Initial configuration
A sudo user, SSH key login, direct root login disabled, timezone and NTP set, and a sensible partition layout.
Start here
Free Audits 5Urgent help
Emergency 4Ongoing support
Managed IT 5Projects
Servers & Hosting 6 Network & Virtualization 9 DevOps 5 Security & Recovery 5 Hardware & Licensing 2 IoT & smart automation 3Find out where your servers, security, backups and performance stand, at no cost.
Server down, network out, site hacked or data lost? Call us now.
Monthly network and server support with response times written into the contract.
Setup, configuration, management and migration of Linux and Windows servers, panels and mail.
Network design and cabling, MikroTik, VoIP, branch links, virtualization and private cloud.
Networking & communications
Virtualization & cloud
Containers, Kubernetes, automated delivery, infrastructure as code and observability.
Server hardening, firewalls, backup and DR, ransomware recovery and incident forensics.
Advice, supply and installation of servers and network gear, plus genuine enterprise licenses.
Connected homes, workplaces and devices, from scenario design to control and monitoring.
Start here
Urgent help
Ongoing support
Projects
You may have just received a fresh VPS, or you may run a server nobody has dared to update in years. Our Linux server management takes configuration, updates and troubleshooting off your hands, and we document every change. We work on the common distributions, such as Ubuntu, Debian and AlmaLinux.
Linux server management covers the configuration, updating, securing and troubleshooting of Linux servers, and every change gets documented. People usually come to us after receiving a fresh VPS with nobody to configure it, when a site throws occasional 502 errors for no clear reason, or when nobody can say when the server was last updated. On Ubuntu, Debian and AlmaLinux, PikoSystem creates a sudo user with SSH key login, sets up Nginx or Apache with PHP-FPM and auto-renewing Let's Encrypt certificates, and configures UFW or nftables with Fail2ban. Daily restic or Borg backups go to a destination outside the server, and security patches are installed in an agreed window. You can book it as a one-time setup or as monthly management; either way you end up with a configured server whose credentials are all handed over, configuration documentation, backup scripts with a test restore result, and a change log.
A sudo user, SSH key login, direct root login disabled, timezone and NTP set, and a sensible partition layout.
Nginx or Apache, PHP-FPM, Node.js or Python runtimes, and Let's Encrypt certificates with automatic renewal.
MySQL, MariaDB or PostgreSQL installed, memory settings sized to the server, and the slow query log enabled.
nftables or UFW with only the required ports open, and Fail2ban to block SSH brute-force attempts.
Daily file and database backups with restic or Borg to a destination outside the server, with test restores.
Security patches in an agreed window, journalctl and web server log review, and fixes for 502 errors, full disks or high memory use.
Choose a distribution by its security support, CPU architecture and application or control-panel compatibility. Ubuntu 24.04 LTS has standard security maintenance until May 2029; repository coverage and expanded support have separate terms. Debian 12 has been superseded by Debian 13 and is now in LTS until 30 June 2028, with limited architecture and package coverage. AlmaLinux 9 has active support until 31 May 2027 and security support until 31 May 2032; each minor release reaches end of life when its successor is released. Check the compatibility matrix for the exact application and control-panel version before choosing.
CentOS Linux 7 reached end of life on 30 June 2024. Ubuntu 18.04 left standard security maintenance in May 2023; Ubuntu Pro coverage has separate terms. Plan a migration around application compatibility, a recoverable backup, a transfer test and a rollback path. Whether to upgrade in place or move to a fresh server depends on the environment's constraints.
Exposing database port 3306 to the internet, setting site folders to 777 or running an application as root can increase the attack surface and the impact of a compromise. None alone proves that an attacker controls the whole server: the outcome depends on authentication, permissions, vulnerabilities and service isolation. Restrict database network access to required sources, set file permissions for the service's actual needs and run applications with the minimum permissions required.
When available memory is insufficient and the kernel cannot reclaim enough, the OOM killer may stop a process; MySQL is not a fixed victim. Selection depends on the scope of memory pressure, process usage, oom_score_adj and system policies. Swap does not guarantee prevention of OOM. Check kernel logs and service or cgroup limits to identify why a process stopped. Uncontrolled log growth can also fill the disk and interrupt database writes; configure logrotate and journald limits for the retention needs.
Make sure you can log in to your VPS provider's control panel. That is where snapshots are taken and where the VNC console lives, which is the only way back in if an SSH change goes wrong. Also gather the list of domains, sites and services on the server, plus access to your DNS management.
If you are not sure who has had access to the server over the years, tell us. At the start we go through the keys in authorized_keys and the system users, and rotate old passwords.
Server count is only part of it. A server running one WordPress site on Nginx needs far less attention than one running several applications, a message queue, PostgreSQL and Docker containers side by side. A control panel, the volume of data to back up and the need for out-of-hours changes also shape the estimate.
The starting condition matters too. A server that has gone years without updates or documentation needs a cleanup phase first, and monthly management becomes simpler after that.
We check the operating system, running services, open ports, disk usage and logs, and list the problems we find.
We describe the changes needed and their order, and agree a time window with you.
Before any major change we snapshot the server or its config files, and we apply changes during low-traffic hours.
Documentation and credentials are handed over. If you choose monthly management, updates and backup checks continue on schedule.
It is a server you rent from any hosting provider, with a technical team looking after updates, security, backups and fixes. We provide that management on the server you already have, so there is no need to move it.
We need administrative access to start. We recommend creating a separate sudo user with an SSH key for us, so our access is traceable and you can revoke it at any time.
We start with logs, CPU and memory usage and disk health to find the cause. If it is urgent, we bring the service back first and fix the underlying cause after. For outages, see our emergency services as well.
Both are available. A one-time setup ends with the documentation handover. Monthly management covers updates, backup checks and issue handling, under terms set in the contract.
Quote
A few lines on where things stand and what you want is enough. An engineer calls you back directly; requests don't go through sales.