A full clone of the ZimaOS system SSD is not the only—or necessarily the most important—recovery asset. The source user had a 512 GB OS NVMe plus four 2 TB NVMe drives planned as RAID 5 and wanted to know whether losing the OS drive would endanger the array.
Current ZimaOS recovery design gives a clearer answer than the original 2025 thread: ZimaOS uses dual system slots for quick boot recovery, data lives separately from the system partitions, and IceWhale now documents backing up local-storage.db so RAID configuration can be restored after a system reinstall. The highest-value backup target remains your actual data and application state.
ZimaOS Has a Dual-Partition System Recovery Design
Current IceWhale System Recovery documentation says ZimaOS uses Slot A and Slot B, each around 6 GB, so a user can boot the alternate partition when one system slot fails.
This handles certain software/boot failures without replacing the SSD.
Use the current ZimaOS dual-slot recovery procedure before assuming every boot issue requires a full disk restore.
A Dead System SSD Is Different from a Broken System Slot
If the NVMe itself has physically failed, both A/B slots are gone with it. The recovery path is then a replacement system disk plus a fresh ZimaOS installation.
The goal is to make that reinstall able to rediscover or restore the data/storage configuration without treating the RAID members as blank disks.
Back Up local-storage.db Before You Need It
IceWhale's current RAID rebuild guide tells users to save:
/ZimaOS-HD/.casaos/db/local-storage.dbbefore reinstalling the system. After the new ZimaOS install, the guide has the user rename the new database, upload the backed-up one to the same location, and reboot.
Follow the official RAID rebuild-after-reinstall workflow rather than editing the database manually.
Do Not Recreate the RAID Just Because a Fresh OS Shows Disks Differently
RAID member disks can still contain valid RAID metadata and data even when a new system installation no longer has the previous ZimaOS storage database.
If those disks contain important files, do not click Create RAID, format them, or zero metadata until the recovery path is understood.
Back Up AppData and Application Databases
The 2025 community reply simplified application state as living under /DATA/AppData. Current ZimaOS makes that separation clearer but also supports different app-data locations and custom bind mounts.
Back up persistent application folders and use application-aware exports for databases such as Nextcloud, Immich, or other services where a folder copy alone may not be enough.
Current ZimaOS app-storage path guidance helps identify what survives container recreation.
A Full OS Clone Can Still Be Useful, but It Has Tradeoffs
A block-level clone can reduce recovery time if it is recent, consistent, and created with a method that correctly captures the live filesystem. But a clone also becomes stale quickly as application settings, storage metadata, and OS updates change.
If the system disk is actively writing while the image is captured, consistency matters. Do not assume “copy every block while running” automatically creates a reliable restore image.
Your Data Backup Matters More Than the Replaceable OS
Current ZimaOS includes a Backup app that can protect data to local disks, another NAS, USB, or cloud and supports scheduled/versioned workflows.
Use the current 3-2-1 backup strategy for the files that cannot simply be reinstalled.
A Practical System-Drive Failure Plan
- Keep a recent copy of
local-storage.dboutside the OS drive. - Back up application state and user data independently.
- Document custom app volumes and Compose stacks.
- Know how to boot the alternate ZimaOS system slot.
- If the SSD physically dies, replace it and reinstall ZimaOS.
- Restore the storage database using the official procedure.
- Verify the RAID before making any destructive storage change.
- Reinstall/reconnect applications to their persistent data.
OS Drive Backup FAQ
Does a ZimaOS boot failure always mean the OS SSD must be replaced?
No. Current ZimaOS has dual system slots, and the alternate slot can recover some system failures.
What small file is especially useful before a system reinstall?
IceWhale's current RAID recovery guide tells users to back up local-storage.db.
Is RAID 5 itself a backup of the data?
No. RAID protects availability after disk failure; keep independent backups for deletion, corruption, theft, and disaster.
