as per comments - all jobs in GitHub run in parallel, but to have cleaner example you could use matrix strategy. To make it work you need it to depend on a 'setup' job that will install everything and create the artifacts (this is done using the needs keyword).
In the following example the install job would setup everything and then the example job, which uses matrix strategy, will run the lint/test/e2e flow in separate parallel jobs after it downloads the required artifacts.
jobs:
install:
runs-on: ubuntu-latest
- uses: actions/checkout#v3
- name: Compile artifacts
run: *install fixed versions/create artifacts*
- uses: actions/upload-artifact#v3
with:
name: my-artifact
path: path/to/my_artifact
example:
runs-on: ubuntu-latest
needs: install
strategy:
matrix:
action: [lint, test, e2e]
steps:
- uses: actions/download-artifact#v3
with:
name: my-artifact
path: path/to/artifact
- name: lint
if: ${{ matrix.lint }}
run: echo "lint"
- name: test
if: ${{ matrix.test }}
run: echo "test"
- name: e2e
if: ${{ matrix.e2e }}
run: echo "e2e"
It will create a workflow like this:
Building a GitHub action based on the commit message I'm trying to base a step on whether the commit message contains a particular string, set it to a variable and then in the next step check with a condition.
My current implementation of my action works:
name: Smoke Test
on:
push:
branches:
- main
permissions:
contents: read
issues: write
jobs:
smoking:
runs-on: [ubuntu-latest]
steps:
- name: Run smoke tests
if: ${{ !contains(github.event.head_commit.message, 'smoke_test') }}
run: |
echo 'Smoke Test not requested'
exit 1
stuff:
needs: smoking
runs-on: ubuntu-latest
steps:
- uses: actions/checkout#v2
- uses: JasonEtco/create-an-issue#v2
env:
GITHUB_TOKEN: ${{ secrets.TOKEN }}
with:
filename: .github/ISSUE_TEMPLATE/smoke-test.md
id: create-issue
- run: 'echo Created issue number ${{ steps.create-issue.outputs.number }}'
- run: 'echo Created ${{ steps.create-issue.outputs.url }}'
but with the implementation of:
exit 1
causes the action to indicate it error'ed out in the action panel and while that works that isn't technically accurate because I don't need it to error I just don't want the remaining steps to run.
I've tried setting a variable:
if: ${{ contains(github.event.head_commit.message, 'smoke_test') }}
with:
run-smoke-test: true
run: |
echo 'Smoke Test requested'
but it's not passing to the next step.
Research
Use environment variable in github action if
How to pass variable between two successive GitHub Actions jobs?
github-action: does the IF have an ELSE?
How to fail a job in GitHub Actions?
GitHub Actions - trigger another action after one action is completed
Without relying on another GitHub action is there a way in step smoking to set an env variable that step stuff would need to validate for before running the step?
Edit
After reading the answer and implementing job outputs I've written:
name: Smoke Test
on:
push:
branches:
- main
permissions:
contents: read
issues: write
jobs:
commitMessage:
runs-on: ubuntu-latest
outputs:
output1: ${{ steps.isSmoke.outputs.test }}
steps:
- id: isSmoke
if: ${{ contains(github.event.head_commit.message, 'smoke_test') }}
run: echo "::set-output name=test::true"
smokeTest:
runs-on: ubuntu-latest
needs: commitMessage
steps:
- uses: actions/checkout#v2
- uses: JasonEtco/create-an-issue#v2
if: steps.isSmoke.output.test == true
env:
GITHUB_TOKEN: ${{ secrets.DEV_TOKEN }}
with:
filename: .github/ISSUE_TEMPLATE/smoke-test.md
but when the commit message of smoke_test is used it bypasses create-an-issue:
and I'm basing my condition after reading "Run github actions step based on output condition" and reading:
Contexts
Expressions
Using conditions to control job execution
Can a condition come before a step and/or what is the correct way to run a step based off the previous step?
You are looking for job outputs, which allow you to send data to the following jobs.
This is the workflow I'm currently using on GitHub
name: Windows10 - CI
on: [ push ]
jobs:
run-test:
runs-on: [ self-hosted, windows, DO ]
steps:
- uses: actions/checkout#v2
with:
clean: false
- name: Run nds2 CI - Sanity Test
if: github.ref == 'refs/heads/master'
run: cd c:\actions-runner\_work\nds2\nds2 ; python3 ci_host.py --master
- name: Run nds2 CI - Build Installer
if: github.ref != 'refs/heads/master'
run: cd c:\actions-runner\_work\nds2\nds2 ; python3 ci_host.py
I have a windows 10 computer which listens for an incoming job by using the GitHub runner.
