Mistakes in production operations can immediately impact customer experience, so Shoreline was designed to maximize safe operations.
Automate mundane work to eliminate human errors
Using pre-approved actions eliminates the potential for human errors that can easily happen when operating quickly under the pressure of an outage
Shoreline replaces the need for SSH access to production boxes
Shoreline’s network and data are secure
Shoreline provides authorization of user requests by role:
- View metrics and alarms (default role)
- Execute actions created by others
- Create derived metrics, alarms, actions, and bots
- Administer the system, including user management
By separating roles for executing actions vs. creating actions, an ops team can manage the operations that can be executed by L1 and L2 support teams.
Run Terraform definitions through CI/CD pipelines
Shoreline includes a verified Terraform provider to bring Gitops best practices to production operations
All artifacts (metrics, alarms, actions, bots and scripts) can be published via a “Terraform apply,” while disallowing artifact definition in production
Ensure that all components of Shoreline automations go through the same CI/CD processes, including version control and code reviews before deployment
Define circuit breakers and blast radius limits
Limit the scope of resources affected by a single command by setting a blast radius by role (max x nodes)
Use circuit breakers to limit frequency of automated actions (max x/hour)
Provide privilege escalation for “break glass” scenarios
