System Requirements

Telebugs can run on almost any hardware. This includes a VPS, cloud server, home server, or even a Raspberry Pi. It supports both AMD64 (also known as x86-64, x64, x86_64, and Intel 64) and ARM64 (also known as AArch64) architectures. In short, if Docker runs on it, Telebugs will too.

Supported Platforms

Telebugs works with over 100 programming languages and frameworks through Sentry SDKs. Popular options include JavaScript (Node.js, React, Angular), Python (Django, Flask), Ruby (Rails), Java (Spring), PHP (Laravel), .NET (ASP.NET Core), and Go. For the full list, see Supported Platforms in the appendix.

Error Throughput

Use this table to estimate sustained, fully processed error reports. These numbers are about reports that have made it through the ingest queue and are visible in Telebugs, not just HTTP requests accepted by the ingest endpoint.

The 2 vCPU / 4 GB RAM row is based on 10-minute benchmarks on a Hetzner CX23 VPS with local disk. In those tests, Telebugs processed about 50 reports per second end-to-end under ingest-only and mixed ingest/UI workloads. The ingest endpoint also accepted short bursts above 160 errors per second, with queued reports draining afterward.

Larger rows are planning estimates. Actual throughput depends on CPU, RAM, disk speed, application payload size, retention settings, notifications, and how much UI traffic the same server is handling. Throughput is shared across all projects in one Telebugs installation. With two busy projects, each gets about half. With four busy projects, each gets about a quarter, and so on.

Benchmark Methodology

Telebugs includes a source-level benchmark harness at bin/load. This is not part of the customer-facing telebugs CLI. It is meant for repeatable release and sizing benchmarks.

The benchmark starts an isolated Dockerized Telebugs instance in the performance environment, creates a throwaway project token, runs k6 against the Sentry envelope endpoint, waits for the ingest queue to drain, and prints both intake and processing rates.

For the 2 vCPU / 4 GB RAM baseline above, we ran these 10-minute tests on a Hetzner CX23 VPS:

bin/load --users 100 --duration 10m --read-ratio 0 --groups 200 --drain-timeout 3600
bin/load --users 100 --duration 10m --read-ratio 10 --groups 200 --drain-timeout 3600
bin/load --users 100 --duration 10m --read-ratio 20 --groups 200 --drain-timeout 3600

The important result fields are:

• Envelope accept rate: how quickly the ingest endpoint accepted incoming reports during the k6 run. This is the burst intake rate.
• Reports processed end-to-end: how many reports per second were fully processed after including any queue drain time. This is the sustained sizing number used in the table.
• HTTP p95 latency: the 95th percentile response time during the load test.
• HTTP failures and failed jobs: both should stay near zero for a healthy run.
• Peak pending ingest payloads: how much backlog accumulated while Telebugs absorbed the burst.

For capacity planning, prefer Reports processed end-to-end. A high Envelope accept rate with a large pending queue means Telebugs absorbed the burst successfully, but the server would need more CPU, faster disk, or a lower incoming error rate to sustain that load indefinitely.

Recommended Minimum Specs

• RAM: 1 GB
• Disk space: 40 GB
• CPU: 1 core

Recommended Specs for Small to Medium Projects

• RAM: 4 GB
• Disk space: 40-80 GB
• CPU: 2 cores

Operating System Compatibility

• Tested on Linux (Ubuntu, Debian, Alpine, Rocky) and any OS that supports Docker.
• Also works on macOS.

Rocky Linux 9 note: Rocky works well once Docker is installed, but the Telebugs installer may not be able to install Docker automatically there. If you see a message that Docker installation is not supported on this platform, install Docker first, then run your Telebugs installation command again:

sudo dnf config-manager --add-repo https://download.docker.com/linux/rhel/docker-ce.repo
sudo dnf -y install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin
sudo systemctl --now enable docker
sudo usermod -a -G docker YOUR_USER

Replace YOUR_USER with the Linux user that will run Telebugs. Log out and back in, or run newgrp docker, so the new group membership is active. If automatic updates complain that cron is not installed, install or enable Rocky’s cronie package; if needed, add a daily telebugs update job manually with crontab -e.

See the Rocky Linux Docker guide for the latest Docker installation details.