Skip to content
CloudOps
Newsletter
All prompts
AI for OpenStack Difficulty: Intermediate ClaudeChatGPT

OpenStack Security Hardening Prompt

Harden OpenStack — service-to-service TLS, secure RBAC, secure defaults audit, CVE response, audit logging.

Target user
OpenStack security engineers
Difficulty
Intermediate
Tools
Claude, ChatGPT

The prompt

You are a senior OpenStack security engineer who has hardened production deployments — TLS everywhere, secure RBAC, CIS-aligned configurations, CVE patching processes.

I will provide:
- Current security posture (TLS state, RBAC config, audit settings)
- The OpenStack release
- Specific concern (TLS audit, RBAC review, CVE response, hardening checklist)

Your job:

1. **TLS audit**:
   - Service-to-service TLS (Keystone, Nova, Neutron talking)
   - API endpoints with valid certs
   - RabbitMQ TLS
   - MariaDB TLS (Galera replication)
   - Cert rotation procedure
2. **RBAC**:
   - Policy files for each service
   - Default policies vs custom
   - Service accounts (heat, octavia) with minimum scope
   - System scope vs project scope
3. **Network exposure**:
   - Management network isolation
   - Public-facing endpoints (Keystone, Nova, Neutron API)
   - Tenant networks segmentation
4. **Secure defaults**:
   - `nova.conf`, `neutron.conf`, etc. with hardened values
   - Disabled deprecated mechanisms
   - Token format (fernet), encryption (aes256)
5. **Audit logging**:
   - Service-level audit events
   - Forward to SIEM (CADF format)
   - Long-term retention
6. **CVE response**:
   - Monitor security-impact list
   - Test patches in staging
   - Coordinated patching across services
   - Rolling updates without downtime
7. **For supply chain**:
   - Image signing for instances (via Barbican + Glance)
   - Container image verification (if Kolla)
   - SBOM tracking
8. **Multi-tenancy**:
   - Network isolation between tenants
   - Storage isolation (separate pools or LUKS per volume)
   - Compute isolation (host aggregates for sensitive tenants)

Mark DESTRUCTIVE: changing token format without coordination (existing tokens invalid), tightening RBAC during peak (locks legitimate users), disabling services to "secure" (loses functionality).

---

Posture: [DESCRIBE]
Release: [DESCRIBE]
Concern: [DESCRIBE]

Why this prompt works

Hardening is multi-faceted. This prompt walks the areas.

How to use it

  1. TLS first — everywhere.
  2. RBAC audit with system scope adoption.
  3. CVE process documented.
  4. Test in staging before all changes.

Useful commands

# TLS check
openssl s_client -connect <endpoint>:443 -servername <fqdn> < /dev/null | grep -E "subject|issuer|Not After"

# Service endpoints
openstack endpoint list
openstack endpoint show <id>

# Token introspection
openstack token issue

# Policy audit
openstack role assignment list --names --effective | head -20
sudo cat /etc/keystone/policy.yaml | head -50
sudo cat /etc/nova/policy.yaml | head -50

# Audit logs
sudo journalctl -u keystone -n 100 --no-pager | grep -i audit

# Pending CVE check (release notes)
# https://wiki.openstack.org/wiki/OSSA

# RabbitMQ TLS check
sudo rabbitmqctl status | grep -i tls
sudo rabbitmq-diagnostics listeners

Hardening checklist

  • All API endpoints TLS-enabled, valid certs
  • Service-to-service TLS (especially RPC)
  • RabbitMQ TLS + auth required
  • Galera replication over TLS
  • Keystone fernet keys rotated regularly
  • Audit logs forwarded to SIEM
  • No default credentials anywhere
  • Service accounts have minimum scope
  • Custom RBAC policies version-controlled
  • Compute hosts patched + tracked
  • Image signing enforced
  • Tenant network isolation verified
  • CVE patching runbook tested
  • Backups encrypted at rest

Common findings this catches

  • HTTP endpoints exposed — switch to HTTPS only.
  • Self-signed certs not trusted — proper PKI.
  • Service accounts with admin — scope to specific roles.
  • No audit log forwarding — compliance gap.
  • Default policy not customized — review for org needs.
  • Old token format (uuid) — migrate to fernet.
  • RabbitMQ unencrypted — major risk.

When to escalate

  • CVE requiring fleet patch — coordinated response.
  • RBAC overhaul — staged migration.
  • Suspected compromise — incident response.

Related prompts

Newsletter

Free: the DevOps AI Incident-Triage Cheat Sheet

Subscribe and we’ll send you the one-page cheat sheet — plus weekly AI prompts, automation ideas, and tool reviews for infrastructure engineers. One email a week. No spam, unsubscribe anytime.

  • AI Incident-Triage Cheat Sheet (PDF)
  • Access to 1,603 DevOps AI prompts
  • One practical workflow email per week