Upon an incoming job, if a push is being made to the master branch the script ci_host.py is being run with the '--master' flag which spins up a VM and runs multiple tests on it. Eventually, at the end of the tests, the script restores the VM to a pre-configured snap shot.
So basically what I'm trying to achieve is, when the job is being canceled through the GitHub actions web interface then the script which handles the test is being canceled mid-job and doesn't have the chance to restore the VM to its prior clean state (snapshot).
How can I run a script which will be run at the end of the workflow even if the job was canceled?
So no matter what happens my VM could be restored to its clean state
Thanks in advance for your help :)
You can use Job status check functions to execute a step depending on what happened on the job before it (or not):
success: Returns true when none of the previous steps have failed or been canceled.
always: Always returns true, even when canceled. A job or step will not run when a critical failure prevents the task from running. For example, if getting sources failed.
cancelled: Returns true if the workflow was canceled.
failure: Returns true when any previous step of a job fails.
Example of use:
steps:
...
- name: Execute if the job succeeded
if: ${{ success() }}
- name: Execute if the job failed
if: ${{ failure() }}
- name: Execute if the job was cancelled
if: ${{ cancelled() }}
- name: Always execute
if: ${{ always() }}
For (mainly) pedagogical reasons, I'm trying to run this workflow in GitHub actions:
name: "We 🎔 Perl"
on:
issues:
types: [opened, edited, milestoned]
jobs:
seasonal_greetings:
runs-on: windows-latest
steps:
- name: Maybe greet
id: maybe-greet
env:
HEY: "Hey you!"
GREETING: "Merry Xmas to you too!"
BODY: ${{ github.event.issue.body }}
run: |
$output=(perl -e 'print ($ENV{BODY} =~ /Merry/)?$ENV{GREETING}:$ENV{HEY};')
Write-Output "::set-output name=GREET::$output"
produce_comment:
name: Respond to issue
runs-on: ubuntu-latest
steps:
- name: Dump job context
env:
JOB_CONTEXT: ${{ jobs.maybe-greet.steps.id }}
run: echo "$JOB_CONTEXT"
I need two different jobs, since they use different context (operating systems), but I need to get the output of a step in the first job to the second job. I am trying with several combinations of the jobs context as found here but there does not seem to be any way to do that. Apparently, jobs is just the name of a YAML variable that does not really have a context, and the context job contains just the success or failure. Any idea?
Check the "GitHub Actions: New workflow features" from April 2020, which could help in your case (to reference step outputs from previous jobs)
Job outputs
You can specify a set of outputs that you want to pass to subsequent jobs and then access those values from your needs context.
See documentation:
jobs.<jobs_id>.outputs
A map of outputs for a job.
Job outputs are available to all downstream jobs that depend on this job.
For more information on defining job dependencies, see jobs.<job_id>.needs.
Job outputs are strings, and job outputs containing expressions are evaluated on the runner at the end of each job. Outputs containing secrets are redacted on the runner and not sent to GitHub Actions.
To use job outputs in a dependent job, you can use the needs context.
For more information, see "Context and expression syntax for GitHub Actions."
To use job outputs in a dependent job, you can use the needs context.
Example
jobs:
job1:
runs-on: ubuntu-latest
# Map a step output to a job output
outputs:
output1: ${{ steps.step1.outputs.test }}
output2: ${{ steps.step2.outputs.test }}
steps:
- id: step1
run: echo "test=hello" >> $GITHUB_OUTPUT
- id: step2
run: echo "test=world" >> $GITHUB_OUTPUT
job2:
runs-on: ubuntu-latest
needs: job1
steps:
- run: echo ${{needs.job1.outputs.output1}} ${{needs.job1.outputs.output2}}
Note the use of $GITHUB_OUTPUT, instead of the older ::set-output now (Oct. 2022) deprecated.
To avoid untrusted logged data to use set-state and set-output workflow commands without the intention of the workflow author we have introduced a new set of environment files to manage state and output.
Jesse Adelman adds in the comments:
This seems to not work well for anything beyond a static string.
How, for example, would I take a multiline text output of step (say, I'm running a pytest or similar) and use that output in another job?
either write the multi-line text to a file (jschmitter's comment)
or base64-encode the output and then decode it in the next job (Nate Karasch's comment)
Update: It's now possible to set job outputs that can be used to transfer string values to downstream jobs. See this answer.
What follows is the original answer. These techniques might still be useful for some use cases.
Write the data to file and use actions/upload-artifact and actions/download-artifact. A bit awkward, but it works.
Create a repository dispatch event and send the data to a second workflow. I prefer this method personally, but the downside is that it needs a repo scoped PAT.
