Amplify nginx Monitoring Tool: Setup & Use
- Requests per minute
- req/min
- Result
- Number of requests
- 1000
- You entered
- Time period (seconds)
- 60 s
- You entered
Amplify is a free, open-source monitoring tool built specifically for nginx and nginx Plus. You install a small agent on the same machine as nginx, and it collects metrics about your server’s performance, health, and configuration, then sends them to a web dashboard where you can view graphs, set alerts, and spot problems before they affect users.
Formula
The core metric Amplify calculates is requests per second (RPS), which is simply the number of requests divided by the time period in seconds:
RPS = N / T
- N = total number of requests processed (count)
- T = time period in seconds (s)
Amplify itself does not use a single formula; it aggregates dozens of metrics like active connections, request time, and upstream response times. But RPS is the foundational throughput metric, and understanding it helps you interpret the dashboard.
Worked Example
Using the default values from the calculator above: 1000 requests over 60 seconds.
- Divide requests by seconds: 1000 / 60 = 16.666…
- Round to two decimal places: 16.67 requests per second.
- For per-minute rate: 1000 / (60/60) = 1000 requests per minute.
The calculator shows 16.67 RPS and 1000 requests per minute, matching the manual calculation.
What Amplify Actually Monitors
Amplify collects data from three main sources: the nginx stub_status module (or the Plus API), system-level metrics like CPU and memory, and a metadata agent that reads your nginx configuration files. The dashboard then presents this in a timeline view, letting you correlate a spike in traffic with a rise in CPU usage, for example.
Key metrics you will see include:
- Active connections (waiting, reading, writing)
- Requests per second and per minute
- Response codes (2xx, 3xx, 4xx, 5xx)
- Upstream response times and error rates
- System CPU, memory, disk I/O, and network traffic
Unlike generic server monitors, Amplify understands nginx’s internal state, so it can tell you if your worker processes are too few, if your keepalive timeout is too short, or if your upstream servers are slow.
How to Install and Configure
Installation is straightforward, but the exact steps depend on your operating system. The general process is:
- Create a free account on the Amplify web dashboard (you will get an API key).
- Install the Amplify agent package using your system’s package manager (e.g.,
aptfor Ubuntu/Debian,yumfor RHEL/CentOS). - Edit the agent configuration file (
/etc/amplify-agent/agent.conf) and add your API key. - Enable the nginx stub_status module in your nginx config (usually at
/stub_status) and make sure the agent can access it. - Start the agent service and check the dashboard to see if your server appears.
If you are using nginx Plus, you do not need stub_status; Amplify uses the Plus API instead. The agent runs as a separate process and does not modify your nginx binary or configuration, so it is safe to install alongside production traffic.
Common Problems and Misconceptions
A frequent issue is that the agent cannot connect to the stub_status endpoint. This usually happens because the nginx configuration does not expose it on a port the agent can reach, or because a firewall blocks localhost access. Check that the URL http://127.0.0.1/nginx_status returns JSON or plain text when you curl it.
Another misconception is that Amplify is a load tester. It is not; it only observes. If you want to generate traffic to test your server, you need a separate tool like ab or wrk. Amplify will then show you the resulting metrics.
Some users expect Amplify to alert on every anomaly. By default, it only alerts on severe issues like the agent losing connection or nginx being down. You can configure custom alert rules in the dashboard, but it is not a full-featured alerting system like PagerDuty; it is primarily a visualization and analysis tool.
How to Choose Between Amplify and Alternatives
Amplify is not the only nginx monitoring option. Here is how it compares to common alternatives:
- NGINX Plus built-in dashboard: If you already pay for nginx Plus, its dashboard gives real-time metrics without extra setup, but it lacks historical graphs and configuration analysis.
- Prometheus + Grafana: More flexible and powerful for custom dashboards, but requires you to set up exporters, write queries, and manage storage. Amplify is easier for a quick start.
- Datadog or New Relic: Commercial SaaS tools that offer nginx integrations plus broader infrastructure monitoring, but they cost money and may collect more data than you need.
- Netdata: A lightweight real-time monitor that works with nginx, but it focuses on live metrics rather than long-term analysis and configuration checks.
Choose Amplify if you want a free, dedicated nginx tool with minimal setup and a clear web UI. Choose Prometheus if you already use it for other services and want full control. Choose a commercial tool if you need integrated alerting and support.
Costs, Safety, and Data Privacy
Amplify is free to use, including the agent and the cloud dashboard. There is no paid tier for the core functionality. However, the dashboard is hosted on nginx’s cloud, which means your server’s metrics and configuration metadata are sent to their servers. If your organization has strict data residency rules, this could be a concern.
The agent runs with minimal privileges and does not require root access, though it may need to read certain system files. It does not modify nginx configuration files; it only reads them. The main safety consideration is that the stub_status endpoint should be restricted to localhost or to the agent’s IP, not exposed publicly, to avoid leaking metrics to outsiders.
For long-term storage, Amplify keeps a limited history on the free tier, so if you need months of data for capacity planning, you may want to export metrics periodically or use a self-hosted solution.
Frequently Asked Questions
Does Amplify work with nginx open source?
Yes. Amplify works with both nginx open source and nginx Plus. For open source, you need to enable the stub_status module, which is included in most pre-built packages. For nginx Plus, the agent automatically uses the Plus API, which provides more detailed metrics like upstream health.
Is Amplify still free in 2025?
Yes, Amplify remains free to use. There is no charge for the agent, the dashboard, or the basic alerting features. nginx offers it as a way to encourage adoption of their ecosystem, similar to how other companies offer free tiers for their monitoring tools.
Can I monitor multiple nginx servers with one Amplify account?
Yes. You install the agent on each server and use the same API key. The dashboard will show all your servers in a list, and you can switch between them or view them side by side. There is no limit on the number of servers for the free tier.
What happens if the Amplify agent crashes?
If the agent stops running, nginx itself is unaffected; the agent is a separate process. You will stop seeing new data in the dashboard, and after a few minutes you will get an alert that the agent is offline. You can restart the agent service, and it will resume collecting data.
Does Amplify slow down my nginx server?
No. The agent uses the same stub_status endpoint that any monitoring tool would use, and it polls it every few seconds. The overhead is negligible compared to the traffic nginx handles. The agent also reads system metrics, which have minimal impact.
Can I run Amplify behind a firewall or in an offline environment?
Amplify requires outbound HTTPS access to the nginx cloud dashboard. If your servers cannot reach the internet, you cannot use the hosted version. In that case, you would need a self-hosted monitoring solution like Prometheus. There is no offline mode for Amplify.
Related
Formula used
in t | Number of requests | - | 1000 | 0 in d | Time period (seconds) | s | 60 | 1 out rps | Requests per second | req/s | t / d | 2 out rps_min | Requests per minute | req/min | t / (d / 60) | 2
Each line is one input (in, sel, list) or one result (out: label, unit, formula, decimals). How the calculators work
Method: the standard formula for this calculation, rounded as each result states. See how the calculators work or report an error.