Operate Kubernetes with a clear next step.
Connect a cluster, understand every feature, investigate safely, and resolve common problems without guessing.
Search across installation, product, security and troubleshooting guides.
Go from sign-in to useful data
Choose what you need
Getting started
Account setup, prerequisites, agent installation, verification, reconnect, upgrade and uninstall.
Start here if this is your first cluster →Product guide
Learn Overview, Applications, Dashboards, Krevo AI, Optimize, Deployments, Integrations, Teams & Access and SSO.
Understand the complete product workflow →Security, agent & RBAC
Agent architecture, exact Kubernetes permission boundaries, data handling, retention, tenant isolation and identity controls.
For security and platform reviews →Troubleshooting
Symptoms, causes, safe checks and fixes for common platform and agent problems.
Find a problem and its next safe action →Investigation benchmark
Read the dated, reproducible accuracy result, scenario coverage, scoring method and limitations.
Review the published evidence →Trust & evidence
Review verified product evidence, implemented controls, operational readiness and explicit claim boundaries.
Open the public trust report →Buyer trust package
Open the 90-second walkthrough, benchmark, security one-pager and exact agent RBAC from one page.
Review the complete package →Agent security review
Inspect RBAC, configuration, network, schema, redaction and security documents before installation.
Open the security documents ↗About KrevoPilot
Read the canonical product description, company ownership, product boundaries and official public resources.
Review the product facts →CrashLoopBackOff
Read the exit code to separate an application crash from a missing ConfigMap, a strict probe, a bad entrypoint or an OOM kill.
Diagnose a restart loop →OOMKilled and exit code 137
Separate a container memory limit from node pressure, then choose a limit from observed peak instead of doubling it.
Size a memory limit →ImagePullBackOff
Five causes that look identical until you read the registry error. Only one of them is fixed by a pull secret.
Read the registry error →Resource rightsizing
Requests decide capacity and cost, limits only cap it. How to size both from observed usage without causing OOM kills.
Size requests from data →No direct match
Try a shorter term, or open Troubleshooting for symptom-based guidance.
100% strict accuracy in the dated production-path run. This is a bounded benchmark, not a claim about every possible incident.
Still not sure where to start?
Use the troubleshooting guide for a current problem, or ask the KrevoPilot team to review your onboarding plan.