The microservice contains an API that accepts any JSON object from the browser and outputs it to the container's stdout. The container configuration should have a Logging Driver setting that will read the stdout and pass it to your log aggregation system.
Features:
- Allows you to collect logs from the browser using server-side logging tools
- Does not limit you in choosing a server logging tool, choose any
- Includes exporter of metrics for Prometheus,
To try the microservice features, run the container with the command:
docker run -it --rm -p 3000:3000 mtsrus/logsNow you can check the work with the command:
curl -d '{"message": "Sample log"}' -H 'Content-Type: application/json' -X POST http://localhost:3000/logs/logJSON should be displayed in the container console. This JSON must be read by the Logging Driver and sent to the logging system.
To start the microservice in production, use the command:
docker run -d --restart always -p 3000:3000 mtsrus/logsAll endpoints use the POST request method and accept a JSON as the request body
/logs/log- information level log/logs/warning- warning level log/logs/error- error level log
The microservice has built-in Prometheus monitoring and is located on the endpoint /metrics.
Block this endpoint on the proxy if you do not need to provide access to metrics from outside your network.
You can use any log collector on the browser side. To send to a microservice, you need to write a send function to your logging tool.
Example:
const log = (message, level, error) => {
fetch(
`http://localhost:3000/logs/${level}`,
{
method: "POST",
body: {
message: message,
userAgent: navigator.userAgent,
location: location.href,
stack: error.stack
}
}
);
}