Probot Metrics

[lswith]

Feature Request

Is your feature request related to a problem? Please describe.
There is a hard limit on how many github api calls can be made from a specific user. The current api limit is 5000 per hour. Being able to track how many github api calls have been made is needed.

Describe the solution you'd like
I'd like to be able to collect metrics from probot. The metrics I'd like to collect are:

• how many github api calls have been made
• how many webhook requests have been received
• how long it took for a specific event to be handled

Describe alternatives you've considered
Adding it into the bot itself through editing the server directly.

Teachability, Documentation, Adoption, Migration Strategy
If you can, explain how users will be able to use this and possibly write out a version the docs.
Maybe a screenshot or design?

[issue-label-bot]

Issue-Label Bot is automatically applying the label enhancement ➕ to this issue, with a confidence of 0.91. Please mark this comment with 👍 or 👎 to give our bot feedback!

Links: app homepage, dashboard and code for this bot.

[gr2m]

We handle the rate limit in latest probot by throttling requests as needed. It’s more complex that counting requests. The implementation is here: https://github.com/octokit/plugin-throttling.js

GitHub

octokit/plugin-throttling.js

Octokit plugin for GitHub’s recommended request throttling - octokit/plugin-throttling.js

[lswith]

That's perfect! I agree that it is more complex than counting requests but it's nice to have metrics from an operations point of view. Are metrics still in the works?

[tcbyrd]

We typically track metrics in external solutions like New Relic/Datadog internally. The challenge with this ask is tracking API requests per installation. I think it's definitely useful, but I'm not sure how it could be done in Probot itself. For example in a Lambda function, we simply couldn't track calls inside the framework. What you're suggesting would require an external service to monitor what's happening across the entire app. For Stale I'm experimenting with Logz.io, which is basically an ELK/Grafana solution as a service. This has given me some decent observability and is free for up to 3GB of data/day.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.