| name | env-troubleshoot |
| metadata | {"author":"rhoai3-demo","version":"2.0.0","platform-family":"rhoai","platform-baseline":"repo","ocp-baseline":"repo","skill-group":"Demo Environment"} |
| description | Diagnose and fix issues in the live rhoai3-demo AWS/OpenShift environment once active deploy and validate scripts exist. During the reimplementation, use this skill to rebuild troubleshooting coverage from legacy references. Use when a deployment stage fails, validate.sh reports errors, pods are in CrashLoopBackOff/Pending/Error/ImagePullBackOff, ArgoCD shows OutOfSync or Degraded, operators are not installing, GPU nodes are not joining, InferenceServices are not Ready, LlamaStack returns errors, Guardrails Orchestrator is unhealthy, MCP tools are failing, or the user reports any problem with their OpenShift AI environment. Also use when the user asks "why isn't X working?" for any demo component. Do NOT use for chatbot UI/prompt changes (use rhoai-chatbot-customization) or model evaluation workflows (use rhoai-model-evaluation).
|
RHOAI Troubleshooting
Structured diagnostic workflow for resolving issues with the RHOAI demo on the
active product baseline in docs/PLATFORM_BASELINE.md.
Reimplementation Status
The active implementation is being rewritten. No active deploy or validate
scripts exist yet. Treat legacy step names, validation commands, and diagnostic
patterns in this skill as reference material for rebuilding troubleshooting
coverage, not as active-project instructions.
Do not run scripts from backup/legacy-implementation-2026-06-09/ unless the
user explicitly asks to restore or inspect the legacy implementation.
When to Use
- A
validate.sh script reports failures
- A
deploy.sh script errors out or hangs
- Pods are stuck in CrashLoopBackOff, Pending, or Error
- ArgoCD shows OutOfSync, ComparisonError, or Degraded
- LlamaStack, Guardrails, or MCP tools are not working
Instructions
Never guess. Every diagnosis must be backed by official documentation or observable cluster state.
Step 1: Run the Validation Script When It Exists
./stage-YXX-slug/validate.sh
Check exit code: 0 = pass, 1 = failures, 2 = warnings only. During the
reimplementation, skip this check until active validation scripts are recreated.
Step 2: Consult Official RHOAI Documentation
Use official docs for the active baseline in docs/PLATFORM_BASELINE.md.
Focus on the relevant section by using .agents/references/red-hat-doc-map.yaml
and the stage family:
| Stage family | Primary docs route |
|---|
1xx | OCP, ODF, RHOAI install, DSCI/DSC, users/groups, accelerators, observability |
2xx | RHOAI model serving, model catalog, model registry, MaaS, RAG, guardrails |
3xx | RHOAI Llama Stack, Gen AI Studio, MCP, connected applications |
4xx | RHOAI AI Pipelines, MLflow, evaluation, monitoring, distributed workloads |
5xx | Edge or applied AI product docs selected by the stage PLAN.md |
Step 3: Gather Cluster State
Start with the readonly workflow in references/cluster-inspection.md, then
run targeted commands from references/diagnostic-commands.md for the failing
component.
Step 4: Match to Known Pattern
Check references/diagnostic-patterns.md for the symptom. Common patterns cover ArgoCD, operators, pods, LlamaStack, guardrails, and MCP.
Step 5: Apply Fix and Verify
Execute the smallest change that fixes the issue. Re-run the active
validate.sh to confirm when it exists.
Step 6: Update Knowledge Base
If this was a new issue, update the env-deploy-and-evaluate skill's references/deploy-notes.md with the new known issue.
Escalation Protocol
If unresolved after one diagnostic cycle:
- Document what was tried and observed
- Include exact error messages and
oc output
- Suggest manual checks
- Report to the user with full context
References
references/cluster-inspection.md
references/diagnostic-commands.md
references/diagnostic-patterns.md