RTMP SERVER HOSTING
RTMP Server Hosting: Run Your Own Ingest and Restream From One VPS
An RTMP server is the ingest point your encoder streams to — instead of renting one from a streaming platform, you run it yourself on a Linux VPS with Nginx-RTMP, MediaMTX, or SRS. Your OBS or hardware encoder pushes a single stream to the server, which restreams it to YouTube, Twitch, and Kick simultaneously, or repackages it to HLS for playback on your own site. Deploy in about a minute across seven datacenters in the EU and US East.
X-ZoneServers Streaming VPS plans pair KVM virtualization with a 10 Gbps network port, unmetered traffic, and a published estimated minimum sustained throughput per plan, so you can size a 24/7 ingest against a known floor. Full root access on Ubuntu, Debian, AlmaLinux, or Rocky Linux means you install Nginx with the RTMP module, MediaMTX, or SRS exactly how you want — RTMP in, HLS out, SRT for contribution links, push directives for multistreaming. Relaying and repackaging are network-bound, so even the entry plan runs a continuous ingest with CPU to spare. 1 Tbps always-on DDoS protection (L3/L4) is included on every plan.
~60s
Deployment time
Unmetered
Network traffic
7
VPS locations
1 Tbps
Always-on DDoS (L3/L4)
01Why us
Why it works
Infrastructure matched to the workload — KVM isolation and full root, not a generic box.
One Ingest, Every Platform
Push a single stream from OBS or a hardware encoder and fan it out to YouTube, Twitch, and Kick with Nginx-RTMP push directives or FFmpeg — one upload, simultaneous everywhere.
10 Gbps Port, Unmetered
Streaming plans run on 10 Gbps ports with unmetered traffic and a published minimum sustained throughput, so 24/7 egress never triggers overage bills.
Your Stack, Full Root
Install Nginx-RTMP, MediaMTX, SRS, Owncast, or datarhei Restreamer on a clean Linux image. No managed-platform rules, watermarks, or feature gates.
24/7 Streams Allowed
Round-the-clock relays, looped broadcasts, and unattended channels are expected use on streaming plans — not a fair-use violation waiting to happen.
RTMP, SRT & HLS
Ingest over RTMP or SRT and deliver HLS to browsers — modern stacks like MediaMTX and SRS handle all three protocols on one server.
EU & US Ingest Locations
Seven datacenters across the EU and US East keep the first hop from your encoder short, which is what keeps a live stream stable.
02Use cases
Ideal for
Self-hosting your RTMP endpoint suits streamers who want to multistream without per-month restreaming-service fees, producers who need a stable ingest near their encoder, and teams embedding low-latency live video into their own site. Latency to the ingest matters more than raw bandwidth for stream stability, so pick the datacenter closest to your encoder — Frankfurt, Amsterdam, Vienna, Warsaw, Bucharest, Sofia or Ashburn. Live transcoding is the only CPU-heavy step: budget roughly 4–6 vCPU cores per 1080p30 x264 rung, or skip transcoding entirely and pass the encoder's bitrate through. You must hold the rights to all content you broadcast.
- Streamers multistreaming to YouTube, Twitch, and Kick without restreaming-service fees
- 24/7 radio, lofi, and looped channels run unattended with FFmpeg and systemd
- Event producers needing a private, stable RTMP ingest near the venue
- Embedding self-hosted live video (HLS) into your own website
- A contribution relay that turns one SRT uplink into multiple RTMP outputs
- Replacing a desktop PC that currently stays on just to keep a stream alive
03Questions
Frequently asked questions
How do I set up an RTMP server on a VPS?
Deploy a Linux VPS, install Nginx with the RTMP module (or MediaMTX/SRS as a single binary), open port 1935, and point your encoder at rtmp://your-server/live with a stream key you define. A basic ingest-plus-restream config is around 20 lines and is running in well under an hour.
How much bandwidth does an RTMP restream server need?
Upstream from your encoder costs one stream (say 6 Mbps for 1080p), and each restream destination costs one more copy — pushing to three platforms totals roughly 24 Mbps. Every X-ZoneServers streaming plan publishes an estimated minimum sustained throughput of at least 1 Gbps, so multistreaming uses under 3% of the smallest plan's floor.
Can my RTMP server convert the stream to HLS?
Yes. Nginx-RTMP, MediaMTX, and SRS all repackage RTMP input into HLS segments without re-encoding, which uses minimal CPU. Browsers then play the stream natively from your server or through a CDN in front of it.
Do I need a GPU to run an RTMP server?
No. Ingest, restreaming, and HLS repackaging are network-bound and run comfortably on CPU-only plans. Only live transcoding to multiple resolutions is CPU-heavy — budget about 4–6 vCPU cores per 1080p30 x264 rung, or pass the original bitrate through and skip transcoding.
Is a self-hosted RTMP server allowed to run 24/7?
Yes. Streaming VPS plans are built for sustained egress: unattended 24/7 relays and looped broadcasts are normal use, traffic is unmetered, and there is no overage billing. The only requirement is that you hold the rights to the content you broadcast.
04Related
Related products & use cases
Get started
Deploy your RTMP server
Spin up a streaming VPS with a 10 Gbps port, unmetered traffic, and full root in about a minute across seven locations in the EU and US East.