Here is an example of how the second way could work. It uses repository-dispatch action.
name: "We 🎔 Perl"
on:
issues:
types: [opened, edited, milestoned]
jobs:
seasonal_greetings:
runs-on: windows-latest
steps:
- name: Maybe greet
id: maybe-greet
env:
HEY: "Hey you!"
GREETING: "Merry Xmas to you too!"
BODY: ${{ github.event.issue.body }}
run: |
$output=(perl -e 'print ($ENV{BODY} =~ /Merry/)?$ENV{GREETING}:$ENV{HEY};')
Write-Output "::set-output name=GREET::$output"
- name: Repository Dispatch
uses: peter-evans/repository-dispatch#v1
with:
token: ${{ secrets.REPO_ACCESS_TOKEN }}
event-type: my-event
client-payload: '{"greet": "${{ steps.maybe-greet.outputs.GREET }}"}'
This triggers a repository dispatch workflow in the same repository.
name: Repository Dispatch
on:
repository_dispatch:
types: [my-event]
jobs:
myEvent:
runs-on: ubuntu-latest
steps:
- run: echo ${{ github.event.client_payload.greet }}
In my case I wanted to pass an entire build/artifact, not just a string:
name: Build something on Ubuntu then use it on MacOS
on:
workflow_dispatch:
# Allows for manual build trigger
jobs:
buildUbuntuProject:
name: Builds the project on Ubuntu (Put your stuff here)
runs-on: ubuntu-latest
steps:
- uses: actions/checkout#v2
- uses: some/compile-action#v99
- uses: actions/upload-artifact#v2
# Upload the artifact so the MacOS runner do something with it
with:
name: CompiledProject
path: pathToCompiledProject
doSomethingOnMacOS:
name: Runs the program on MacOS or something
runs-on: macos-latest
needs: buildUbuntuProject # Needed so the job waits for the Ubuntu job to finish
steps:
- uses: actions/download-artifact#master
with:
name: CompiledProject
path: somewhereToPutItOnMacOSRunner
- run: ls somewhereToPutItOnMacOSRunner # See the artifact on the MacOS runner
It is possible to capture the entire output (and return code) of a command within a run step, which I've written up here to hopefully save someone else the headache. Fair warning, it requires a lot of shell trickery and a multiline run to ensure everything happens within a single shell instance.
In my case, I needed to invoke a script and capture the entirety of its stdout for use in a later step, as well as preserve its outcome for error checking:
# capture stdout from script
SCRIPT_OUTPUT=$(./do-something.sh)
# capture exit code as well
SCRIPT_RC=$?
# FYI, this would get stdout AND stderr
SCRIPT_ALL_OUTPUT=$(./do-something.sh 2>&1)
Since Github's job outputs only seem to be able to capture a single line of text, I also had to escape any newlines for the output:
echo "::set-output name=stdout::${SCRIPT_OUTPUT//$'\n'/\\n}"
Additionally, I needed to ultimately return the script's exit code to correctly indicate whether it failed. The whole shebang ends up looking like this:
- name: A run step with stdout as a captured output
id: myscript
run: |
# run in subshell, capturiing stdout to var
SCRIPT_OUTPUT=$(./do-something.sh)
# capture exit code too
SCRIPT_RC=$?
# print a single line output for github
echo "::set-output name=stdout::${SCRIPT_OUTPUT//$'\n'/\\n}"
# exit with the script status
exit $SCRIPT_RC
continue-on-error: true
- name: Add above outcome and output as an issue comment
uses: actions/github-script#v5
env:
STEP_OUTPUT: ${{ steps.myscript.outputs.stdout }}
with:
github-token: ${{ secrets.GITHUB_TOKEN }}
script: |
// indicates whather script succeeded or not
let comment = `Script finished with \`${{ steps.myscript.outcome }}\`\n`;
// adds stdout, unescaping newlines again to make it readable
comment += `<details><summary>Show Output</summary>
\`\`\`
${process.env.STEP_OUTPUT.replace(/\\n/g, '\n')}
\`\`\`
</details>`;
// add the whole damn thing as an issue comment
github.rest.issues.createComment({
issue_number: context.issue.number,
owner: context.repo.owner,
repo: context.repo.repo,
body: comment
})
Edit: there is also an action to accomplish this with much less bootstrapping, which I only just found.
2022 October update: GitHub is deprecating set-output and recommends to use GITHUB_OUTPUT instead. The syntax for defining the outputs and referencing them in other steps, jobs.
An example from the docs:
- name: Set color
id: random-color-generator
run: echo "SELECTED_COLOR=green" >> $GITHUB_OUTPUT
- name: Get color
run: echo "The selected color is ${{ steps.random-color-generator.outputs.SELECTED_COLOR }}"