100% Self-Hosted • Zero Data Leaks

Kubernetes pod crash alerting
with AI diagnostics.

Detect CrashLoopBackOffs, OOMKilled containers, and node resource bottlenecks in real time. Local, telemetry-free, and fully self-hosted. Catch crashes before your users do.

$curl -fsSL https://raw.githubusercontent.com/Akshatsainiaks/srevox-setup/main/setup.sh | bash

Interactive Incident Console Demo

Experience Live Incident Diagnosis

Click "Run AI Diagnosis" below to watch Srevox analyze pod logs and generate fix steps in real time.

Back to incidents list

CRITICAL● openCLUSTER: ASCA

simulated-payment-auth-api-914

Namespace: production

Incident Details

Crash ReasonCrashLoopBackOff

Cluster asca

Restart Count8

Namespaceproduction

Containerauth-api

Exit Code1

First Seen7m ago

Last Seen7m ago

Acknowledged Byadmin

Resolved Byadmin

Pod Labels

app=auth-apienv=productiontier=backend

AI Incident Diagnosis

diagnosing with logs

AI-Powered Root Cause Analysis

On-demand Kubernetes log parsing and zero-config automated fix generation.

Container Terminal Logs (auth-api)

2026-07-26T16:40:01.102Z [server] Starting Srevox Payment Authentication Service v2.4.1...
2026-07-26T16:40:01.345Z [server] Loading configuration from /etc/config/auth.json...
2026-07-26T16:40:02.012Z [database] Connecting to Primary PostgreSQL Database at postgres.prod.svc.cluster.local:5432...
2026-07-26T16:40:02.450Z [database] Database connection established successfully (pool_size=20).
2026-07-26T16:40:03.118Z [redis] Initializing Redis Cache Connection at redis-cluster.prod.svc.cluster.local:6379...
2026-07-26T16:40:08.125Z [redis] [ERROR] Redis Connection Timeout after 5000ms: ECONNREFUSED redis-cluster.prod.svc.cluster.local:6379
2026-07-26T16:40:08.126Z [server] [FATAL] ConnectionRefusedError: Failed to connect to cache cluster
    at RedisClient.connect (/app/node_modules/ioredis/built/redis/index.js:284:14)
    at async initializeCache (/app/dist/services/redis.js:42:5)
    at async bootstrap (/app/dist/index.js:18:3)
2026-07-26T16:40:08.127Z [server] [CRASH] Process exited with status code 1. State: CrashLoopBackOff (Restart count: 8)

Complete Observability Platform

Engineered for High-Compliance Teams

Everything you need to monitor workloads, isolate pod crashes, and automate root-cause remediation without sending metrics to third-party clouds.

Real-Time Incident Stream

Captures Kubernetes CrashLoopBackOffs, OOMKilled events, and LivenessProbe failures in under 1 second. Stream live pod logs directly in your browser.

AI Root-Cause Diagnostics

Analyzes stacktraces and container exit codes using OpenAI, Anthropic, or 100% local Ollama LLMs to provide immediate fix manifests.

Granular Alert Controls

Mute platform system alerts globally, configure custom Node CPU/Memory threshold rules, and target explicit Slack, Teams, or Email integrations.

Node & Telemetry Metrics

Historical CPU, Memory, Pod count, and Node health telemetry. Built-in background evaluator automatically triggers warning alerts before nodes crash.

Service Owner Routing

Assign services to explicit engineering team owners. Srevox routes incident alerts directly to responsible service maintainers automatically.

Audit Log & Governance

Full activity audit trails tracking user invites, alert channel edits, cluster configuration updates, and RBAC permission changes.

Docker Image Pulls

Flexible Cluster Connectivity

Two Connection Architecture Options

Choose between agentless Service Account token connections or lightweight cluster watcher agents.

Direct Service Account / Kubeconfig

Agentless

Connect remote EKS, GKE, AKS, or bare-metal clusters directly via API token or Kubeconfig. Srevox uses CoreV1Api to query pod logs and metrics without installing anything inside the cluster.

Zero pod footprint inside your cluster
Custom CA certificate & skipTLSVerify support

Srevox Cluster Watcher Agent

In-Cluster Daemon

Deploy a lightweight DaemonSet or Helm watcher inside your cluster. The agent streams Kubernetes events outbound via secure HTTPS webhooks to your Srevox server instance.

Sub-second incident event streaming
Air-gapped network outbound proxy support

Quick 3-Step Setup

Deploy Srevox on any Linux machine or server in under 2 minutes.

Run Setup Script

Execute the single-line bash script on your server to pull the required docker containers.

curl -fsSL https://raw.githubusercontent.com/Akshatsainiaks/srevox-setup/main/setup.sh | bash

Open http://YOUR_SERVER_IP:3000 in your browser and log in with default credentials.

admin@srevox.local / admin123

Connect Cluster & AI

Add your Kubernetes API token or Kubeconfig and select your preferred AI provider (Ollama or OpenAI).

✓ Ready for Monitoring

Frequently Asked Questions

Everything you need to know about operating Srevox in production.

Does Srevox need inbound firewall ports opened into my Kubernetes cluster?

No. Srevox supports agentless Service Account API queries and outbound agent webhook streaming. It requires zero inbound open ports or public load balancer endpoints inside your Kubernetes cluster.

Can I run Srevox completely offline in air-gapped environments?

How does System Alerts global muting work?

What user authentication and RBAC structure is supported?

Kubernetes pod crash alerting with AI diagnostics.