Server monitoring
Updated
Every Wolkra cloud server comes with CPU, disk and network graphs in the console, so you can see at a glance how busy a server is and how much traffic it handles. The graphs are included at no cost and need no agent inside the server.
What you can see
The server page in the console shows three graphs:
- CPU: how much of the server's vCPU capacity is in use
- Disk: read and write activity on the server's disk
- Network: inbound and outbound traffic on the server's interfaces
Nothing to install
The graphs are part of the console. You do not have to set up a monitoring stack before you can see whether a server is busy, and they are there for every server, whichever operating system it runs.
Reading the graphs
CPU
A server that sits close to full CPU for long stretches is short on compute. Options are a bigger plan in the same family or a move to the CPU-optimised c family. Short spikes during deploys or cron jobs are normal.
Disk
Constant heavy disk activity often points to a database that has outgrown its memory. More RAM lets it cache more and read less, so the memory-optimised m family is a good next step. A large block storage volume helps when you run out of space, not speed.
Network
Traffic on Wolkra is unmetered, so the network graph is for understanding your workload, not for counting euros. It shows release days, backups leaving the server, crawlers and unexpected spikes. Sudden outbound traffic you cannot explain can also be a sign that something on the server is misbehaving. See network for how unmetered bandwidth and fair use work.
From graphs to action
| What you see | What to try |
|---|---|
| CPU near full for hours | Resize, or move to CPU-optimised plans |
| Heavy disk activity and slow queries | More RAM with memory-optimised plans |
| Traffic peaks at release time | A load balancer and more servers |
| Unexplained outbound traffic | Check processes, tighten firewalls |
Cloud servers can be resized to a bigger plan from the console; the disk can grow but not shrink.
Combining with your own tools
The built-in graphs answer the first question: is the server busy? For application metrics, logs and alerts, run the stack you prefer. With root access you can install Prometheus exporters, Netdata, Uptime Kuma or anything else. A small monitoring server on a private network can watch the rest of your fleet without exposing dashboards to the internet. See self-hosting for ideas.
The API lets internal tools read server details with a read-only token.
A simple routine
A few minutes a week with the graphs pays off:
- Open each production server and look at the last few days.
- Note the normal level of CPU, disk and network for that server.
- Compare peaks with what happened: a release, a newsletter, a backup job.
- Act on trends early, before a busy day turns a slow server into a down one.
When you change something, such as a new cache, a bigger plan or an extra server behind a load balancer, check the graphs again to see whether it helped. It is the quickest way to learn what your workload really needs and to avoid paying for capacity you do not use.
Service status
Platform status is shown on the status page in the console. If something affects the Frankfurt location, you will find it there, and you can open a support ticket from the same place.
Typical uses
- Checking whether a WordPress site needs a bigger plan
- Sizing database servers
- Watching build load on CI runners
- Following traffic on game servers and VPN servers
Limits to know
- Graphs cover CPU, disk and network per server. Application metrics and alerts are up to your own tools.
- Graphs are shown per server in the console.
Related products
- Cloud servers
- Network
- Load balancers
- Teams and security for the activity log
See your servers clearly
Graphs are on from the moment a server starts. Sign up for Wolkra and launch a server in Frankfurt.
Frequently asked questions
What does Wolkra monitoring show?
Each server has CPU, disk and network graphs in the console. They show how busy the server is and how much data it reads, writes, sends and receives.
Do I need to set anything up?
No. The graphs appear on each server's page in the console once the server is running.
Does monitoring cost extra?
No. Graphs are included with every cloud server.
How do I know when to upgrade my plan?
If CPU stays close to full for long periods, or the server is slow while disk graphs show constant heavy activity, a bigger plan or a different family is likely to help. You can resize from the console.
Can I use my own monitoring tools as well?
Yes. You have full root access, so you can run any monitoring or logging stack you like next to the built-in graphs.