The Curious Incident in your Software in the Day Time
summary
Advocates a concrete incident plan: centralised reporting in Slack workflows with dedicated incident channels, Slack calls for auditable timelines, email templates and a status page; fallback to Teams if Slack fails. Gives real failures — GitHub leaking repo URIs, Atlassian accidentally deleting ~400 sites, a Huddle identity bug that let six sessions share tokens out of ~5 million logins (fixed within hours, code changes and nonce added), and an Azure uk-south outage lasting eight and a half hours — and recommends rehearsals and biannual DR.
AI-written summary · about
chapters (13)
- 0:00Introduction to incidents
- 1:51Why incident plans matter
- 3:42Real-world incident examples
- 6:24Defining and reporting incidents
- 10:58Planning and rehearsal
- 13:18Reporting tools and processes
- 18:32Communication and roles
- 22:43Post-incident analysis
- 23:48Huddle case study
- 30:10Azure and cloud incidents
- 35:03Weird incidents and lessons
- 39:38Key takeaways and Q&A
- 41:40Incident response team structure
related talks
57mThe Curious Incident in your Software in the Day-Time
15mIncident Ready: Chaos Engineer Your Incident Response
20mHow to Have an Operational Incident
35mIncident Management: Lessons from Emergency Services
19mHow to handle incidents at scale
30mHow To Have An Operational Incident (a crash course)
22mSmarter Disasters: End-to-End Automation for Incidents
27mLessons from Incident Management and Postmortems at Atlassian