Infrastructure on your own servers
We build and run infrastructure — from the cloud to the application.
Private clouds on OpenNebula and Ceph, S3 storage, an edge with WAF and TLS 1.3, PHP made faster and hardened. At Hetzner, OVH, Selectel or on your own hardware — on accounts that belong to you.
strict-transport-security: max-age=31536000
content-security-policy: default-src 'self'; base-uri 'none'; object-src 'none'; frame-ancestors 'none'; form-action 'self'; img-src 'self' data:; font-src 'self'; style-src 'self' 'unsafe-inline'; script-src 'self' 'unsafe-inline'; connect-src 'self'; manifest-src 'self'; upgrade-insecure-requests
x-frame-options: DENY
x-content-type-options: nosniff
referrer-policy: strict-origin-when-cross-origin
permissions-policy: accelerometer=(), ambient-light-sensor=(), autoplay=(), battery=(), camera=(), display-capture=(), document-domain=(), encrypted-media=(), execution-while-not-rendered=(), execution-while-out-of-viewport=(), fullscreen=(), gamepad=(), geolocation=(), gyroscope=(), hid=(), idle-detection=(), magnetometer=(), microphone=(), midi=(), payment=(), picture-in-picture=(), publickey-credentials-get=(), screen-wake-lock=(), serial=(), sync-xhr=(), usb=(), web-share=(), xr-spatial-tracking=()
alt-svc: h3=":443"; ma=2592000This is for teams that have outgrown a couple of VPSes or pay AWS more than their load is worth; whose Bitrix or WordPress site falls over during a sale; who publish something worth attacking — and have nobody whose job is the infrastructure. If you run one server and it works, we have nothing to sell you. We will say so.
Services
What we take on
Private Cloud
We build a cloud on OpenNebula and Ceph over dedicated servers: VMs, volumes and S3 at hardware cost rather than at a hyperscaler's list price.
- OpenNebula on KVM: virtual machines, networks, quotas and self-service for your team through the web UI and the API
- Ceph under VM disks (RBD) and object storage (RGW, S3): three nodes or more, with no single point of failure
- Servers at Hetzner, OVH, Selectel or on hardware you already own — ordered on your account, not ours
from $4000 to build, then from $600/mo
Read the detail →PHP: Speed and Hardening
We find why a Bitrix, WordPress or Laravel site is slow and what it leaves open, and fix it at the server, the database and the configuration.
- Measured before and after: TTFB and p95 on key pages, under load rather than on a single request
- PHP-FPM, OPcache and the PHP version sized to your code and the server's memory, not left on defaults
- Slow MySQL, MariaDB or PostgreSQL queries found in the log and fixed with indexes and tuning
audit from $400, fix quoted from it
Read the detail →Managed Edge
We run the public entry point to your service so it stays up under load and never leaks its origin.
- HAProxy and OpenResty in front of your backend, with the origin unreachable from the internet
- TLS 1.3 and HTTP/3, certificates issued and rotated without your involvement
- WAF in blocking mode, with rules tuned to your traffic rather than left on a vendor default
from $400/mo
Read the detail →S3 Storage
Your own S3-compatible storage on Ceph or MinIO — or your buckets moved out of AWS to somewhere egress is not half the bill.
from $1200 to build, then from $250/mo05Mail Deliverability
We make your mail arrive — SPF, DKIM, DMARC, reputation, and the work of getting unblocked when it stops.
$500 setup, then from $200/mo06Mesh & Access
We take your servers off the public internet and give your team a way in that does not depend on a shared password.
$800 setup, then from $300/moProviders · account in your name
Hetzner · OVHcloud · Selectel · AWS · your own hardware →02Engagement
How the work runs
Three steps, and stopping after the first is a normal outcome — the report is yours either way.
- 01
Read what is there
Every engagement starts with the audit, because we do not quote a rebuild for something we have not read. About a week, and it ends in a written list of findings, ordered by what they would cost you, with numbers wherever speed or money is involved.
- 02
Build it
On your infrastructure, with our automation, at the fixed price the audit named. Nothing moves until it is built and you have seen it working.
- 03
Cut over, then operate
Traffic and data move in a window you pick, and the old path stays in place until you decide to drop it. After that we run it: upgrades, disks, certificates, WAF rules, and the response at three in the morning.
03Proof
What we run ourselves
The same automation we sell is the one we run: 63 Ansible roles covering edge, mail, monitoring, hardening and backups; 46 runbooks; 32 playbooks. The tooling is ours too — the heartbeat monitor, the support desk and the API gateway we would deploy for you have been running on our own fleet for months. Servers are ordered in your name and the configuration lives in your repository: if we stop working together, you keep everything needed to carry on without us.
Who does the work
The work is done by the RX Team — the same engineers who write our automation and run our fleet. The audit, the build and the operation stay with one team, with no hand-offs between departments: whoever read your system is the one who fixes it.
- Automation roles
- 63
- Runbooks
- 46
- Playbooks
- 32
- Own MTA
- DKIM
The audit is the door and it is priced to be one: $500–1500 depending on size — how many servers, domains and applications. It takes about a week. Everything after it is optional and quoted from what the audit found.