I will troubleshoot kubernetes deployment and audit cluster setup
About this Gig
Is your Kubernetes workload stuck in CrashLoopBackOff, failing health checks, or rolling out unpredictably?
I provide evidence-driven Kubernetes troubleshooting and reliability audits for engineering teams and SaaS builders running workloads on AWS EKS, GCP GKE, or self-managed clusters.
What I Deliver (Scoped by Package):
- Basic (Workload Triage): Root-cause diagnosis of 1 failing Deployment, StatefulSet, or Job from your redacted manifests, kubectl describe output, and logswith tested YAML fixes.
- Standard (Reliability Audit): Audit up to 5 workloads covering CPU/memory requests & limits, liveness/readiness probes, HPA, PodDisruptionBudgets, RBAC, and NetworkPolicies.
- Premium (EKS/GKE & GitOps Review): Full cluster reliability, Argo CD / Rollouts drift & canary rollback, Prometheus/OpenTelemetry observability, and Workload Identity review.
Background & Boundaries:
My failure-oriented methodology is backed by independent GCP GKE & AWS EKS engineering benchmarks (auto-healing recovery, Argo CD drift self-healing, and scheduler race resolution). I do not perform unapproved live production changes; message me first for custom cluster execution.
Tools:
Kubernetes
•
Docker
•
Amazon EKS
•
Google Kubernetes Engine
Frameworks:
Npm
•
Terraform
•
Ansible
•
Chef
•
Crossplane
Programming language:
Bash
•
C
•
Go
•
Java
•
JavaScript
•
PHP
•
Python
•
Golang
Expertise:
Debugging
•
Configuration
My Portfolio
Other DevOps Engineering Services I Offer
FAQ
Do you need direct access to my live production cluster?
No. Most reviews and root-cause triages are completed safely offline using your redacted YAML/Helm manifests, kubectl describe outputs, and pod logs. If hands-on cluster verification is needed, we scope least-privilege read-only or staging access via a custom offer first.
How should I share configuration files and logs safely?
Always redact passwords, API tokens, TLS private keys, and .env secrets before uploading files. Never paste raw credentials into Fiverr messages or order requirements.
What if my issue spans multiple microservices or requires hands-on fixes?
Basic covers 1 workload and Standard covers up to 5 workloads. If your environment requires multi-service remediation or live cluster execution, message me with your scope and I will prepare a fixed-price custom offer with a clear rollback plan.
Can you guarantee zero downtime or 100% security after the review?
No honest engineer can guarantee zero downtime or absolute security. What I deliver are concrete, tested configuration fixes, prioritized risk findings, and documented verification steps so your team can eliminate known failure modes.

