GoDaddy: Sensu as a Service
Role: Sr Dir of SRE (Observability & ITSM)
Overview: Overhauled synthetic monitoring by replacing our home-grown Nagios fork with Sensu. This enabled developers to deploy an instance for their monitoring needs and have a core set of infrastructure and security checks enabled for their application by default as well as develop their own.
Situation: GoDaddy’s existing synthetic monitoring, based on a homegrown Nagios fork, was difficult to scale, manage, and customize. Developers faced delays and dependencies on a central team for deploying or modifying checks, hindering agility.
Task: To implement a modern, scalable, and self-service synthetic monitoring platform. This involved selecting a suitable tool (Sensu), designing a service model, and enabling developers to easily deploy and manage their own application-specific checks while ensuring a core set of infrastructure and security checks were applied by default.
Action:
- Led the replacement of the homegrown Nagios fork with Sensu.
- Designed and implemented “Sensu as a Service,” allowing internal teams to self-service their monitoring needs.
- Enabled developers to deploy their own Sensu instances for application-specific monitoring.
- Ensured a core set of infrastructure and security checks were automatically enabled for all applications.
- Facilitated the development of custom checks by engineering teams.
- Integrated Sensu with Kafka for big data integration and broader event visibility.
Tech Stack Used: Sensu, RabbitMQ, Redis, Kafka (Big Data integration).
Result: Allowed internal teams to self-service their monitoring, significantly reducing the time required for modifications and distributing monitoring responsibilities across engineering teams.
Context: This initiative modernized GoDaddy’s approach to synthetic monitoring, aligning with DevOps principles by giving developers more ownership and control over their application’s observability. It improved the speed and effectiveness of monitoring across the company.
Visuals:
