TryCI

A script to emulate a soft-dev buildbot CI job run.

If a build job fails, TryCI can conduct a post-mortem (using the --post-mortem flag) by attaching a shell which allows you to prod around inside the image.

This must be run from within a softdev repo which uses buildbot for CI, in the same directory as the repo's .buildbot.sh file.

Example usage

The example below shows how a CI build for Alloy (located on my local machine) can be run on remotely on [email protected] (with the bencher16 alias in my ~/.ssh/config):

cd /path/to/alloy
tryci --remote bencher16 --post-mortem

The --remote arg specifies the ssh host where the build should be run.

The --post-mortem flag attaches a shell to the build if it fails so that you can prod around inside the container.

The docker image and container for the build are removed before the script exits.

Checking out a specific CI job

By default, TryCI uses the working tree as the build context for the CI job. This works well for iterating on features, but can be annoying when you want to debug a remote CI job which failed.

You can specify which version of the repo you want to TryCI with the -c/--checkout <ref> option. <ref> can be either a branch, commit, or tag of the local repository; or a remote URL of the git repository. Here are some examples:

# run tryci on the local HEAD
tryci -c HEAD

# run tryci on the local feature-branch
tryci -c feature-branch

# run tryci on the master branch of jacob-hughes/yk
tryci -c https://github.com/jacob-hughes/yk

# run tryci on the 'trying' branch of ykjit/yk
tryci -c https://github.com/ykjit/yk#trying

# run tryci on a specified commit ykjit/yk
tryci -c https://github.com/ykjit/yk#0a6902a

This works with the --post-mortem flag, so -- provided docker is installed -- you can even use TryCI to prod around in a CI job on machines which you haven't cloned the original repo or setup to develop on.

Using bindmounts for faster build times

Sometimes a CI build will depend on a submodule or external program which the .buildbot.sh script builds from source each time. For large projects (like ykllvm, which can take up to an hour to build) this can make build times significantly slower.

Instead, you can mount arbitrarily many prebuilt programs from the host to be used inside the docker container using -m/--mount flag. You can specify multiple bind-mounts by repeating the option or by providing a comma-separated list:

-m /src:/dst -m /foo:/bar:ro
-m /src:/dst,/foo:/bar:ro

If using this with the --remote flag, you must ensure the directory you wish to bind-mount exists on the remote machine -- it does not scp / rsync it over.

Troubleshooting

• Docker must be installed on both the local machine and remote machine used to compile and run the build.
• Your user account on both the local and remote machine must be a member of the docker group.
• If running CI on a remote machine (--remote <server_name>) you must use ssh passwordless login (using public/private key pair).