Skip to main content

sre.mttr_current

Latest state of every incident, one row per incident.

The latest known state of every CanadaLogin incident, one row per incident.

Each row is a single incident. The table is overwritten on each upload with the most recent values, so it never grows beyond the current incident set. For point-in-time history, see mttr_history.

Keywords: MTTR, MTTA, MTTM, incident, incident response, postmortem, severity, reliability, SRE, mean time to resolve

Provenance

Fetched automatically every Monday at 06:00 ET from the CanadaLogin MTTR Dashboard Google Sheet using the sre-mttr-sheets-fetcher Lambda. Each upload overwrites this table wholesale with the current incident state and appends a dated snapshot to mttr_history.

  • Updated: Weekly, Every Monday 06:00 ET
  • Retention: 7 years
  • Steward: EDCP Data and Research Team
  • Contact: Yinlin Zhang (Data and Research team) or Obie Okongwu (CL Platform Team)

Fields

One row per incident, keyed by incident_id. See the schema-level Data quality note for caveats that apply to the string timestamps, URL columns, free-text categories, and rounded durations.

Field Name Type Description
incident_id string Unique identifier for the incident, format INC-NNN
slack_channel string URL of the incident’s dedicated Slack channel
severity string Incident severity. One of “Sev1”, “Sev2”, “Sev3”, or “Sev4” (Sev1 most severe, Sev4 least). Free text in the source, not strictly enforced
service string The CanadaLogin service affected
component string The specific subsystem affected, more granular than service. Free text
short_description string One-line free-text summary of the incident
detected_at string When the incident was detected, America/Toronto local time at minute precision (YYYY-MM-DD HH:MM)
acknowledged_at string When the incident was acknowledged. Same format and timezone as detected_at
mitigated_at string When the incident was mitigated. Same format and timezone as detected_at
resolved_at string When the incident was fully resolved. Same format and timezone as detected_at
root_cause_category string Free-text root-cause classification
mtta_alert_ack integer Minutes from detected_at to acknowledged_at, rounded to the nearest minute
mttm_detected_mitigated integer Minutes from detected_at to mitigated_at, rounded to the nearest minute
mttr_detected_resolved integer Minutes from detected_at to resolved_at, rounded to the nearest minute
postmortem_link string URL of the incident’s postmortem Google Doc. Optional
week integer ISO-8601 week number (1-53) of detected_at
month string Calendar month of detected_at, first-of-month formatted as mmm-yy text (for example “Feb-26”)
fetched_at timestamp When the pipeline ingested this snapshot, in UTC