The new, more powerful Dynatrace problem feed allows you to quickly identify trends in your environment, automate problem resolution, or to see a high-level management summary of what's going on in your environment.
In order to resolve problems before your customers are affected by them, it’s critical that you instantly see the reasons underlying performance issues. Dynatrace Davis®, our radically different AI causation engine, automatically processes billions of dependencies to identify the root causes with unmatched precision.
Davis-detected anomalies are reported as problems within your Dynatrace monitoring environment. You can review the list of currently active and resolved problems for any given timeframe by navigating to the Problems page (just select Problems from the navigation menu).
Particularly for large cloud environments, it’s crucial that you get a fast, clear, and flexible overview of all your Davis-detected problems, which you can then slice and dice according to your needs. Dynatrace version 1.204 addresses this need with a completely new and much more powerful problem feed, which now allows you to:
- Get a better overview by easily sorting the new, table-based view by columns
- Identify trends in your environment with the new, adaptive aggregation chart
- Quickly compare large timeframes by loading millions of problems in just one second
- Easily identify which problems affect the most entities by sorting them based on blast radius
- Immediately see if a detected problem was delivered to a first-responder team or went unnoticed
- Automate problem resolution by seamlessly exporting any query result via our public REST API
Get a better overview by easily sorting the new, table-based view by columns
One of the most important improvements in the new problem feed is the switch from a tile-based design to a table-based view. With the table-based view, you now have a much clearer and more compact representation of each problem with an organized format for viewing and comparing problem details.
Catch the most recent problem by sorting the list by start time
In a table-based feed, the ability to sort all problems by their respective start times is much easier and more natural than in a tile-based view. Sorting by start time allows you to easily identify the most recent problem in your environment.
Identify the longest-lasting problems by sorting the list by duration
Another valuable sort option is the newly introduced Duration column. All loaded problems can be sorted by duration, which provides a quick overview of how long each problem has been active. This allows you to quickly focus on the problems that have negatively influenced your infrastructure for the longest time.
Identify trends in your environment with the new, adaptive aggregation chart
The new and more powerful problem feed is fully capable of showing aggregated query results in the chart, no matter how complex your query is.
The aggregate view gives you an immediate, high level management summary of what’s going on in your whole environment. By glancing at the chart you can easily see when critical outages occurred and when they were resolved. This is an especially useful tool for quickly seeing the difference before and after an outage on the environment level.
See the resulting chart and problem list below of a multifaceted query that includes several tags, a severity level, and an impact level:
Quickly compare large timeframes by loading millions of problems in just one second
The new problem feed is no longer limited in terms of the query timeframe. We’ve dramatically improved the performance of long timeframe queries. In addition to aggregated results and totals, the response loads the first 50 results for each query. This results in load times of under a second, even for queries that look back over the last 365 days.
This allows your Operations teams to quickly compare the number of problems this month versus last month.
In the screenshot below, a query into all problems over the last 365 days results in a dedicated chart and a total of 1 million problems—the results are displayed immediately.
You can expect the same performance when you run any filtered query, such as a query into all problems that contain one or more entity tags.
Easily identify which problems affect the most entities by sorting them based on blast radius
The new problem list view now shows the number of Entities affected. This number is known as the “blast radius,” a term that’s widely used within the DevOps continuous-delivery community that measures the overall negative impact that a deployment can have. The count of affected entities is a good indicator of the overall impact of a problem.
By sorting according to blast radius, you can now quickly pinpoint the problems that affect the most entities in your environment. Also, this feature clearly shows how hundreds of individual alerts are reduced into one—a level of noise reduction that only the Dynatrace Davis AI is capable of!
Immediately see if a detected problem was delivered to a first-responder team or went unnoticed
Alerting profiles are used to send out alerts for detected problems over your various configured alerting channels, such as email or Slack.
Based on our discussions with customers, we’ve found that information about who’s been notified is crucial because it tells you if a detected problem was delivered to a first-responder team or if it went unnoticed. Now Dynatrace displays the notified alerting profiles for each problem directly in the problem feed, so you can see at a glance which problems were delivered to which responders.
See the notified alerting profiles example below for each listed problem. Of course, you can also filter and query the problem list by Alerting profiles.
Automate problem resolution by seamlessly exporting any query result via our public REST API
The new problem feed is built on top of and fully integrated with the newly introduced user-facing Problems REST API v2. As a result, any query result you see in the Dynatrace web UI can also be generated and exported by directly addressing the public REST API. This API-first strategy also allows us to offer a seamless experience for third-party integrations.
For example, if you deploy a new service version, you might want to automatically check your environment and regularly query Dynatrace on the API to give you the amount of problems for that certain entity. In case there is a problem, you can trigger the automatic rollback of that service version.
How do I get the new problem feed?
You can enable the new problem feed via the teaser banner at the top of the Problems page (see below, select Try it out). In the coming months, the new view will become the default, replacing the current tile-based view.
This side-by-side offering of both the new and the existing problem feeds will help us collect your feedback before we switch all Dynatrace environments over to the new problem feed, sometime around the end of this year.
Your feedback matters! We hope you like the new problem feed and we look forward to your comments and questions. Please post your feedback in this Dynatrace Product Ideas forum thread.