Starting today, we also extract error incidents from Heroku Logplex data and provide you with a magic dashboard for Heroku status codes.
With the latest update to our Logplex parser, we can now extract Heroku errors from the data stream. We add a "heroku" namespace in AppSignal, where we capture errors such as "H12: Request timeout", "H14: No web dynos running," and many other error codes, complete with samples like any other error.
These errors can help you debug issues that might otherwise not be captured, such as the timeout error and no dyno's running error. The AppSignal agent requires a running (Ruby/Elixir/Node.js) process to catch errors, and with these Logplex errors, we can capture the last missing errors in your Heroku stack.
A logdrain is easily added by using the Heroku CLI and running the following command in your application directory:
heroku drains:add "https://appsignal-endpoint.net/logplex?api_key=<push_api_key>&name=<app_name>&environment=<app_environment>"
Replace the placeholders with your AppSignal Push API key, app name, and environment. You can find these values on the "Push & Deploy" page on AppSignal.
Errors are captured automatically after adding the logdrain.
A Magic Dashboard with Heroku Status Codes
Together with these errors, we'll also add a new Magic Dashboard to your app that tracks the overall response counts for each HTTP status code group (2xx, 3xx, 4xx, 5xx). This gives an overview of the overall response rates of your Heroku application. These metrics also allow you to set up Anomaly Detection triggers on the response codes to track any anomaly on Heroku.
Thanking Our Customers
We want to thank our customers who have contacted the support team for suggesting these features! If you would like to see something not already available on AppSignal, don't hesitate to reach out, we'd love to hear from you!
If you are not using AppSignal yet, maybe this encourages you to try us out and experience our developer-to-developer support.