24/7 NOC Monitoring and Incident Response in Rwanda
The EGreed Network Operations Centre watches your servers, networks, applications and backups around the clock from Kigali — with defined alert thresholds, escalation trees, one-hour incident response and monthly uptime reporting against SLA.
On this page
What a NOC actually does
A network operations centre is the difference between finding out your system is down from a customer and finding out from an engineer who is already fixing it. The EGreed NOC runs 24/7/365 from the Kigali headquarters, with a fully staffed secondary NOC at RWA-DC7 that assumes control if Kigali is unreachable.
Monitoring without response is just a dashboard. Every EGreed alert has an owner, a runbook and an escalation clock.
What we monitor and what triggers an alert
| Signal | Alert condition | Action |
|---|---|---|
| Host availability | Two consecutive failed checks (60s apart) | P1 — immediate engineer engagement |
| CPU | Sustained above 85% for 10 minutes | P3 investigation, capacity review |
| Memory | Above 90% or swap thrashing | P2 investigation |
| Disk capacity | Above 80% used, or 5 days to full by trend | P3 with growth forecast |
| Disk health | SMART pre-failure attribute or degraded RAID | Pre-emptive replacement scheduled |
| Service and port checks | Process down or port not answering | P1 restart per runbook |
| Application response time | p95 above agreed threshold for 5 minutes | P2 with trace capture |
| HTTP status | Non-2xx error rate above 2% | P2 with log correlation |
| TLS certificates | Expiry within 21 days | P4 renewal task |
| Backup jobs | Failed, skipped or shrinking backup size | P1 within the hour |
| Database replication | Lag above agreed seconds | P2 investigation |
| Network links | Packet loss above 1% or latency spike | P2 with carrier ticket |
| Security events | Brute force, new admin, privilege change | Security escalation |
| Environment (on-prem) | UPS on battery, room temp above 27 °C | On-call dispatch |
The escalation ladder
- Automated check fails and the alert is enriched with recent metrics and logs
- Tier 1 NOC engineer acknowledges within the SLA and applies the documented runbook
- If unresolved in 15 minutes for P1, Tier 2 systems or network engineering joins
- At 30 minutes, the customer's nominated technical contact is called, not emailed
- At 60 minutes, the service delivery manager opens a formal incident bridge
- Resolution is confirmed with the customer before the incident is closed
- A root cause analysis is issued within 3 business days for every P1
Onboarding — from contract to live monitoring
- Discovery workshop: systems, business hours, what 'down' means to you
- Inventory and dependency mapping of hosts, services and network paths
- Agent and probe deployment, including read-only access where needed
- Threshold tuning against two weeks of baseline data to kill false positives
- Runbook authoring per service, reviewed and signed off by your team
- Escalation tree and contact rota loaded, with a live test call
- Go-live, followed by a 30-day tuning review
Reporting you can take to a board
- Monthly uptime achieved per service, measured against the contractual SLA
- Incident log by severity with response and resolution times
- Mean time to detect and mean time to restore trends
- Capacity forecast: which resource runs out first, and when
- Top recurring alerts and the permanent fixes proposed for each
- Patch and vulnerability posture across monitored estate
Service tiers
| Tier | Coverage | P1 response | Best for |
|---|---|---|---|
| Watch | 8×5 monitoring, email alerts | 4 hours | Small offices, non-critical systems |
| Respond | 24/7 monitoring with runbook remediation | 1 hour | Growing businesses with production systems |
| Assure | 24/7 monitoring, remediation, capacity and security correlation | 15 minutes | Banks, telcos, government, healthcare |
Request this service
Use the service request form at the bottom of this page. It routes your request to the owning EGreed team, confirms a first-response time and lists exactly what happens next. You can also email support@egreedtech.org or call +250 795 822 290.
Delivery is backed by 1 Kigali headquarters, 8 Rwandan data centres and 3 branch service centres, with servers in 10 African countries and presence in 70 countries worldwide.
Official EGreed links
- EGreed Technology — egreedtech.org
- All EGreed services
- Request a quote or site survey
- EGreed Cloud
- EGREED ID
See also the services hub, the FAQ hub and the locations and infrastructure map.
Frequently asked questions
Is the EGreed NOC staffed at night and on weekends?
Yes. The Kigali NOC is staffed 24/7/365 with an on-call rota, and a secondary NOC at RWA-DC7 takes over if Kigali is unreachable.
How quickly do you respond to a critical alert?
Fifteen minutes on the Assure tier and one hour on Respond, measured from alert generation to engineer acknowledgement.
Can you monitor systems hosted outside EGreed?
Yes. The NOC monitors on-premise servers, other clouds and SaaS endpoints alongside EGreed Cloud workloads.
How do you avoid alert fatigue?
Thresholds are tuned against two weeks of baseline data before go-live, alerts are deduplicated and correlated, and a monthly review permanently fixes the top recurring alerts.
Do you fix problems or only report them?
The Respond and Assure tiers include remediation under agreed runbooks — restarts, failovers, capacity extensions and vendor escalation — not just notification.
What reporting do we receive?
A monthly report with uptime against SLA, incident log, MTTD and MTTR trends, capacity forecast and the permanent fixes proposed for recurring alerts.