Runbooks

Documentation > Specialized Systems > OpenStack Cloud Platform > Runbooks

Runbooks

44 step-by-step operational runbooks for the GSD Skill Creator OpenStack deployment, organized by service with dual navigation indexes for goal-oriented lookup (find by task) and incident-response lookup (find by symptom). Every runbook follows the standard format: preconditions, numbered procedure steps, TAID verification criteria, and rollback plan. Runbooks are aligned with CAP-137 NASA SE Documentation traceability standards and cross-reference the CAP-138 V&V Infrastructure compliance matrix.

Find by Service

Browse runbooks by OpenStack service area. Each service group contains runbooks covering both routine procedures and incident response for that service's operational domain.

  • General Platform (4 runbooks) — Platform health check, credential rotation, certificate renewal, and Kolla-Ansible reconfiguration procedures applicable across all services.
  • Keystone — Identity (5 runbooks) — User and project management, token provider reconfiguration, LDAP sync failure recovery, application credential rotation, and service catalog repair.
  • Nova — Compute (6 runbooks) — Instance evacuation, live migration, compute node maintenance mode, hypervisor fence-and-recover, flavor management, and scheduler filter reconfiguration.
  • Neutron — Networking (6 runbooks) — Network agent restart, OVN/OVS database repair, floating IP reassignment, security group flush, router HA failover, and VLAN trunk recovery.
  • Cinder — Block Storage (5 runbooks) — Volume backend failover, snapshot cleanup, backup restoration, storage pool rebalancing, and Ceph OSD failure recovery.
  • Glance — Image (4 runbooks) — Image import and conversion, corrupt image cleanup, store backend migration, and image metadata repair.
  • Swift — Object Storage (4 runbooks) — Ring rebalancing, replication lag recovery, container database repair, and capacity expansion.
  • Heat — Orchestration (4 runbooks) — Stack update rollback, stuck resource cleanup, auto-scaling group reset, and template validation workflow.
  • Horizon — Dashboard (3 runbooks) — Session store flush, SSL certificate replacement, and panel customization recovery.
  • Kolla-Ansible — Deployment (3 runbooks) — Full redeployment, container image upgrade, and configuration rollback.

Find by Symptom

Use this index when responding to an active incident and the affected service is not immediately obvious. Symptoms are grouped by observable impact category.

  • Authentication failures — Users cannot log in, API calls return 401 → Keystone token provider, LDAP sync, credential rotation runbooks
  • Instance launch failures — Nova scheduler errors, no valid host, resource exhaustion → Nova compute node, scheduler filter, evacuation runbooks
  • Network connectivity loss — Instances unreachable, floating IP not routing, security group lockout → Neutron agent restart, OVN repair, floating IP runbooks
  • Volume attach failures — Cinder timeout, multipath errors, backend unreachable → Cinder backend failover, OSD recovery runbooks
  • Dashboard inaccessible — Horizon 502/503, session errors, SSL mismatch → Horizon session flush, SSL replacement runbooks
  • Orchestration stack stuck — Heat stack in UPDATE_FAILED or DELETE_FAILED state → Heat rollback and resource cleanup runbooks
  • Image operations slow or failing — Glance upload timeout, format conversion errors → Glance store backend, image import runbooks
  • Object storage degraded — Swift replication lag, container not found errors → Swift ring rebalancing, replication recovery runbooks

Standard Runbook Format

Every runbook in this collection uses the following structure to ensure consistent execution and traceability:

  • Preconditions — Required access level, service state checks, and safety confirmations before beginning the procedure
  • Step-by-step procedure — Numbered actions with exact commands, expected outputs, and decision branches for error conditions
  • Verification criteria — TAID-classified checks confirming the procedure completed successfully
  • Rollback plan — Actions to reverse the procedure if verification fails, restoring the previous known-good state

Related Documentation

  • Operations Manual — Full per-service operational procedures providing the service context for these runbooks
  • Verification & Validation — Compliance matrix and safety-critical test suite that runbook verification steps reference

Best for: On-call operators responding to incidents, SysAdmins executing planned maintenance procedures