recordatus listens with an endpoint /log that ingests logs/metrics from your web app and sends to a specified logstash instance.
The data sent to recordatus can be in any format you choose, depending on your application and configuration.
Send data to recordatus with tools like browser-bunyan, StrackTrace.js, or anything that can PUT/POST JSON.
"logstashUrl": "http://logstash:9080"
This is your logstash instance running something like the example pipeline configuration below.
"listenPort": 3050
This is the port recordatus listens on.
"index": "browser-%Y.%m.%d"
This is the index recordatus will write data to. strftime format is supported.
"type": "browser"
This is the type of elastalert document that will be created. The _type field on your record will be this value.
"routeName": "/log"
If you want to change the endpoint route name, you may do so here.
docker run -v ${PWD}/config.json:/home/node/app/config.json -it -p 3050:3050 servercentral/recordatus
Note the port in this invocation must match the listenPort in your config.
npm install
npm run start
recordatus should now be listening on the specified port, ready for browsers to send it data.
[PUT, POST] /log
Send a JSON object, or array of objects, to this endpoint. The objects can be in any structure you choose, depending on your needs. They will get sent to the the logstash index and type specified in the config.
This configuration on a fresh logstash install will send your data to the specified elasticsearch host.
input {
http {
port => 9080
type => "%{type}"
}
}
output {
stdout {
codec => rubydebug {
metadata => true
}
}
elasticsearch {
hosts => ["elasticsearch:9200"]
index => "%{index}"
}
}
If you already have a logstash instance you will have to integrate these rules into your existing config, noting the %{type} and %{index} variables. If you do not include these variables in your logstash config, and instead enter your own values, recordatus will use those values instead.