Part 3 of 6 · 1 min read
Roles
Common practice, not from the books.
UDS tells the engineer to post actions to a shared channel, and BM says each developer is responsible for running their own services. Beyond that the books do not set out roles. In larger incidents, teams commonly name a few jobs (common practice, not from the books):
| Role | Job |
|---|---|
| Incident lead | Decides what to try next and keeps the group focused |
| Responders | Investigate and carry out the fixes |
| Communicator | Updates other teams, support staff and sometimes customers at fixed intervals |
| Scribe | Records the timeline: what was seen, tried and decided, and when |
One person can hold several roles in a small incident. The point is that someone decides, someone talks to the outside world, and someone writes things down, so that the people fixing the fault are not also answering questions.
Good updates (common practice) state what users see, since when, what is being done, and when the next update will come, even if only to say "no change".
Incident lead
The incident lead is the person who coordinates the response and makes the calls, so that responders can work without arguing about the next step.
Example: two engineers disagree about rolling back or scaling out. The lead picks rollback because it is faster to reverse.
Check yourself: 1. What does the scribe record? 2. Why separate communication from investigation? 3. What should an update say?