When your app breaks, how long does it take to find root cause?

Now imagine a tool that gives it to you instantly!

Zebrium is machine learning that finds root cause automatically

Zebrium uses machine learning to automatically find root cause. It works by detecting hotspots of abnormally correlated anomalous patterns across your logs and metrics.

It does not require manual training and works with any application. To get started, just install a lightweight collector or fork a copy of your logs. Accurate ML-driven RCA occurs within the first day.

Automated root cause analysis (RCA)

With an incident management tool

The process is seamless - when an incident is created in PagerDuty, Opsgenie or Slack, Zebrium automatically updates the open incident with a root cause report. More details here.

Without an incident management tool

When your app breaks, view the Zebrium incident summary list. Click into a detailed root cause report to see the log lines and correlated metrics anomalies that explain what happened.

Proactive detection for engineering

Zebrium constantly scans and automatically creates incident RCA reports for abnormally correlated anomalous patterns across multiple data streams. Engineers can use these reports to improve product quality and proactively fix latent bugs, before they manifest as production P1 incidents. At Zebrium, our VP of engineering swears by this capability!

Use standalone or to augment your log manager

Augment your log manager - You can continue using your existing logger for aggregation, dashboarding and search. But, when you have a problem, use Zebrium to find root cause without the need for manual hunting.

Use with the Elastic Stack - The ZELK Stack option lets you use Zebrium in conjunction with Elastic Stack to catch software incidents and see root cause right inside Kibana. Integration is simple and doesn't require endpoint reconfiguration.

Zebrium as a log manager - Our log management and monitoring UI is optimized for Developers and DevOps with a focus on troubleshooting. Beyond ML-driven RCA, it provides easy incident drill-down, plus a full suite of aggregated log exploration features including regex search, filtering, one-click charting, alerting and more.

Plain language root cause summaries

Zebrium produces concise root cause reports with relevant log lines and correlated metrics anomalies. Now you can turn these into plain language summaries, thanks to the OpenAI GPT-3 language model.

This feature is currently in private beta. Click here to learn more or request beta access.

Platform Security

Zebrium adheres to industry best practices for security including: encryption of data in flight, AES-256 encryption of data at rest, granular removal of sensitive records or fields, secure isolation of customer data and an option for a dedicated instance and VPC. All customer data will be deleted upon termination of service or upon request. Read our security policy here.

Supported Platforms

VMware, Windows and more available by using our Logstash integration.

Sweetwater, the world's leading music technology and instrument retailer, relies on Zebrium to detect and root cause production incidents across logs and Prometheus metrics, and to manage all their production logging.

“We were getting very odd error messages from one of our processes. Soon after installing Zebrium, it detected an incident that let me quickly figure out what the actual issue was.” – Yosef Deray, Lead Software Engineer @ Iralogix

“We used Litmus Chaos Engine for K8s to induce failures in our OpenEBS platform. Zebrium automatically detected every single failure and also identified its root cause.” - Murat Karslioglu, VP of Product Mgmnt @ MayaData

Experience Zebrium for yourself

Try with a sample microservices app

Our simple guide and video shows you how to setup a minikube Kubernetes cluster, install a cloud-native app, break the app and then see Zebrium detect root cause. The entire process takes 20-30 minutes.

Try with your own data

It takes less than 5 minutes to sign-up for an account and install our log and metrics collector. The rest is automatic. For best results, you will need to experience a real or induced failure in your environment.