Add scripterase-zfs.sh which does fulla / metadata erase and verification

README.md

@@ -4,6 +4,35 @@
 a device as a ZFS vdev. It is intended for decommissioning or repurposing a
 device after it has been cleanly removed from its pool.
 
+## Unified Eraser
+
+`erase-zfs.sh` is the recommended entry point. It performs a full-device wipe
+by default and uses 10 overwrite passes. Select `--metadata` when only the
+partition tables and ZFS vdev labels should be overwritten.
+
+```bash
+chmod +x erase-zfs.sh
+sudo ./erase-zfs.sh /dev/sdX
+sudo ./erase-zfs.sh --metadata /dev/sdX
+sudo ./erase-zfs.sh --passes 3 /dev/sdX
+sudo ./erase-zfs.sh --source random --verify /dev/sdX
+```
+
+`--wipe` explicitly selects the default full-device mode. `--source zeroes`
+is the default, while `--source random` uses `/dev/urandom`. `--passes` accepts
+a positive integer and defaults to 10. `--verify` runs after wiping and checks
+`wipefs` signatures, readable ZFS labels through `zdb -l`, and ZFS pool
+discovery. A full zero wipe additionally compares every readable byte against
+`/dev/zero`. Random data cannot be byte-for-byte checked after it is written,
+so its verification is limited to the metadata and ZFS recovery checks. The
+verifier marks unavailable or ambiguous ZFS discovery as a failure rather than
+declaring the wipe successful.
+
+Pool discovery searches the target device's parent directory. Any reported
+pool may use a neighboring device, so inspect its vdev paths before treating
+the result as evidence about the erased target. A failed verification exits
+with a nonzero status and leaves all evidence in place for investigation.
+
 ## What It Erases
 
 The script performs 10 overwrite passes over the first 4 MiB and the last