eBPF Based Troubleshooting
Comprehensive root cause analysis and forensics for containers, Kubernetes metrics
The Five Most Difficult Problems to Troubleshoot
Troubleshooting Use Cases
Kubernetes Resource Utilization
Troubleshoot issues such as crashloop backoffs, pod evictions, and resource allocation and limits. See which containers are running out of CPU, memory, or file system resources leading to throttled or killed containers.
Event Correlation
Contextualize metrics and understand the "why" faster with a unified view of both metrics and events. You can configure Event Overlay to display events from Kubernetes such as deployments and node failures, as well as alert events, security violations, and custom events sent using Sysdig's open REST API.
Golden Signals
Application And Service Golden Signals. Identify issues in latency, errors, traffic, or saturation to make sure your application environment is performing optimally. Sysdig's out-of-the-box golden signals dashboards provide immediate visibility into what matters most without additional instrumentation.
Analysis of Syscall Capture Files
Capture in-depth system call activity to more effectively triage issues in your container environment even if containers are already killed. Reproduce step by step run processes and commands, reconstruct read and written files including logs, visualize traffic on network connections, and much more.
Metric Segmentation
Segmentation of key metrics to focus investigation. Other tools give you resource usage information to the container level. Sysdig goes deeper giving you resource usage information at the process level inside the container. In addition, our Explore tool provides the ability to inspect every aspect of your environment both physically and logically. Then Sysdig Monitor enriches those views including network connections with Kubernetes context, and event overlays.
How It Works
The Sysdig Agent uses an eBPF probe to capture in-depth system call activity to more effectively triage issues in your container and Kubernetes environment. Then, Sysdig enriches these metrics with host, cloud, and Kubernetes context to help you make sense of all that data. Sysdig helps you find the needle in the haystack even if containers are killed.
“Without Sysdig, we would have scattered visibility into our multiple databases. That would be atrocious for debugging issues. On top of that, Sysdig helps me to determine what is causing issues sooner. If I had to do my debugging manually with other tools, I would spend three to four more time debugging.”
Thyrus Gorges, DevOps Engineer, Pike13