Toolbox Reference#

This document describes additional open source command line tools that are provided by Semaphore and available in all VMs.

cache#

Semaphore cache tool helps optimize CI/CD runtime by reusing files that your project depends on but are not part of version control. You should typically use caching to:

  • Reuse your project's dependencies, so that Semaphore fetches and installs them only when the dependency list changes.
  • Propagate a file from one block to the next.

Cache is created on a per project basis and available in every pipeline job. All cache keys are scoped per project. Total cache size is 9.6GB.

The cache tool uses key-path pairs for managing cached archives. An archive can be a single file or a directory.

cache supports the following options:

cache store key path#

Examples:

cache store our-gems vendor/bundle
cache store gems-$SEMAPHORE_GIT_BRANCH vendor/bundle
cache store gems-$SEMAPHORE_GIT_BRANCH-revision-$(checksum Gemfile.lock) vendor/bundle

The cache store command archives a file or directory specified by path and associates it with a given key.

As cache store uses tar, it automatically removes any leading / from the given path value. Any further changes of path after the store command completes will not be automatically propagated to cache. The command always passes, i.e. exits with return code 0.

In case of insufficient disk space, cache store frees disk space by deleting the oldest keys.

cache restore key[,second-key,...]#

Examples:

cache restore our-gems
cache restore gems-$SEMAPHORE_GIT_BRANCH
cache restore gems-$SEMAPHORE_GIT_BRANCH-revision-$(checksum Gemfile.lock),gems-master

Restores an archive which partially matches any given key. In case of a cache hit, archive is retrieved and available at its original path in the job environment. Each archive is restored in the current path from where the function is called.

In case of cache miss, the comma-separated fallback takes over and command looks up the next key. If no archives are restored command exits with 0.

cache has_key key#

Example:

cache has_key our-gems
cache has_key gems-$SEMAPHORE_GIT_BRANCH
cache has_key gems-$SEMAPHORE_GIT_BRANCH-revision-$(checksum Gemfile.lock)

Checks if an archive with provided key exists in cache. Command passes if key is found in the cache, otherwise it fails.

cache list#

Example:

cache list

Lists all cache archives for the project.

cache delete key#

Example:

cache delete our-gems
cache delete gems-$SEMAPHORE_GIT_BRANCH
cache delete gems-$SEMAPHORE_GIT_BRANCH-revision-$(checksum Gemfile.lock)

Removes an archive with given key if it is found in cache. The command always passes.

cache clear#

Example:

cache clear

Removes all cached archives for the project. The command always passes.

Note that in all commands of cache, only cache has_key command can fail (exit with non-zero status).

Cache Dependencies#

The cache tool depends on the following environment variables which are automatically set in every job environment:

  • SEMAPHORE_CACHE_URL: stores the IP address and the port number of the cache server (x.y.z.w:29920).
  • SEMAPHORE_CACHE_USERNAME: stores the username that will be used for connecting to the cache server (5b956eef90cb4c91ab14bd2726bf261b).
  • SEMAPHORE_CACHE_PRIVATE_KEY_PATH: stores the path to the SSH key that will be used for connecting to the cache server (/home/semaphore/.ssh/semaphore_cache_key).

Additionally, cache uses tar to archive the specified directory or file.

checkout#

The GitHub repository used in a Semaphore project is not automatically cloned for reasons of efficiency.

The libcheckout script includes the implementation of a single function named checkout() that is used for making available the entire GitHub repository of the running Semaphore project to the VM used for executing a job of the pipeline. The checkout() function is called as checkout from the command line.

Shallow clone#

By default, the implementation of the checkout command uses shallow clone during the clone operation.

With shallow clone, you get all the files in your repository and a small number of recent revisions. This makes the checkout process faster comparing to downloading the entire history of the repository.

Note that some deployment scenarios, like Heroku, require presence of the full Git history.

If you'd like to do a full clone, execute checkout with the --use-cache flag, or use the following code snippet:

checkout
git fetch --unshallow
git config remote.origin.fetch '+refs/heads/*:refs/remotes/origin/*'
git fetch --all

The checkout() function of the libcheckout script depends on the following three Semaphore environment variables:

  • SEMAPHORE_GIT_URL: This environment variable holds the URL of the GitHub repository that is used in the Semaphore project (git@github.com:mactsouk/S1.git).

  • SEMAPHORE_GIT_DIR: This environment variable holds the UNIX path where the GitHub repository will be placed on the VM (/home/semaphore/S1).

  • SEMAPHORE_GIT_SHA: This environment variable holds the SHA key for the HEAD reference that is used when executing git reset -q --hard.

All these environment variables are automatically defined by Semaphore.

The --use-cache flag#

The checkout command supports the --use-cache flag. The purpose of that flag is to tell checkout to get the contents of the GitHub repository from the Semaphore Cache server instead of the GitHub servers because it is faster.

If there is no cache entry for the active GitHub repository, the functionality of the --use-cache flag will create it.

When using the --use-cache flag, checkout supports the following environment variables:

  • SEMAPHORE_GIT_CACHE_AGE: This environment variable specifies how often the cache for that GitHub repository will get updated. Its value is always in seconds and by default it is 259200, which is 3 days. The value that you are going to choose depends on your project and how ofter it gets updated.

  • SEMAPHORE_GIT_CACHE_KEEP: Each time there is an update to the cache, the key in the Cache server is also being updated. Before updating the key, the previous key that is related to the current project is deleted. This environment variable tells Semaphore Cache server to keep a history. The default value of SEMAPHORE_GIT_CACHE_KEEP is 0.

Note: When used with the --use-cache flag, checkout does not use shallow clone.

Examples:

checkout

Notice that the checkout command automatically changes the current working directory in the Operating System of the VM to the directory defined in the SEMAPHORE_GIT_DIR environment variable.

The following command will tell checkout to use the Semaphore Cache server to get the contents of the GitHub repository instead of using GitHub server:

checkout --use-cache

If you set SEMAPHORE_GIT_CACHE_KEEP to 1 then it will keep two copies in the Semaphore Cache server: the active one and the previous one.

If you set SEMAPHORE_GIT_CACHE_AGE=86400, then the cache for the GitHub repository will get updated after 1 day.

checksum#

The libchecksum scripts provides the checksum command. checksum is useful for tagging artifacts or generating cache keys. It takes a single argument, a file path, and outputs an md5 hash value.

Examples:

$ checksum package.json
3dc6f33834092c93d26b71f9a35e4bb3

retry#

The retry script is used for retrying a command for a given amount of times at given time intervals – waiting for resources to become available or waiting for network connectivity issues to be resolved is the main reason for using the retry command.

The retry bash script supports two command line parameters:

  • -t or --times: this is used for defining the number of retries before giving up. The default value is 3.
  • -s or --sleep: this is used for defining the time interval between retries The default value is 0.

The retry bash script only depends on the /bin/bash executable.

Examples:

$ retry bundle install

$ retry --times 10 go get ./...

$ retry --times 3 --sleep 5 make install.deps

install-package#

The install-package script can be used to cache missing packages and their dependencies. It works by creating a "temporary package cache" folder (if its not present), downloading packages (that are not present),
installing the packages from the "temporary package cache", then uploading the folder to the projects cache.

Usage:

install-package [--update|--skip-update|--update-new] pkg1[=version] [pkg2[=version]]

---update forces repository list update before installing packages.
---skip-update skips repository list update before installing packages.
---update-new updates only repository lists added in the last hour.

-pkg1 is the desired package which accepts the =version parameter. Multiple package names with versions are supported.

See also#