Engineering / ENG-1497
Status page and incident updates
Description
Shaped in Emargen: Status page and incident updates, started the day of OPS-33.
Why
On Wednesday 17 June phone codes to UK and Irish numbers arrived up to two hours late, and customers learnt it from their depot staff before they heard from us. Support told administrators one by one.
What we build
- status.acme-hq.example (set-up is OPS-34): sign-in, single sign-on, portal, email notices and SMS codes, per region. A static site outside our three regions' stacks, so it stays up when production doesn't.
- Fed by the outside checks we already run every minute: two failures in a row mark the component and page on-call. The incident lead writes the words; nothing is posted automatically.
- Who posts and when; how we write.
- Email subscriptions for customers' IT, by component.
Done when
Customers can subscribe and the on-call engineer can post. Target 1 July.
Child issues
- ENG-1507 Portal shows a banner while an incident is open on the status page Done
- ENG-1508 Status page: email subscriptions for customers' IT Done
Linked issues
- relates to OPS-33 Textbridge outage: SMS codes to UK and Irish numbers delayed for about two hours Done
- relates to SUP-965 Northgate: tell us when codes are down Done
- relates to OPS-34 Status page: set up status.acme-hq.example Done
- relates to ENG-1505 Phone-code screen says when codes are arriving late Done
Public since 10:00. Rules in Confluence OPS › Status page. Closing.