| name | openshift-lvm-storage |
| description | Use when planning, installing, configuring, validating, upgrading, expanding, shrinking, backing up, restoring, maintaining, or troubleshooting LVM Storage (LVMS) on OpenShift/OKD — including Single Node OpenShift, multi-node clusters, volume group provisioning, thin pool configuration, TopoLVM CSI filesystem and raw block volumes, volume group expansion, disk management, StorageClass defaulting, and lifecycle safety. |
OpenShift LVM Storage Lifecycle
Use this skill as a lifecycle router for LVM Storage (LVMS) on OpenShift/OKD. Do live discovery first, choose the relevant reference runbook, and write an actionable plan or command sequence with explicit safety gates.
Routing
- Discovery, install, or prerequisite planning: start with
references/install-and-preflight.md.
- Volume group provisioning, thin pool configuration, and disk management: use
references/volume-group-provisioning.md.
- Filesystem volumes (ext4/xfs) via TopoLVM CSI: use
references/filesystem-volumes.md.
- Raw block volumes via TopoLVM CSI: use
references/block-volumes.md.
- Volume group expansion, disk addition/removal, or shrink: use
references/expand-shrink.md.
- LVMS operator or TopoLVM CSI upgrade: use
references/upgrade.md.
- Backup target, restore, or DR planning for local storage: use
references/backup-restore-dr.md.
- Node maintenance, disk removal, VG cleanup, operator uninstall, or cluster cleanup: use
references/maintenance-uninstall.md.
- Validation, hardening, post-reboot drift checks, or troubleshooting: use
references/validation-hardening.md.
Core Safety Rules
- Never run or recommend
pvcreate, vgcreate, vgremove, lvremove, wipefs, partitioning, or LVMS uninstall until the user gives explicit destructive confirmation for the exact target disk and intent.
- Before every destructive disk action, require
readlink -f, lsblk -f, pvs, vgs, lvs, and wipefs -n evidence for the target.
- Use stable
/dev/disk/by-id/* or /dev/disk/by-path/* paths for destructive block-device targeting. Never use /dev/sdX, /dev/nvmeXnY, or guessed paths.
- Warn that
LVMCluster changes and node-level VG operations can disrupt running workloads. On SNO, warn that node maintenance removes API access until the single node returns.
- Keep exactly one default StorageClass unless the user explicitly requests another policy.
- On SNO, use one TopoLVM replica or document the single-node constraint. Do not copy SNO single-node defaults into multi-node production plans without explicit direction.
volumeBindingMode: WaitForFirstConsumer is required for TopoLVM so capacity-aware scheduling works. Do not change this to Immediate on TopoLVM StorageClasses.- Do not downgrade the LVMS operator. For major upgrades, verify all
LogicalVolume CRs are healthy and all PVCs are bound before upgrading.
- Document the difference between LVMS operator version and TopoLVM CSI driver version when applicable.
- Thin pool over-provisioning risks running out of physical space. Monitor thin pool usage and set alerts before the pool reaches critical capacity.
Required Source Checks
For install, upgrade, and uninstall operations, verify the current LVMS / TopoLVM docs and release notes when network access is available. Use pinned version docs for commands and live cluster discovery for the installed version; do not assume the latest version is the target unless the user asks for it or the cluster already runs it.
For OpenShift channel, patch, or one-hop upgrade-path questions, use openshift-versions. Release availability is not cluster upgrade readiness and is not LVMS product compatibility.
Inputs To Collect
- Cluster topology: SNO or multi-node, OpenShift/OKD version, node roles, target MachineConfigPool.
- Current LVMS state: absent, operator installed,
LVMCluster CR exists, installed versions, VG health, thin pool usage, StorageClasses, LogicalVolume CRs.
- Target lifecycle action: install, VG provisioning, filesystem volume, block volume, expand/shrink, upgrade, backup/restore, maintenance, uninstall, validation, hardening, or troubleshooting.
- Target disk inventory by node, preferably
/dev/disk/by-id/*, plus whether each disk may be destroyed and whether it is raw or already has LVM signatures.
- Thin provisioning policy (over-provisioning ratio,
thinLvPercent, sizePercent, spare GB), default StorageClass intent, filesystem type (ext4/xfs), volume mode (filesystem vs block), backup/restore requirements, and maintenance window constraints.
Output Expectations
- Start with discovered facts, assumptions, and safety gates.
- Name the exact reference runbook(s) used.
- Separate read-only discovery from mutating or destructive actions.
- Show commands with placeholders for cluster-specific values instead of fabricating node, disk, or version identifiers.
- Include post-change validation and rollback or stop conditions.
- For production work, include backup, monitoring, and restore validation guidance.
