Alert System Tools Investigations

Overview

Klaudia's Alert System Tools enable seamless integration with your existing alerting and monitoring infrastructure. When investigating issues, Klaudia can automatically query your alert systems to correlate Kubernetes resource problems with active alerts, providing richer context during root cause analysis and troubleshooting sessions.

By connecting your alert providers to Klaudia, you unlock the ability to:

• Automatically correlate Kubernetes issues with relevant alerts
• Gain deeper visibility into the timeline of incidents
• Reduce mean time to resolution (MTTR) by accessing alert context directly within investigations
• Streamline troubleshooting by eliminating context switching between tools

Supported Alert Providers

Klaudia supports the following alert system integrations:

Prerequisites

Before Klaudia can use alert system tools, you must configure the appropriate integration for each provider in Komodor.

What Alert System Tools Can Do

General Capabilities

When connected to your alert providers, Klaudia can:

• Retrieve active alerts: Query current alerts affecting specific services, namespaces, or clusters
• Access alert history: Look up historical alert data to understand incident patterns
• Correlate alerts with Kubernetes events: Match timing of alerts with deployment changes, pod restarts, and other cluster events
• Enrich investigations: Automatically include relevant alert context in root cause analysis
• Identify alert sources: Determine which monitoring rules triggered during an incident

Provider-Specific Capabilities

Datadog Alerts

• Query monitor statuses and alert states
• Retrieve triggered alert details with full context
• Access related metrics and traces linked to alerts
• Correlate APM data with Kubernetes workload issues

Sentry

• Retrieve error details, stack traces, and breadcrumbs
• Access issue frequency and user impact data
• Correlate application errors with pod crashes or restarts
• Link release information to deployment changes

OpsGenie

• Query active incidents and their status
• Retrieve on-call information for escalation context
• Access alert notes and responder actions
• Correlate incident timelines with Kubernetes events

Grafana Alerts

• Query alerting rule states and thresholds
• Retrieve alert annotations and labels
• Access dashboard links for visual context
• Correlate metric-based alerts with resource conditions

Custom Webhook

• Query any custom alerting system via standardized API calls
• Support for various authentication methods
• Flexible payload formatting for different alert schemas

When Klaudia Uses Alert System Tools

Klaudia automatically leverages alert system tools in the following scenarios:

Root Cause Analysis (RCA) Investigations

When Klaudia performs an automated RCA on an unhealthy resource, it queries connected alert systems to:

• Identify any alerts that fired around the time the issue started
• Correlate application-level errors (from Sentry) with infrastructure issues
• Match monitoring alerts (from Datadog/Grafana) with deployment or configuration changes

Troubleshooting Unhealthy Resources

When you ask Klaudia to investigate a failing deployment, crashlooping pod, or degraded service:

• Klaudia checks for active alerts related to the affected resource
• Alert context is included in the investigation summary
• Historical alerts help establish whether this is a recurring issue

Chat Sessions with Klaudia

During interactive chat sessions, you can:

• Ask Klaudia to check for alerts affecting a specific service
• Request alert history for a particular time window
• Ask Klaudia to correlate a Kubernetes issue with your monitoring data

Example prompts:

• "Are there any Datadog alerts for the payment-service?"
• "Check Sentry for errors in the checkout namespace from the last hour"
• "What OpsGenie incidents are currently open for the production cluster?"