aboutsummaryrefslogtreecommitdiffstats
path: root/.github (follow)
Commit message (Collapse)AuthorAgeLines
...
* CI: rename jobsGravatar MarkKoz2021-12-27-10/+6
|
* CI: clean up Docker stuff in self-hosted runnerGravatar MarkKoz2021-12-27-8/+14
| | | | | | | | | | | | | | Use docker-compose run instead of docker-compose up. This is more appropriate since the container is only needed for one command. The latter was actually starting the whole snekbox server. Furthermore, the former has the --rm option to remove the container when the command finishes. As an extra precaution, use docker-compose down in the self-hosted runner to also remove images, volumes, networks, and any other containers that were somehow missed. Removing images will also prevent the disk usage from building up. This is not necessary for the GH-hosted runner since a new VM is used for each run.
* CI: fix pre-commit logs step conditionGravatar MarkKoz2021-12-27-1/+1
| | | | The step was running even if the pre-commit hooks step never ran.
* CI: fix flake8 being skippedGravatar MarkKoz2021-12-27-0/+7
|
* CI: run linters outside the containerGravatar MarkKoz2021-12-27-36/+44
| | | | | | | | Remove the dependency on the container so the lint job can run in parallel with the build job. More time has to be spent installing Python dependencies, but this is made up for by not having to download and load the image artefact in addition to not having to wait for the build job.
* CI: specify service name to coverallsGravatar MarkKoz2021-12-27-1/+1
| | | | See https://github.com/TheKevJames/coveralls-python/issues/240
* CI: remove SHA from image build GHA cache scopeGravatar MarkKoz2021-12-26-1/+1
| | | | | | Unlike the cache action, the build-push action's GHA cache feature seems to only do an exact comparison for the scope. Thus, new commits lead to cache misses.
* CI: combine coverage data from all matrix jobsGravatar MarkKoz2021-12-26-21/+38
|
* CI: pass artefact name as an input to workflowsGravatar MarkKoz2021-12-26-10/+28
| | | | | | Make the artefact and file names identical to simplify things. The artefact name doesn't have to be unique anyway since it can only be downloaded by the same workflow run.
* CI: use buildx's integrated GH Actions cache featureGravatar MarkKoz2021-12-26-12/+18
|
* CI: use build-push-action to output an image tarballGravatar MarkKoz2021-12-26-4/+1
| | | | | | | load: true was already creating a tarball, but it was getting immediately loaded. Since no other Docker builds run in this job, it's useless to load it. The action can still be leveraged to create the tarball instead of manually invoking `docker save`.
* CI: use reusable workflows as jobs in main workflowGravatar MarkKoz2021-12-26-242/+27
|
* CI: create a reusable deployment workflowGravatar MarkKoz2021-12-26-0/+96
|
* CI: create a reusable test workflowGravatar MarkKoz2021-12-26-0/+74
|
* CI: create a reusable lint workflowGravatar MarkKoz2021-12-26-0/+52
|
* CI: create a reusable build workflowGravatar MarkKoz2021-12-26-0/+88
|
* Do not run Kubernetes deployment on self-hosted runnerGravatar Joe Banks2021-12-26-1/+1
|
* CI: disable setup-python action on self-hosted runnerGravatar MarkKoz2021-12-25-1/+1
| | | | | | Python 3.9 is available on the self-hosted runner and is sufficient to run coveralls. Trying to get the setup-python action supported on the self-hosted runner proved to be problematic.
* CI: create a matrix with a self-hosted runner with cgroupv2Gravatar MarkKoz2021-12-21-6/+18
| | | | | | The self-hosted runner has cgroupv2 enabled. It's only needed to run the tests on a cgroupv2 system. Only lint, push the image, and deploy it on one runner to avoid redundancy.
* Remove myself from the codeownersGravatar Matteo Bertucci2021-12-10-1/+1
|
* Bumps Python Version To 3.10Gravatar Hassan Abouelela2021-10-11-1/+1
| | | | Signed-off-by: Hassan Abouelela <[email protected]>
* Add a step to lint-test to show pre-commit logsGravatar ToxicKidz2021-05-18-0/+7
|
* Move from GHCR_TOKEN to GITHUB_TOKEN in Docker publish stageGravatar Joe Banks2021-04-05-1/+1
|
* master => mainGravatar Joe Banks2021-03-13-6/+6
|
* Docker: improve caching & install numpy in containerGravatar MarkKoz2021-03-07-6/+1
| | | | | | | | | | | | | CI was building the image twice: once with dev dependencies and again without. Separating the pipenv command into separate layers allows the second build in CI to take advantage of the cache for the base dependencies that it will share across both builds. Install numpy along with the dev dependencies within the container. Previously it was installed in CI only, but this meant extra work for those running tests locally. Install numpy to the correct site.
* CI: fix test step not failing when tests failGravatar MarkKoz2021-02-04-3/+6
| | | | | Generating the report in the same step resulted in the report exit code overriding the exit code of the test runner.
