Honeybadger Insights
You can use Honeybadger Insights to dive into the data collected by Honeybadger and the logs and other events that you send to our Events API. We provide a query language (that we lovingly call BadgerQL) that enables quick discovery of what's happening inside your applications. The Insights UI also lets you chart the results of those queries and add those charts to dashboards that you can share with your team.
Querying and Visualization
Our query language strives to be minimalist, yet powerful. With it you can specify which fields you want to see, filter the kinds of events that should be returned, perform aggregations and calculations, and more. When you first load the Insights UI, you will see a query box that has a default query to help you get started:
fields @ts, @preview
| sort @ts
This query selects a couple of special fields — the timestamp and a preview of the fields that are available in the event — and sorts the results by time, with the most recent results first. Each row of the query is piped through the following row, which allows you to apply filters, formatting functions, and so on. Let's do a quick walk-through to see how it works, and to see how it can be used to create visualizations of your data.
Walk-through
Here's an example of working with some Honeybadger data. First, filter the data to see only the results of uptime checks:
fields @ts, @preview
| filter event_type::str == "uptime_check"
| sort @ts
You can see that we've piped the initial results through filter, which accepts a variety of conditions, such as the string comparison shown here. You'll also notice that we specified the data type of the event_type field (str) so the query parser can validate the functions and comparisons that you use on the field data.
Clicking on the disclosure arrow will show the all the fields that were stored for an event:
Additional disclosure controls appear inside the event detail view when the event has nested objects.
Let's filter on some additional data that is present in these events. We can limit the results to show only the uptime checks that originated from our Virginia location, and we can change the fields that we display so we can see some info about the results of each check:
fields @ts, location::str, response.status_code::int, duration::int
| filter event_type::str == "uptime_check"
| filter location::str == "Virginia"
| sort @ts
Now let's summarize the data to find the average response duration for all successful checks:
fields duration::int
| filter event_type::str == "uptime_check"
| filter location::str == "Virginia"
| filter response.status_code::int == 200
| stats avg(duration) by bin(1h) as time
| sort time
We use stats to perform all kinds of calculations, such as averages, and by allows us to specify the grouping for those calculations. Grouping by bin gives us time-series data, which makes it easy to create a chart by clicking the Line button.
From there you can experiment with different visualizations, update the query to change the chart (try changing 1h to 15m), and add the chart to a custom dashboard.
Of course, this functionality isn't limited to only the data that is generated by Honeybadger. Your error data is also available for querying (event_type::str == "notice"), and you can send logs and events to our API to be able to query and chart your own data.
Working with Dashboards
Dashboards allow you to collect different types of charts and query results on a single page. Any query or chart that you generate can be added to a dashboard, which will then be shared with the rest of your team. Each widget on a dashboard will include a link to view the query and raw results behind the widget:
If you change the query or the visualization, you can then save those changes back to a custom dashboard, or add them as a widget on another dashboard.
We provide some automatic dashboards to get you started:
- Honeybadger Dashboard: Includes widgets that show Honeybadger data, such as the current status of your uptime checks, volume of error notifications, etc.
- Heroku Dashboard: Once you add a Heroku drain to your app, this dashboard will show data like the number of requests grouped by response code that can be automatically gathered from Logplex.
Adding data from other sources
Insights includes all the events that Honeybadger collects, such as error notifications, uptime checks, and check-in reports, but you can send your own event data as well.
Our API accepts newline-delimited JSON, where each line is a JSON object that describes an event that you care about. You can send user audit trail events, metrics, or any other data you'd like to query and analyze. The type of data most frequently sent to Insights is application log data.
Sending structured logs in a JSON format (like lograge produces) allows you to correlate what's happening in your app with the error data that Honeybadger is already recording for you. See our integration guides to learn how you can easily send log events from sources such as Heroku apps and CloudWatch Logs.