Thundra .NET agent now supports intelligent sampling so you can reduce your data usage. The agent provides you built-in samplers for Metrics, Traces and, Logs;
You need to upgrade the agent version to 1.4.0 to use sampling. Documentations are here and the sample code is here. Come to our community slack and get instant help!
With this update, we are starting to send you weekly and daily performance reports of your functions! Reports include general metrics of your functions during the corresponding period, the problematic functions in terms of health and the most costly functions. You can choose the type of reports you want to receive under the profile settings page. Both weekly and daily reports are enabled by default.
With this update, you will be able to deep dive into your traces with Thundra's advanced query capabilities. For example filtering transactions that has interaction with DynamoDB and its interaction duration is greater than 1000 ms may help to find out bottlenecks related to DynamoDB in your application. After finding out the related trace, you can easily see invocation details, metrics and logs related to that trace which will help you fix the problem.
You can search following aspects of your transactions with Thundra:
There are more of them that you can utilize while searching your traces, please refer to our documentation for more details and you can read related blog post.
Please provide your feedbacks and feature requests to support@thundra.io or on our Slack channel.
With this new release you can resolve your policies that you no longer want to see in alert state. When you resolve a policy Status goes back to OK state. Resolved policies go back to alert state if an alert event related to policy occurs again.
With this update to Thundra console, you'll be able to filter and sort your AWS Lambda functions according to their health which unlocks you the ability to keep control your serverless architecture better. You'll also be able to set up alerts using health queries.
Ability to filter according to error count is a good way of understanding what may go wrong in your system. However, this might be misleading sometimes. For example; you can filter the functions which has 10 errors in last 1 hour and it can return some results. This might not be a big problem because function has over a million invocations and having 10 erroneous invocation can be acceptable. For this reason, there should be a way to query the functions according to health. You will be able to filter functions whose health diminishes. For example; you can query the functions whose health is less than 80% and sort them ascending with respect to health. Now, you have a way of understanding the functions with poorest health.
The biggest advantage of this update arises as the ability to save health queries as alerts. Now, you can set up an alert saying that "Notify me when the number of functions in the user service whose health is less than 95% becomes more than 5.". In this way, you don't need to keep track of the count of errors. You can focus on what really matters: Health!
Please provide your feedbacks and feature requests to support@thundra.io or on our Slack channel.
.NET agent v1.3.0 has been released, and with this release, Thundra .NET agent supports HTTP instrumentation by default, no code change required. You can see the URL, path, query parameters, HTTP method type, HTTP response code, and the HTTP request in the span. What you can see is configurable as stated here.
With a new addition to our query console, you can now share the queries with your colleagues to report an anomaly in your system or to share your insights.
In order to achieve this, you need to click on the `Share` button in the query console. This will copy a special URL to clipboard. When you open this URL in a separate browser window, you'll see that Thundra arranged the global time range at top-right to that specific interval and run the query.
Alternatively, you can share the queries just copying the URL from browser address bar. In this case, you can only share the query but not the time interval.
We're glad to answer one of the most wanted items in our backlog with this improvement. This update will foster the collaboration between the serverless teams with nice monitoring. Happy serverless-ing!
Thundra .NET agent v1.2.4 is released. Let's have a quick look at what features are in version 1.2.x:
Check our documentation and come to our Slack and get instant support there.
Today, we are very glad to announce that we are introducing Thundra alerts. This feature is taking advantage of our flexible queries that you are already familiar of. Using alerts, you'll be able to stay on top of your serverless issues and will be able to maintain a peaceful serverless stack.
We defined default alert policies into your accounts checking the erroneous invocations. You may be receiving emails coming from this alert policy. You can use this as an opportunity to improve your system or creating more queries which fits to your use case more. All you need to do is saving a query as an alert policy and stay on top of your problems.
You can catch more details from our documentation: https://docs.thundra.io/docs/alerts-page. Please come to our Slack and get instant support there.
