We’ve formalised a lot over the past few months, however Tech:Incidents on Meta is still guidelines and doesn’t cover clear approaches to what is an incident, what isn’t and what must be reported and what can be left to discretion.
|Open||None||T8793 Create a formal Incident Response/Management Process|
|Resolved||John||T8843 Implement documentation for all monitoring checks|
|Resolved||John||T8844 Allow defining an alert as critical|
|Open||None||T8845 Allow Icinga to generate Phabricator tasks for Critical alerts|
Considering this resolved. John has made edits to the original Tech:Incidents guideline page which addressed this task. No objections have been made clear in almost a year since those edits were made, and as mentioned above I also have none. Therefore, the page is now policy and should be followed. Any suggestions for changes can be made afterwards.