Blog / Why We Replaced 4 Commercial IT Tools W…

Why We Replaced 4 Commercial IT Tools With One Self-Hosted Platform

Mt
Antony Njagi KigunduCore Mtaalam Technologies

Every IT department eventually hits the same wall. The organisation grows, the servers multiply, and suddenly you are paying for four overlapping tools — an RMM agent here, a monitoring suite there, a patching product from a third vendor, a SIEM nobody configured properly — and you still cannot answer the simplest question:

“Is everything actually okay right now?”

We were managing 32+ servers across a mixed Windows/Linux environment with Active Directory, SAP Business One, MSSQL, web servers and UPS power management. The commercial stack was costing a fortune, and it still had holes. So we made a decision: build the platform ourselves.

The cost of the old way

Let us be fair to the tools — each one did one thing well. The problem was the sum:

  • No single pane of glass. Server health in one dashboard, patch compliance in another, security events in a third. Comparing them meant opening three tabs and doing the mental math yourself.
  • Vendor blame games. When a patch window failed, the RMM vendor blamed the patching tool, which blamed the monitoring suite. Nobody owned the outcome.
  • Per-seat pricing that grew with every server. The bill kept climbing while the coverage stayed patchy.
  • UPS events were a coin flip. Commercial tools would notify you when the power died — but nobody was orchestrating an ordered server shutdown that could keep the fleet safe.

    That last one was the tipping point. Power failures in an agribusiness are not a question of if but when, and a messy shutdown corrupts SAP databases, MSSQL logs and file servers. We needed software that understood the whole power-down sequence, not just a dashboard with pretty graphs.

    What we built instead: KIFAA

    KIFAA is a full-stack, self-hosted platform we architected from scratch:

  • A lightweight Go agent deployed to every endpoint — one codebase cross-compiled for Windows (amd64/arm64) and Linux, polling every 5 seconds.
  • A Python/FastAPI backend with 43 API routers, backed by PostgreSQL with TimescaleDB for time-series metrics.
  • A React/Tailwind frontend with 45 pages — no training manual needed.
  • Celery + Redis running every background job: metric collection, alert evaluation, AD sync, UPS polling, report delivery.

    One platform now covers what four products used to:

  • Monitoring: CPU, RAM, disk, network, services, software inventory, hardware specs — 137M+ metric rows and counting.
  • Patching: Windows Update integration, compliance dashboards, scheduled patch windows with group targeting.
  • Active Directory: read/write control of users, groups and OUs, bulk operations, stale account detection.
  • Security: SIEM syslog listener, IOC scanning, phishing simulation, Sophos integration, CIS-aligned compliance checks.
  • UPS orchestration: custom raw SNMPv2c polling of APC devices, with ordered shutdown policies, cancellation windows and Telegram/WhatsApp notifications.
  • Nine integrations: VMware, Proxmox, Nutanix, Sophos, O365, SAP B1, Unitrends, APC UPS and more.

    Why self-hosted was the right call

    People assume self-hosted means “more work for us”. It is the opposite. When you own the platform:

  • Your data stays yours. No telemetry, no cloud dependency, no “we are migrating you to our new platform” emails.
  • You can fix anything. A bug in a commercial tool means a support ticket and a two-week wait. A bug in ours means a fix by the end of the day.
  • It adapts to your business. Want a shutdown policy that starts with the SAP server and ends with the UPS itself? That is a configuration, not a feature request.

    The honest trade-offs

    We should be straight about what self-hosting costs:

  • You carry the maintenance. The platform is only as current as your update discipline — which, ironically, KIFAA itself helps enforce.
  • You need one competent engineer. Not a team — one person who knows the stack. That is the difference between owning your platform and being owned by it.
  • First build is slower than first purchase. Buying a tool takes a day. Building one takes months. But the payoff compounds from day one, and you stop paying forever.

    What we learned

    Three years in, the numbers speak for themselves: 32+ servers, 137M+ metrics, 99.9% uptime, and a security posture that went from “we think we are fine” to “we know we are fine, and here is the quarterly report proving it”.

    The biggest lesson was simpler than any architecture decision: tools you understand beat tools you merely subscribe to. That is the philosophy behind everything Core Mtaalam builds — and the service we offer to businesses that want the same advantage without building it themselves.

    KIFAA is available for deployment in your environment. Request a demo, or read the deployment guide to see what running it involves.

    Keep reading

  • 137 million metrics and counting: taming time-series data
  • Tutorial: monitor your SMB network in under an hour
  • Keep building

    Want this kind of capability in your business?

    Tell us what you are trying to run — we will tell you the honest way to run it. Free consultation, no obligation.

    Talk to the team
    Chat with us