FIELD MANUAL 001 // HELP DESK → INFRASTRUCTURE
Get out of the queue.
Fix the stuff behind it.
Hands-on break/fix material for help desk, NOC, desktop support, and junior sysadmin people who are tired of being told to “learn cloud” and would rather troubleshoot something real.
$ kubectl get pods -n breakfix-01
NAME READY STATUS RESTARTS
api-65c877c9bd-7qgnm 1/1 Running 0
$ kubectl get endpointslice -n breakfix-01 -l kubernetes.io/service-name=api-svc
NAME ADDRESSTYPE PORTS ENDPOINTS
api-svc-7m2zt IPv4 80 <none>
$ kubectl get pods -n breakfix-01 --show-labels
api-65c... Running app=api
$ kubectl describe svc api-svc -n breakfix-01
Selector: app=backend
> theory: the Service selects zero pods
> status: investigate, then fix
FREE SAMPLER // NO SIGNUP
Three things are broken.
You get to find out why.
These are complete labs, not teaser screenshots. You get the bad config, hints, the full solution, and cleanup files. If they are useless, you have spent zero dollars and learned something important about our business model.
Service has no endpoints
Healthy pods, dead Service. Diagnose the selector mismatch before blaming DNS.
LAB 02 · EASY · 5–10 MINPending forever
A scheduler refusal hidden behind a perfectly healthy node. Read the event before guessing.
LAB 03 · MEDIUM · 10–15 MINPVC never binds
Enough storage exists. It is still the wrong storage. Trace what the claim actually requires.
THE ESCALATION PATH
Coming from support?
You probably do not need to start over. The useful path is Linux, networking, scripting, containers, one cloud, IaC, then orchestration — with troubleshooting reps the whole way.
Read Help Desk → Cloud Infrastructure →Already know the basics?
Good. Skip the motivational speech and open the broken systems. If the free labs are too easy, that tells you more than a certificate badge does.
Open the free labs →FIELD MANUAL 001 // FULL PACK
Kubernetes Break/Fix Labs
10 labs · $10 · one-time
The free three plus seven more scenarios covering NetworkPolicy, probes, ConfigMaps, RBAC, CronJobs, ResourceQuota, and startup/readiness/liveness interactions.
- Broken manifests you diagnose and repair yourself
- Detailed solutions explaining what failed and why
- Cleanup scripts and reusable lab files
- No recurring subscription and no account to maintain
INCIDENT: BF-007
STATE: INTENTIONALLY BROKEN
Symptoms:
pod: Running
readiness: 0/1
service: reachable
user traffic: dead
Known facts:
- container started
- process exists
- probe disagrees
Your job:
1. inspect
2. form a theory
3. prove it
4. fix the smallest thing
WHY THIS EXISTS
Learn troubleshooting,
not incantations.
Cloud training has no shortage of videos telling you where to click. That is useful right up until something breaks. Escalation Path Labs is built around deliberately broken systems, observable symptoms, and the process of figuring out what actually happened.
The material is for working IT people who already know how to follow a ticket, use a terminal, and search an error message, but want the judgment required to own more of the stack.