Linux infrastructure that stays online.
When servers crash for no visible reason, clusters misbehave or an outage keeps coming back - we find the real cause and fix it for good.
ITdept is a Kraków-based engineering consultancy for teams that run their business on Linux. We step in where standard runbooks and vendor support come up empty.
When to call us
Clients usually come to us with one of these.
- Servers freeze, reboot or panic - and nobody knows why.
- The same outage keeps returning after every "fix".
- The vendor's support ticket has been open for weeks.
- Kubernetes or storage is unstable under real load.
- You need a migration or upgrade with zero downtime.
- Monitoring is noisy, and real incidents get missed.
- Your team is strong, but needs a senior Linux expert on call.
- You want an independent opinion on how reliable your setup really is.
Services
What you get - and the technology behind it.
Incident response & root-cause analysis
Your production is down or unstable. We find the actual cause - down to the kernel if needed - fix it, and make sure it doesn't happen again.
- Crash & freeze investigation - kernel panics, hangs, driver bugs and silent reboots explained from crash dumps and traces.
- Performance problems - memory, I/O and CPU bottlenecks found and removed.
- Postmortem you can share - a clear write-up of what happened and what was changed.
- Fixes that last - patches and backports instead of restarts and workarounds.
kdump · crash · GDB · eBPF · perf · ftrace · strace · ACPI/PCIe · kernel patching · DKMS
Reliability & high availability
Systems that keep working when a node, disk or data centre fails - and alerts that wake you only when it matters.
- High-availability setups - automatic failover and zero-downtime migrations.
- Useful monitoring - clear service levels and actionable alerts instead of noise.
- Storage & databases - distributed storage and high-load databases that stay consistent.
- Fewer repeat incidents - safeguards built after every serious outage.
Pacemaker · Keepalived · HAProxy · Envoy · BGP/FRRouting · WireGuard · ZFS · Ceph · NVMe-oF · PostgreSQL · SAP HANA · Prometheus · VictoriaMetrics · Grafana · OpenTelemetry
Kubernetes, cloud & automation
Platforms your team can operate confidently - reproducible, documented and secured.
- Production Kubernetes - design, build and operation of clusters with GitOps delivery.
- Infrastructure as code - cloud and bare metal described in code, not in someone's head.
- Security hardening - systems brought in line with CIS benchmarks.
- Clean packaging - your software delivered as proper DEB and RPM packages for Debian, Ubuntu, and RHEL.
Kubernetes · OpenShift · KubeVirt · KVM/Proxmox · Incus/LXD · AWS · GCP · OpenStack · ArgoCD · Flux · Terraform/OpenTofu · Ansible · SELinux · AppArmor · deb/rpm packaging
What you can expect
Three rules we don't bend.
Production stays safe
Hard problems are reproduced in isolated test environments first. Nothing is tried out on your live systems.
The cause, not the symptom
We fix the underlying problem instead of hiding it behind restarts and workarounds - so you don't pay for the same incident twice.
Your team can maintain it
Minimal, documented changes that follow upstream standards. When we leave, your team understands what was done and why.
ITdept is led by Viktor Pashaiev, a Linux systems and kernel engineer with over 20 years of hands-on infrastructure experience and an active upstream contributor to Ubuntu, Debian, and Linux.
Contact
Tell us briefly what's happening and what you've already tried. You'll get a reply within one business day.