What we do · Operations
Network and power telemetry, logging, DDoS detection
We run three dedicated platforms underneath our alarm system: Observium for network and power graphs, Graylog for every device's logs, and Wanguard for denial-of-service detection at the border, and each of them has something independent watching it.
What it is
- Observium polls every switch, router, firewall, UPS and rack power strip over SNMP and keeps the history: traffic per port, errors, optics, temperatures, and the draw on each power feed.
- Graylog receives the system log of every device we run: Cisco routers and switches, firewalls, HPE 3PAR arrays, Brocade fibre channel switches, blade chassis controllers and Linux hosts. Logs are parsed, cleaned and sorted into streams per role, and the events that matter raise alarms.
- Wanguard reads mirrored traffic from the border routers and watches continuously for volumetric and denial-of-service patterns against any address we announce.
How we build it
Every source, normalised. Different vendors write logs differently, and some devices leave out their own hostname. Graylog rewrites every source to a clean, consistent device name on the way in, so a search for one router returns everything that router said, whatever format it used. Logs land in purpose-built streams (core routing, fabric, firewalls, storage, hosts) rather than one undifferentiated firehose.
Alarms for what precedes an outage. Event definitions raise an alarm in the operations chat for the things that come before trouble: a BGP session dropped, a link flapping, BFD reporting a dead path, encryption churn on the backbone, a hardware fault, a fibre channel error on the storage fabric, a NAT pool running out of addresses.
Every power strip, on both feeds. Every rack PDU, on the A feed and the B feed, is in Observium. That means per-rack draw and the headroom left if one feed is lost are measured numbers, not assumptions.
Border traffic, analysed. Port mirrors on the border routers feed the Wanguard sensors, so an attack is something we are already measuring when it starts.
Watching the watchers. A monitor that stops is worse than no monitor, because you trust the silence. The DDoS sensors have their own independent watchdog on separate infrastructure: a deliberately simple service that checks the sensors are collecting data and warns well ahead of any licence renewal.
Measured, not guessed. When a carrier tells us about planned maintenance, we measure what actually happened afterwards from two sides: from outside, using public probe data, and from inside, using link state, BGP session timelines and our customer URL checks. The answer to "did last night's maintenance affect us?" is a timeline with timestamps.
What it protects you from
- A detection system that has silently stopped, with nobody noticing the silence.
- A device that looks healthy on its own console but has stopped sending logs.
- An attack that is first reported by your customers instead of our sensors.
- A rack running closer to its power limit than anyone realised.
- An incident review that has to reconstruct events from memory.
- "We think it was the carrier" when you need a measured answer.
Why EuroVPS
- Every device's logs, from routers to storage arrays, searchable in one place.
- DDoS detection reads real traffic at our own border routers.
- Power telemetry on both feeds of every rack, not just a monthly bill.
- Our monitoring systems are themselves monitored by independent watchdogs.
- Questions about the past are answered with data, not recollection.
FAQ
Can you send my firewall or server logs to your Graylog? Yes, for services we manage. Logs are kept in streams with access limited to the engineers who operate your platform.
How long do you keep logs? It depends on the service and your requirements. Tell us what your auditors or policies need and we design retention to match.
Do you graph my own switch ports or links? For network services we provide, yes. You can ask for the graphs at any time.
If you want answers measured from the wire rather than guessed, Talk to an engineer.
Your next infrastructure partner
Let’s look at
your setup.
Tell us what you run today and what needs to improve. An engineer will review your brief and help you plan the next step.
+357 222 32280
who will look after your infrastructure.
Let’s start with your setup.
A little context helps us start the right conversation.