Loku System Status

This repository is the data source for the Loku Status Page and Loku Changelog.

All service incidents and product changelogs are published here as GitHub Issues and reflected on the public pages in real-time.

---

Status Page

The status page reads GitHub Issues from this repository to display incident history, per-service availability, and auto-calculated SLA.

Live: loku.co.id/status.html

Incident Labels

Label	Required	Description
incident	✅	Marks the issue as an incident. Required to appear on the status page.
database-gateway	optional	Affects Database Gateway status and SLA
rest-api-gateway	optional	Affects Rest API Gateway status and SLA
web-app	optional	Affects Web APP status and SLA
billing-system	optional	Affects Billing System status and SLA

An issue labeled only with incident (no service label) is treated as a system-wide incident — all services will be marked as disrupted and overall SLA will be impacted.

Per-Node Database Labels

The Database Gateway is backed by multiple database cluster nodes. On the status page the Database Gateway row expands to show a per-node breakdown. To scope an incident to a single node, add the node label in addition to incident:

Label	Affects
database-gateway	The whole gateway — all DB nodes marked disrupted
database-gateway-db1	Node db-1 only
database-gateway-db2	Node db-2 only
database-gateway-db3	Node db-3 only

Convention: the node label is the service slug with the node name as a suffix → database-gateway-<node>. To report an outage on one node, open a new issue with labels incident + database-gateway-db2 (for example). A node-specific label keeps the incident isolated: other DB nodes and other services stay green, and per-node SLA is calculated independently. A bare database-gateway label (no suffix) still disrupts every node. Adding a new node later only requires registering its slug in the status page config (DB_NODES).

SLA Calculation

SLA is calculated automatically from the duration of each incident within the last 30 days:

SLA = (total_minutes_30d - total_downtime_minutes) / total_minutes_30d × 100

• Downtime is measured from created_at to closed_at. Open incidents count until the current time.
• Overall SLA is derived from all incident issues.
• Per-service SLA is derived from incident issues that also carry the service label.

Uptime Bar

The 90-day uptime bar on the status page is color-coded per day:

Color	Meaning
Green	No incidents that day
Yellow	Incident occurred and has been resolved
Red	Active incident (issue still open)

---

Changelog

The changelog page reads GitHub Issues from this repository labeled changelog.

Live: loku.co.id/changelog.html

Changelog Labels

Label	Required	Description
changelog	✅	Marks the issue as a changelog entry. Required to appear on the changelog page.
feature	optional	New feature — shown as ✨ Fitur baru
fix	optional	Bug fix — shown as 🔧 Perbaikan
improvement	optional	Enhancement — shown as ⚡ Peningkatan
security	optional	Security update — shown as 🔒 Keamanan

---

Subscribe to Updates

Watch this repository on GitHub to receive email notifications whenever a new incident or changelog is published.

👉 Watch this repo

---

Contact

For service inquiries, please reach us at halo@loku.co.id or visit loku.co.id.