CI/CD Environment Variables Missing Results in Failed Runs on Serverless

[jamestkelly]

Description

As identified by @houstonpearse, previous versions of the lambda, specifically version 10, have no environment (env) variables. At this current point, this is the most likely reason as to why previous deployments via the pipeline have failed to fire the cron schedule correctly. This can be seen from the following snapshot of the serverless dashboard where the deployment coincided with a failure for Harold to fire on schedule.

Resolution

Simple Fix

For the easier implementation to determine whether or not it is the environment variables leading to the cron schedule not firing. @houstonpearse suggested the following:

• We make a commit, let the CI/CD deploy, then add the env variables in the AWS console, then hopefully if a new commit is made and when the CI/CD redeploys it the env variables will persist.

Long-term Solution

• Build an ansible script to be called via GitHub Workflow to deploy Harold to serverless, etc with the env variables.