PROTOTYPE Clickable mockup · simulates the finished app ◃ Product overview
Petanque Life Operator Console
Production OP

Operations

Single pane of glass for active and historical incidents across every federation, club and player — plus SLOs, dependency health, cost and continuity. Times in UTC.

1
active incident
● SEV1 · API latency · 22 min
3
open incidents
● 1×SEV1 · 1×SEV2 · 1×SEV4
99.94%
uptime 30 d
● target 99.9% · error budget 41% left
1
on-call page right now
● Patrik W · ack 2 min
Dependency health (probe-driven)
● API · up ● Worker (job queue) · up ● Postgres · up ● Stripe (PAY) · degraded ● OIDC (AuthN) · up ● Object storage (backup) · up
99.94%
API uptime 30 d
target 99.9%
41%
Error budget left (API)
resets Aug 1
0.42%
Error rate (5xx) 24 h
threshold 0.5%
12,480 SEK
Infra cost this month
64% of budget
SLO & error budget per service · rolling 30 d
ServiceSLITargetActualError budget leftStatus
API availability 99.94%
41%
● OK ✎ edit target
API p95 latency < 800 ms 98.4%
12%
● Near budget ✎ edit target
Worker (job queue) job done < SLA 99.7%
62%
● OK ✎ edit target
Postgres availability 100.0%
100%
● OK ✎ edit target
Public status page availability 99.99%
88%
● OK ✎ edit target
Cost & budget · platform this month
12,480 / 19,500 SEK · budget alert job every 15 min
Container Apps + Postgres + object storage + third parties. Above 90% → budget alert (SEV4).
Public status page curation
Publish status update
What the public sees · petanque.life/status
Partial disruption
Jul 2, 11:52 · Investigating
Elevated response times
We are currently seeing slower response times in the app and are investigating the cause. Next update within 30 minutes.
History · subscribe to updates
Incidents · publication status
#SEVTitlePublic status
INC-451 SEV1 Elevated API latency ● published Update
INC-450 SEV2 Backup verify failed ○ not public (internal ops) Publish →
INC-448 SEV4 Stripe webhook retry queue ○ not public Publish →
States
⏳ Loading operations metrics…
✓ No open incidents — all quiet.
⚠ Could not load SLO data.

SLO targets are set by an operator (audited). The error-budget bar shows colour + number (never colour alone). A burned budget can auto-open an incident — the operator remediates; the system never auto-remediates. Publishing always requires an audit-logged reason.