Join Slack using this link: https://join.slack.com/t/app-perf/shared_invite/zt-3opre7mb-OUBLj1iElF1EBuH86qSG5Q
More images are at the bottom.
NOTE: This application is in extremely beginning stages and I am still working out flows and learning the data model. I will be cleaning code up as I go.
This is a application monitoring app. I am trying to build an open source, easy to setup, performance monitoring tool.
Before you set up AppPerf, you need to make sure the following is installed and functional.
- Yarn
- Node
- PostgreSQL
- Mailcatcher, or another smtp mail server (In Development, if you want to send emails)
If you are on Mac, you should be able to install them with the following commands:
brew install yarn
brew install node
brew install postgresql
Setup should be straightforward and easy. We use foreman to manage processes. Begin by running the setup script:
./bin/setup
This should create and setup the database and any other needed services/apps. Then to start everything up, run foreman:
bundle exec foreman start
AppPerf supports using Docker and Docker Compose. Just run the following and you should be good to go:
docker-compose build
docker-compose up
Navigate to http://localhost:5000. This application is setup to report to itself so you can begin seeing information immediately!
AppPerf can be easily deployed to any Kubernetes cluster using Helm (both v3 and v2 supported) with the chart provided. There are a few simple steps for this:
- With Helm v3 you need to create a namespace first, i.e.:
kubectl create ns app-perf
The above is not required with Helm v2 since it creates the namespace automatically.
-
Edit ./chart/values.yaml as needed with the required information, including connection strings for the Postgres and Redis databases and an optional ingress resource with TLS certificate to expose the app to the web.
-
Finally you can install AppPerf with the following command:
helm upgrade --install \
--namespace app-perf \
-f ./chart/values.yaml \
app-perf \
./chart
AppPerf has a sample account allowing you to check out the application
username: [email protected]
password: password
SQLite has some limitation with concurrency in this app, as well as various date functions used for reporting. If you would rather test against postgresql (Mysql to come), you can run the following command to create a local development database running on port 5443 (Must have postgresql installed):
./bin/setup_psql
Note: Don't forget to update your config/database.yml file.
App Perf will automatically detect new applications that are posting data and display them in the Applications page. From there you can go to each individual application to view the performance metrics.
In order to monitor an application, you have to add the Ruby Agent gem to the Gemfile:
gem "app_perf_rpm"
Once you have add the gem, Add the following lines to your project (or in an initializer):
require 'app_perf_rpm'
AppPerfRpm.configure do |rpm|
rpm.license_key = "License Key"
rpm.application_name = "Application Name"
end
You can get your license key by visiting the Applications tab and clicking the "New Application" button.
Install the App Perf Agent gem:
gem install app_perf_agent
Then run the following command on your server:
app_perf_agent --license LICENSE_KEY --host HOST
More information is on the Edit Organization page.
Adding metrics to App Perf is as simple as posting data to the following endpoint:
POST http://domain/api/listener/:protocol_version/:license_key
Currently the only protocol version supported is 2. License key is generated when you create a new user account. There is a default one that is used for testing in the .env.development file.
Clone the github project at https://github.com/randy-girard/app_perf_rpm or https://github.com/randy-girard/app_perf_agent somewhere locally, then run the following command for the RPM to force bundler to look at that specific path:
bundle config local.app_perf_rpm /path/to/local/app_perf_rpm
To remove this configuration, run the following command:
bundle config --delete local.app_perf_rpm
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent