Presenting some of my recent work on “Agile ways of working in service delivery” (essentially I was talking about DevOps). In this talk I described some of the software engineering practices I’d been using to improve the life of our Operations teams.

Touching on aspects of the system architecture of an automation engine I’d built to handle some common Operational processes, this included:

• Log gathering
• System stop/starts that handled Veritas cluster manager
• App and DB deployments/upgrades
• DB backups/restores
• App and DB Cluster failover
• App and DB Site failover

Essentially this was a kind of “poor man’s Ansible”, before I knew ansible existed. It was built in perl and needed to be distributed to each node under management - early days in my own personal IaC journey. Using the go script principles it created a common language between Development and Operations (and other teams along the way). As our product was called TOPS, I named this the TOPS Ops Console.

• thomson reuters
• unconference
• agile
• service delivery
• devops