Thundra

Thundra: Serverless Observability for AWS Lambda

The black box nature of AWS Lambda and other serverless environments means that identifying and fixing performance issues is difficult and time-consuming. Built for straightforward debugging, monitoring, and observability, Thundra provides deep insight into your entire serverless environment. Thundra collects and correlates all your metrics, logs, and traces, allowing you to quickly identify problematic invocations and also analyzes external services associated with that function. With Thundra’s zero overhead and automated instrumentation capabilities, your developers are free to write code without worrying about bulking up their Lambdas or wasting time on chasing black box problems.

Get Started    Discussions

Graphics Page

This page gives an overall picture of all the invoked functions that fall in the specified filter. You can filter out displayed functions based on profiles, regions, function types, interval, date, and specific function.

Graphs

Invocation counts: Shows total invocation counts per function basis for each time interval.

Average duration: Shows average invocation durations in milliseconds per function basis for each time interval.

Cold start invocation count: Shows total "cold start" invocation counts per function basis for each time interval.

Cold start invocation durations: Shows average "cold start" invocation durations in milliseconds per function basis for each time interval.

Error counts by function names: Shows total occurred error counts per function basis for each time interval. Here, “ occurred error” means that error is thrown from the invocation as an exception.

Error counts by error types: Show total occurred error counts per error type basis for each time interval. Here, “occurred error” means that error is thrown from the invocation as an exception.

Cold Start

"Cold start" invocation means that the container and the application are created and initialized during the invocation so it took more time than the regular ones which are responded by already running containers.”

How to read graphs

You can get exact information for each time interval when you hover over the graphs. Information at a point is the sum of occurrences between the selected point and the next consecutive point on the time axis. For instance in Invocation Graph, invocation count at each point is the sum of invocations between that point and next consecutive point on the time interval axis.

Filter Bar

Profiles: Specifies profiles where the audit data was collected from. Profiles are used for representing different environments such as "development", "staging", "production", etc ... One or more profiles can be specified. "All Profiles" means skipping profile based filtering by accepting all available profiles. The default value is "All Profiles".

Regions: Specifies AWS region(s) where the audit data was collected from. One or more AWS regions can be specified. "All Regions" means skipping region based filtering by accepting all available AWS regions. The default value is "All Regions".

Function types: Filters functions according to language type. Default value is "All Types".

Time interval: Specifies the time interval before the specified "END DATE" when the audit data was collected. Time intervals determine the granularity of time periods which are used for aggregating audit data by time in date histogram. The default value is "DAY". There are 3 types of time intervals:

  • Hour: Specifies the last hour before the specified "END DATE". In this time interval, the time period is "1 minute" for aggregating metrics.

  • Day: Specifies the last day before the specified "END DATE". In this time interval, the time period is "30 minutes" for aggregating metrics.

  • Week: Specifies the last week before the specified "END DATE". In this time interval, the time period is "3 hours" for aggregating metrics.

End Date: Specifies the end date when the audit data was collected before. The default value is "Current Time".

Functions: One or more functions can be specified to be monitored by Thundra. "All Functions" means skipping function based filtering by showing metrics for all functions. The default value is "All Functions".

The Refresh button resets all the filters and updates the current page to the default filters.