Site status & network outages

Real-time site availability derived from core signalling: which sites are down, since when, how many customers are affected — with hierarchy, history and a resolution workflow.

Live

site availability computed from S1AP signalling — independent of vendor EMS

Impact-first

every outage carries an affected-customer count, so priority is obvious

Hierarchy

parent/child site relations show what a single failure really takes down

At a glance

Status: Live

Deployed at: CETIN SK — production

Domain: Network

Built with: S1AP / core alarms · Kafka · BitSwan pipelines · TESU

What it does

The application watches core signalling (S1AP and related alarm sources) and continuously derives which sites are actually up — vendor-independently, without waiting for an EMS to decide something is worth an alarm. Each detected outage is enriched with the site’s customer base, its parent/child relations and its baseband inventory, then tracked until resolution.

Why it matters

An alarm list answers “what is complaining?”. This board answers the question the NOC actually triages by: “who is offline, and how many customers does each of these represent?” The affected-customer count turns prioritisation from tribal knowledge into a sort order, and the site hierarchy makes cascade failures (one transmission site taking children with it) readable at a glance.

Because the same site-status feed powers TESU, a support agent looking up one complaining customer sees instantly that their whole site is down — and the NOC sees how many more calls are coming.

How it works

1. BitSwan ingestors stream S1AP and core alarm sources into Kafka.

2. Pipelines correlate the signalling into per-site availability states, debounce flapping, and attach customer counts and topology from inventory connectors (CROSS, SBR).

3. The status board and the per-site detail (location, basebands, 30-day history) are served as a BitSwan application, with the same data exposed over APIs for downstream tools.

Why it replicates

S1AP is S1AP. The derivation logic, the impact enrichment and the UI are operator-agnostic BitSwan modules; replication is a matter of connecting your signalling feed and your inventory.

From the live system

Screenshots

Site status board. Sites currently down with start time, duration and affected customers — in this view, 6 sites and 3,360 customers. Sorted by impact, not by alarm count.

Site detail. One site with its location, basebands, affected customers and 30-day availability history — outage context without leaving the page.

Screenshots come from the production system and will be anonymised before public launch.