Skip to main content

Platform

Platform metrics for CanadaLogin, in the sre schema. It holds two groups of tables: incident tables, which record outages and how long they took to resolve, and uptime tables, which record whether each monitored site was up and how fast it responded.

The incident tables come from the CanadaLogin Platform team’s CanadaLogin_MTTR_Dashboard Google Sheet, maintained by hand in the CDS Google Drive. The sre-mttr-sheets-fetcher Lambda reads it through the Google Sheets API every Monday at 06:00 ET. Both tables have one row per incident and differ only in history: mttr_current holds the latest state of each incident, while mttr_history adds a dated snapshot of every incident on each upload.

Incident Tables

Table Description
sre.mttr_current Latest state of every incident, one row per incident.
sre.mttr_history Dated snapshots of every incident, one row per incident per snapshot.
No matching items

Data quality

Because the source is a hand-maintained spreadsheet, a few caveats apply to both tables:

  • Timestamps are strings, not native timestamps. detected_at, acknowledged_at, mitigated_at, and resolved_at are text in America/Toronto local time at minute precision (YYYY-MM-DD HH:MM). Parse them before doing date math. (fetched_at is a real timestamp, in UTC.)
  • URLs may carry trailing whitespace or embedded newlines. Trim slack_channel and postmortem_link before joining on or rendering them.
  • Categorical columns are unvalidated free text. severity, service, component, and root_cause_category are typed by hand, so expect occasional typos, casing, and spacing drift. The values listed in each field table are those observed so far, not a closed set.
  • The duration columns are rounded to the nearest minute from second-precision source times, so recomputing them from the displayed (minute-precision) timestamp strings can differ by up to a minute.

Uptime Tables

The uptime tables are sourced from the cds-snc/status-statut public GitHub repository and updated automatically each day. No manual steps required.

Table Description
sre.uptime_summary Daily rolling-window uptime and response-time figures for each monitored site.
sre.uptime_timeseries Status snapshots for each monitored site, one row per upptime bot commit.
No matching items