Ben Johnson
Ben Johnson
This makes it obvious they need to commit and deploy. It also helps to show which files changed.
I want to support this new change to our schema: https://github.com/timberio/log-event-json-schema/commit/e238ae8932686c86da1f5b15f4d7cf43c0b2019a
Through Heroku addons they'll have the drain setup automatically, then they can optionally run the installer. The installer should be aware of this somehow and not prompt the user to...
https://github.com/timberio/timber-ruby/pull/102#issuecomment-307000092 These lines should be suppressed with the `logrageify!` option. And since Timber is now responsible for logging these, adding an option should be fine. The reason we keep them...
All of our docs specify an `Object` return type. We should be specifying the various input and output types.
Rack timeout logs `logfmt` style logs. We should capture these and classify them accordingly.
We need a way to track server restarts. Thin, mongrel, unicorn all log when they boot up.
As part of splitting out the libraries we need to support the `:header_keys_to_sanitize` option in the new `:timber_plug` library. Previously this was defined as a root level timber configuration, but...
Just like our agent, we should attempt to capture EC2 metadata if the Elixir application is on an EC2 server.
https://devcenter.heroku.com/articles/dyno-metadata