Context
A hybrid estate in a large international retail group: an Azure and Google Cloud foundation of about 900 applications, and about 300 on-premises servers.
Challenge
Reducing level 1 and 2 support tasks, speeding up remediation and monitoring the whole estate.
What I did
- CloudOps / SRE from 2018 to 2021: level 3 support, production incidents, post-mortems, on-call; operations automation with Ansible.
- I designed and deployed the observability chain across the whole estate: Prometheus and Thanos, Grafana, Telegraf, AlertManager.
- A patch management architecture for about 300 on-premises servers, with Azure Arc and Azure Update Management.
Outcome
Level 1 and 2 support tasks fell by more than 50%, and remediation is faster. The whole estate is covered by the same observability chain.
Stack
- Ansible
- Prometheus
- Thanos
- Grafana
- Telegraf
- AlertManager
- Azure Arc
- Azure Update Management