Datadog CI Visibility provides a unified platform for monitoring your CI/CD pipelines. Now, we are partnering with CircleCI to extend that same critical visibility to your CircleCI environment. Datadog’s integration uses CircleCI webhooks to capture information about the status and performance of your workflows and associated jobs, such as a job’s duration and whether or not it failed or was canceled. Once you set up the integration, you will be able to:
Having this visibility into your CircleCI environment enables you to track performance trends in CI pipelines and surface problems that could affect a release, such as a job within a key workflow that is suddenly failing more often than others.
Your CI/CD pipelines are key to releasing new features and bug fixes to customers, and a slow pipeline can prevent you from releasing on time. Datadog CI Visibility helps you track the status of your CircleCI workflows (referred to as pipelines in Datadog) and their constituent jobs via a built-in pipelines dashboard. This gives you valuable information about their overall performance, enables you to quickly spot build errors, and find areas that need optimization. For example, you can use the pipelines dashboard to easily compare performance across all running workflows and jobs in order to see which ones took the longest amount of time to execute.
From there, you can decide the best course of action for improving pipeline efficiency, such as implementing a caching strategy for dependencies or configuring your tests (e.g., unit tests) to run in parallel.
Pipeline failures are commonplace in CI environments, so it can be difficult to know when a failure is legitimate or not. Datadog CI Visibility makes it easy to see which jobs in your CircleCI environment are failing the most, so you can prioritize improving the reliability of critical jobs in order to avoid the delay of future releases.
You can drill down to a specific pipeline execution to get more information, including a flame graph breakdown of every job and test that ran as part of that execution and their duration. Failed jobs are automatically highlighted in red, so you can quickly identify which ones you need to review first.
This information gives you more context for why an issue occurred, so you can quickly distinguish between code and pipeline issues and inform the appropriate team of the problem. This not only helps in reducing the number of interruptions to your teams but also decreases the amount of time it takes to resolve the issue.
CI environments generate a large volume of notifications, which can create alert fatigue and cause your team to miss legitimate issues. You can already create alerts based on CircleCI metrics, but you will soon be able to leverage CircleCI events to build more powerful, pipeline-specific alerts that fit your teams, the jobs they maintain, and the activity they care about. For example, instead of triggering a general alert anytime a pipeline fails, you can alert a specific team when a job within a workflow they manage takes longer than expected to execute.
You can also use Datadog’s turn-key integrations for popular communication platforms like Slack to ping a specific channel when an issue occurs. This helps cut down on the number of notifications that are triggered in your CI environment and ensures that performance issues or anomalies are routed to the appropriate teams for troubleshooting.
With our partnership with CircleCI, you have better visibility into the status and performance of workflows and jobs. Datadog not only makes it easy to compare workflow or job performance across your entire CircleCI environment but also gives you deeper insight into the causes of a failure or performance degradation. This allows you to improve your CI pipelines’ day-to-day performance and reduce the number of environment issues that would otherwise cost your team more time and effort to address. Check out our documentation to learn more about our CircleCI integration, or sign up for a 14-day free trial today.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。