* CI: use Docker Compose to run the containerGravatar MarkKoz2021-02-04-18/+8
| | | | | | | | | | | | | | | | | | | | | Avoid redundant specification of Docker settings. The compose file is set up to build all stages. This makes sense for local development; both an interactive shell and running the webserver are useful. Therefore, the image built is tagged "snekbox:dev". However, CI does not need to run a webserver. It is therefore sufficient for it to only build to the venv stage, and it does exactly that. The image in CI is tagged as "snekbox-venv:<git sha>". To facilitate the discrepancy in image tags, the suffix for the image tag can be set with the new IMAGE_SUFFIX environment variable. Docker Compose will use this to determine the image from which to create a container. A TTY needs to be allocated to prevent the container from exiting immediately after starting. This is probably because the entrypoint is Python (inherited from the base image), and the REPL relies on a TTY.
* Merge PR #88 - use protobuf to parse configGravatar Mark2021-02-02-1/+0
|\
| * Replace nsjpy alias with a Python scriptGravatar MarkKoz2021-01-10-1/+0
| | | | | | | | | | | | | | | | | | The Python script uses the same underlying code Falcon uses to invoke nsjail. It allows for the omission of redundant shell code that set up cgroups and nsjail args. This is also a step towards removing dependence on shell scripts and thus resolving #73.
* | CI: don't do a user install for pre-commitGravatar MarkKoz2021-01-12-1/+2
|/
* CI: install git in the containerGravatar MarkKoz2021-01-09-0/+6
| | | | Pre-commit requires git.
* CI: run pre-commit hooksGravatar MarkKoz2021-01-09-0/+6
|
* CI: improve readability of some run commandsGravatar MarkKoz2021-01-09-5/+7
|
* CI: install numpyGravatar MarkKoz2021-01-09-1/+2
| | | | One of the unit tests depends on numpy.
* Fix getting Git SHA in workflowGravatar ks1292020-12-20-1/+1
|
* Create workflow for Sentry release creationGravatar ks1292020-12-20-0/+24
|
* Add Git SHA build argument to Docker build in workflowGravatar ks1292020-12-20-0/+2
|
* Add @Akarys42 to the codeownersGravatar Matteo Bertucci2020-11-27-1/+1
|
* Add Mark to code ownersGravatar Mark2020-11-24-2/+2
|
* Update CODEOWNERSGravatar Dennis Pham2020-11-24-2/+2
|
* Update CODEOWNERSGravatar Joe Banks2020-11-24-1/+2
|
* Update CODEOWNERSGravatar Joe Banks2020-11-24-1/+1
|
* Add production_build == 'true' condition on deploy.Gravatar Leon Sandøy2020-11-19-0/+2
|
* Add Kubernetes auto-deploy to the CI build stage.Gravatar Leon Sandøy2020-11-19-0/+15
|
* Add a Kubernetes deployment manifest.Gravatar Leon Sandøy2020-11-19-2/+0
| | | | Simple stuff. Basically copy-paste from site.
* Use repository caching for master buildGravatar Sebastiaan Zeeff2020-11-18-37/+30
| | | | | | | | One problem that our master builds may have is that they retain more and more layers of old builds, as there is no easy way of purging them from the cache. As such master cache would not have benefits over using repository-based caching, I've removed persistent local caching for non-PR builds.
* Rectify production branch and clean up formattingGravatar Sebastiaan Zeeff2020-11-18-14/+15
|
* Remove escape sequence and improve job nameGravatar Sebastiaan Zeeff2020-11-18-2/+2
| | | | | | I accidentally escaped a single quote in a run command; I've removed it now. I also changed the job name to `lint-test-build-push` to better reflect the contents of the job.
* Migrate build pipeline to Github ActionsGravatar Sebastiaan Zeeff2020-11-18-0/+207
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I've migrated the build pipeline to GitHub Actions and changed the container registry to GitHub Container Registry. In the process, I've made some changes to our docker setup and caching: - We are now using a single multi-stage Dockerfile Instead of three separate dockerfiles, we are now using a single multi-stage Dockerfile that can be used to build the three images we want using build targets. In part, this is because we're now using the docker buildx build action currently recommended by docker. This new engine runs in a sandboxed mode, meaning that while it can export built images to `docker` running in the host, it cannot import local images from it to base builds on. - Docker builds are now cached within GitHub Actions The builds are now cached using the GitHub Actions cache of the build cache directory. The cache keys try to match a cache generated by a build that matches the current build as closely as possible. In case of a cache miss, we fall back to caching from the latest image pushed to the container repository. - The `base` and `venv` images now have an inline cache manifest In order to fall back intelligently to caching from the repository, the final build and push action for the `base` and `venv` images includes an "inline" cache manifest. This means that the build process can inspect, without pulling, if it makes sense to pull layers to speed up the build. The other options, pushing a cache manifest separately (not inline), is currently not supported by GHCR. The custom caching script has been removed. - Linting errors are now added as GitHub Actions annotations Just like for some of our other pipelines, linting now generates annotations if linting errors are observed. - Coverage is pushed to coveralls.io A coverage summary is now pushed to coveralls.io. Each CI run will get a unique job that's linked in the CI output. If the run is attached to a PR, coveralls.io will automatically add a check link with the coverage result to the PR as well. - The README.md, Pipfile, docker-compose, and scripts have been updated As we now need to pull from and link to the GHCR, I've updated the other files to reflect these changes, including Pipfile run commands. I've also changed the CI badge and added a coveralls.io badge.
* Update CODEOWNERSGravatar Joseph2020-02-02-1/+1
|
generated by cgit v1.2.3 (git 2.39.1) at 2025-10-19 14:16:03 +0000