Environment variables

Environment variables are pairs of key and value that can be used to customize the build process and store sensitive data such as access details to deployment servers.

Levels

Environment Variables can be defined on 4 levels:

1. Workspace – available for all projects and pipelines. Configured on the dashboard view.
2. Project – available for all pipelines in the project. Configured on the project view.
3. Pipeline – available only for the pipeline in which it was defined. Configured in the Variables tab of the pipeline.
4. Action – available only for the action in which it was defined. Configured in the Variables tab of the action.

Configuration

When adding a new variable, the following fields are available:

• Scope – defines the level of the variable
• Name – the name of the variable, e.g. $SFTP
• Value – the value conveyed by the name, e.g. 192.168.0.100

Additional options let you hide the value of the variable in the logs, and mark the variable as settable, allowing you to change its value and pass it to the next action or pipeline run.

Example usage

Let's configure some variables to encrypt access to an SFTP server:

1. Go to the Variables tab in your deployment pipeline.

2. Add the variables and set your server credentials as values. For example:

   • $SFTP_HOST / 192.168.0.104
   • $SFTP_LOGIN / admin
   • $SFTP_PASS / buddy1234

Variables in pipeline

3. Now you can use the variables to store access credentials to your deployment server:

Storing credentials with variables

Supported fields

To check if the input supports ENV VARs, click it and see if the dollar icon appears:

Variable icon in action input

Default environment variables

Buddy populates a number of environment variables during each pipeline run, for example, the branch from which you deploy, and the ID of the pipeline run. You can use them in webhooks, notifications, conditional runs, and any place that require passing CI/CD-related information.

Combining into phrases

Variables can be combined with other words to create longer phrases than can be used, for example, to automatically name your infrastructure, releases, or sandboxes. However, you need to obey the following rules:

• Before the variable: any character is allowed.
• After the variable: letters, numbers, and footers (_) are not allowed.

Correct usage:

release-$BUDDY_EXECUTION_ID-staging
release$BUDDY_EXECUTION_ID:staging
release_$BUDDY_EXECUTION_ID[staging]

Incorrect usage:

release-$BUDDY_EXECUTION_ID_staging (footer)
release-$BUDDY_EXECUTION_IDstaging (letter)
release-$BUDDY_EXECUTION_ID1staging (number)

Passing variables

Actions

Below you can see a pipeline that saves action logs to a variable and sends it in a notification.

1. The first action builds the application and saves the log output to logs.txt:

Primary action (always run)

2. In case the build fails, the Local Shell action assigns the content of logs.txt to a variable using export LOGS=$(cat log.txt). The logs are then sent to a Slack channel with $LOGS as the message content:

Conditional actions (run on failure only)

Pipeline runs

You can pass the values of variables between pipeline runs as well. The values can be saved in one run and then used in another one, e.g. you can set a variable with the version type and increment it in every run.

Pipelines

You can pass a variable between pipelines using the Trigger pipeline action:

1. Go into the Variables tab in the action details.
2. Define the name of the variable and the value to pass:

Variables tab

Resetting

To reset settable variables between pipeline runs, you can add an action with a command that will clear the value of the variable (preferably at the beginning or at the end of the pipeline):

export mySettableEnv=$

Reset variable command

On pipeline start

You can parameterize your run using the Pass arguments action. It will pause the pipeline and wait for your input to proceed:

Pass arguments window