ReferenceGlossary

MTTD and MTTA

Mean time to detect is the average time from the start of a failure until a signal shows it. Mean time to acknowledge is the average time from that signal until someone takes the incident. Google's SRE book counts both beside MTTR, because a team that fixes fast but notices late is still slow.

Reliability

In Quorum#

Every Incident records the sim second its first signal lands and the second a player first acts on it, and its Postmortem shows MTTD and MTTA beside MTTR. With Alerting the signal is the pager, which fires the Step a rule's threshold is crossed. Without Alerting the signal is the board, at the first second more than 1% of requests fail. On-call answers the page five calendar minutes after it lands, which is the MTTA a rotation holds. Without On-call, the player's first action is the acknowledgement.

Design#

The incident loop

MTTR, SLO, SLA, SLI, Four golden signals

Learn more#