Startup Systems Engineer's Instruction manual - SREcon17 Europe
7 likes2,582 views
As a sole system engineer at a startup, you will be responsible for much of the work alone initially. However, there are benefits such as gaining experience and taking on significant responsibilities. It is important to plan the infrastructure by understanding dependencies, documenting processes, and prioritizing basic tools. Automate as much as possible, be consistent, and create reusable solutions. Over time, share your knowledge to help the team grow and distribute responsibilities more broadly.
Startup Systems Engineer's Instruction manual - SREcon17 Europe
- 2. 1 The Normality of being a member Share the workload Share ideas Share the responsibility Guidance and mentoring Specialisation Polyphony Bus Factor > 1 You are not alone
- 3. 2 Joining a startup as a sole system engineer You will do most of the work You will talk to yourself You are responsible You are both a generalist and a specialist Bus Factor ≃ 1 2.1 You are alone
- 4. 2 Joining a startup as a sole system engineer Time to grow up Take decisions and responsibility Resourcefulness and creativity Fewer blind spots 2.2 The bright side
- 5. 2 Joining a startup as a sole system engineer Know why you are here Set the standards Bring your experience Stand between devs and infrastructure Find answers and solutions Design and advise for the future 2.3 Significance of your role
- 6. 3 Planning and building your infrastructure Learn about the product and its future Understand components dependencies Learn the current deployment process Gather information Document the current state Document manual steps Detect bad habits 3.1 Observe and listen
- 7. 3 Planning and Building your infrastructure Configuration Management*/Orchestration Provisioning environments Monitoring & Alerting Logging & Metrics Update Management Viable backups Security 3.2 The big picture * where available
- 8. 3 Planning and building your infrastructure Prioritise Start with familiar and basic tools Avoid complex solutions (save it for later) Develop tools Iterate 3.3 Take small steps
- 9. 3 Planning and building your infrastructure Decide how things should be done Automate as much as possible Find reusable and clear solutions Be consistent (e.g. in naming) Be consistent (i.e. be consistent) Hide complex procedures Revise when you should 3.4 Creating processes and rules
- 10. 3 Planning and building your infrastructure Choose wisely what to research, and when Understand the company’s size and needs Learn to let go Be patient Failures will happen, and you’ll fix them Think ahead and design for tomorrow 3.5 Research, deploy, break
- 11. 3 Planning and building your infrastructure Use a work tracking tool Try to plan your week Have a shared wiki (FAQs, Runbooks etc) Write readable code 3.6 Document
- 12. 4 Common pitfalls Always saying yes Always saying no Over-engineering Failure to decode information Underestimation Assumptions Mind the gap
- 13. 5 Working in a development team Manage interruptions Be approachable and compromise Share your knowledge and train Communicate processes and rules Don’t be arrogant You are in the same team
- 14. 6 Becoming a team Recount your experiences Delegate Train Let the team evolve Don’t micromanage Share your Legos
- 15. 7 Summing up Make short iterations Let your solutions mature Revise Your role is to save time Always say “we” Embrace failures and mistakes Don’t be arrogant You can do it!
- 17. Links Ansible: https://scotch.io/tutorials/getting-started-with-ansible Jenkins: https://www.sumologic.com/blog/devops/jenkins-developing-ci-pipeline/ Jenkins: https://www.novoda.com/blog/new-jenkins-pipelines/ Fabric: https://tinyurl.com/q86mer3 Munin: http://blog.bbv.ch/2013/02/16/linux-server-monitoring-with-munin/ Monit: https://www.linode.com/docs/uptime/monitoring/monitoring-servers-with-monit Tools