gitlabrunner-healthcheck

Made this script which will ping alertmanager any time a runner spends too long not contacting gitlab-ci.

Needs 3 env variables:

• GITLAB_URL - the url to the instance being used excluding the api part, e.g https://gitlab.dotmesh.com
• GITLAB_TOKEN - access token with access to the API
• ALERTMANAGER_URL - url of the alertmanager instance to ping, excluding the api part

Optional:

• ALERTMANAGER_USERNAME basic auth username for alertmanager
• ALERTMANAGER_PASSWORD basic auth password for alertmanager
• THRESHOLD_BEFORE_ALERT - a golang standard duration string. How long between now and the contactedAt time is acceptable? Defaulted to 20m.
• PAUSE - golang standard duration string. Time between checking the api on cycle. Defaulted to 10m.