roukien / rundeck-activity-monitor

A small app to monitor rundeck executions across multiple rundeck instances

Rundeck Activity Monitor (RAM)

RAM is a tool to monitor rundeck jobs executions across multiple rundeck instances. It is composed of two main components: a scraper and a UI.

The scraper is in charge of requesting the different rundeck instances to get the execution history, and the UI is rendering the executions across all instances in a swimlane chart.

Documentation / Usage 📝

Documentation is available at https://roukien.github.io/rundeck-activity-monitor/

Contributing

Building RAM

1. build the frontend app (which is basically a react SPA)
2. build the backend app and embedd the frontend inside it

Running locally

1. Start the postgresql container (it is recommanded to run the one described in the docker-compose.yml file)
2. Ensure the schema is up to date by running cd backend && go run ram.go database update
3. Start developping
4. optionnally start the UI cd frontend && npm run dev (requires you to run the backend server: go run ram.go serve)

rundeck-activity-monitor's Issues

Wrong colors being displayed on the timeline

It seems that in some cases, the rendered color in the timeline is wrong.

I think that the colors computation is false.

Timeline blocks browser when it has to render more than 8K executions

Persist Search params in URL

Ensure the search params are kept in the query string so that any user may share a link that will perform the same search on the app.

Supported search params:

• instances
• projects
• statuses

Search params dropdown

Allow the search params to be hidden when there's too many params.

Scraping improvement

Some refactoring to have a better scraping process:

• Better usage of SQL prepared statements when inserting
• Rework of rundeck clients to receive a chanel and directly push retrieved executions in that chanel

New config option to scrape more than 20 items per page when fetching rundeck executions

We use the rundeck default value for scraping x elements max per API call, but it can be improved by sending a max parameter. It would reduce the number of API calls to do if you're OK to have a higher resource consumption on the rundeck side (retrieving more executions at a time increase the API response time)

Have some docs

Build a documentation that allows newcomers to perform the following actions:

Administration guide

• Getting started
• Configuring a Rundeck instance for scraping
• Configuring multiple instances
• Configuring the scraper
• Starting the server

Contributing guide

• starting the app

Manage Rundeck timezones

When scraping the executions, we retrieves the executions history by considering the rundeck Timestamps. Those may be not the same than the one of the user.

scrape daemon

Allow the scrape command to run as a long term process.

Basically by adding a --scrape-interval option that would trigger a scrape every 1h

New option to scrape execution from the past X hours

Right now it is not possible as there is a mandatory begin and end option to scrape jobs.

The idea would be to scrape jobs from the last X hours so that this command could be used when we want to automate scraping.