Slaying the SLAs: Mastering Effective Communication
summary
Built an internal app called Roadrunner using Python FastAPI, Celery, Redis, MongoDB and Jinja2 to manage ~3 billion daily messages and meet a 90‑day archival SLA (90% in 1–7 days, 99.99% in 7–30). Automation targets the <1% of IDs that fail: a daily t‑3 verifier moves stuck IDs to a Recon DB, a 4‑hour pandas refresh updates states, an auto-remediation script retries team APIs, and a daily Jira creation script + synccat notifies teams. Three months of work removes manual toil and prevents potential SLA breaches.
AI-written summary · about
chapters (10)
- 0:00Introduction and Smash overview
- 4:11Understanding SLAs and challenges
- 7:24Problems and manual toil in SLA management
- 8:27Introducing Roadrunner application
- 10:54Daily state verification script
- 12:19Periodic dynamic update script
- 13:02Auto remediation script
- 14:05Daily Jira creation script
- 15:29Results and impact of Roadrunner
- 16:12Q&A on tool impact and data tracking
related talks
24mSLO Creation And You: Or, How We Learned To Stop...
18mSLICK: Driving SLO Culture
28mHow to get away with refactoring
35mAvoiding Goodhart's Law - Use SLO's as Tools Not Cudgels
01h 06mPepperoni, Puppets, and Priorities: Demystifying SLAs, SLOs, and SLIs
39mSRE or Devops - I am an Architect - Why Do I Care?
21mSRE, what is it all about?!
20mUsing Monitoring to Speed Up Your DevOps Journey