SEC.01
The playbook
The run starts read-only: the recon probes decide what happens next. The steps that follow are adapted to the distribution and state found — a real machine is rarely clean.
- 01reconZFS already present (tools + module) + version
- 02recondo the disks exist? consistent sizes?
- 03reconsignature on a disk = data → NEVER overwrite
- 04reconsystem disk to exclude; does a pool with the same name already exist?
- 05reconstable by-id paths of the disks (to use for the pool)
- 06recondistro and kernel; Secure Boot active (unsigned DKMS module → blocking)
- 07reconRAM: 2 GB minimum recommended for ZFS
- 08actPrepare the ZFS repository (Debian and RHEL only)verify
- 09actInstall ZFSverify
- 10actLoad the moduleverify
- 11actSimulate the pool creationverify
- 12actCreate the ${pool_name} pooldestructiveverify
- 13actCheck the mountverify
SEC.02
What it verifies
Each step ships its own check. The run only moves on when the check passes, so what the recipe reports as done is what is actually on the server.
- 01verifyPrepare the ZFS repository (Debian and RHEL only)Verified by
rpm -q zfs-release 2>/dev/null || apt-cache policy zfsutils-linux | head -3 - 02verifyInstall ZFSVerified by
command -v zpool - 03verifyLoad the moduleVerified by
lsmod | grep -c '^zfs ' - 04verifySimulate the pool creationVerified by
zpool list -H -o name ${pool_name} 2>/dev/null || echo absent - 05verifyCreate the ${pool_name} poolVerified by
zpool status -x ${pool_name} - 06verifyCheck the mountVerified by
zfs get -H -o value mounted ${pool_name}
SEC.03
Guardrails
Non-negotiable instructions the run must respect, whatever it finds on the server.
REFUSE to create a pool on disks carrying a signature or on the system disk. Never zpool create -f. Prefer /dev/disk/by-id paths. Ask for explicit confirmation before any zpool create. Secure Boot blocking → stop, do not disable it.
SEC.04
Variables
What you choose before launching — the recipe fills in the rest.
| Name | Meaning |
|---|---|
| ${pool_name} | Pool name |
| ${devices} | Disks |
| ${raid} | RAID |
SEC.05
How a recipe runs in Servor
Servor is zero-knowledge: the platform cannot execute anything on its own. When you launch this recipe, a run reads the reconnaissance, plans the steps and asks for your approval. Each approved command is signed by your browser — with a key the server never sees — then relayed verbatim to the agent, which verifies the signature locally before executing.
search by intent — type “firewall”, get ufw
- 01Read-only reconnaissance before any change
- 02Every command signed by your browser, approved by you
- 03Each step verified before the run moves on, every run kept in the history
SEC.06
Deploy on Servor button
Maintain ZFS pool or a project that needs it? Put this button in your README or on your site: one click and your users install it on their own server.
SEC.07
More storage & backup recipes
- 01Swap fileCreates a persistent swap file of the chosen size. Destructive: modifies /etc/fstab.
- 02restic backupsrestic + a scheduled, encrypted backup of a folder to a repository.
- 03MinIOS3-compatible MinIO object storage via Docker, local-only. Community project archived: frozen image.
- 04Garage (S3 storage)Garage (Deuxfleurs) S3-compatible object storage via Docker, single-node and local-only, with a first bucket and its key. Maintained alternative to MinIO.
- 05Samba shareSamba file server (SMB). Destructive: creates a share and a user.
Verified recipe
Run this recipe on your server
Connect a server, launch the recipe, and approve each step as Servor executes and verifies it. Free plan, no card required.