Skip to content

[Bug] CRITICAL UPDATE: Real-world Bug & Urgent Feature Request Expansion (Why Pool Export/Import GUI is Mandatory) #519

Description

@moratac

🚨 CRITICAL UPDATE: Real-world Bug & Urgent Feature Request Expansion (Why Pool Export/Import GUI is Mandatory)

Following up on my original request for advanced storage management, I recently encountered a severe storage system lockup on ZimaOS v1.6.1 that solidifies why the current storage architecture needs a major overhaul.

Currently, the ZimaOS Storage Panel only provides a destructive 【Format and Break】 option. There is absolutely no GUI workflow to safely Disconnect, Export, or Import an existing pool without wiping the data. This missing lifecycle feature creates a fatal flaw when users physically migrate arrays from standard Linux environments (like Proxmox VE or traditional BTRFS NAS systems) into bare-metal ZimaOS.

💥 The Real-world Failure Scenario (mdadm over Btrfs Migration Bug)

When attempting to migrate a healthy 4-HDD mdadm RAID 5 array with a Btrfs filesystem layer directly into ZimaOS, the system's strict Immutable OS sandbox and the lack of an Import/Export GUI completely broke the storage management layer:

  1. Web UI Desync & 500 Internal Server Error: The Storage Control Panel detected the physical drives but completely choked on the legacy partition metadata. Clicking into the Files App threw a /media/pool: storage not mount error. Because there is no "Disconnect/Import" toggle, attempting to format or manage the ghost pool directly crashed the backend, rendering a persistent 500 Internal Server Error pop-up in the Web UI.

  2. Severe OS Sandbox / Kernel Node Isolation:
    Upon debugging via SSH (tonyadmin), cat /proc/mdstat confirmed that the Linux kernel successfully assembled the array flawlessly as healthy ([4/4] [UUUU]). However, due to ZimaOS's aggressive sandboxing, the system completely omitted the creation of the block device node (/dev/md1 or /dev/md127 did not exist).
    Even though a symlink existed at /dev/block/9:1 -> ../md1, running mount manually returned special device does not exist.

  3. Total CLI Lockout:
    Because it is an Immutable OS, standard administrative commands like sudo su - are completely blocked (command not found). Users are left entirely trapped in a loop where the GUI throws 500 errors and the CLI is too castrated to manually rescue or mount the data.


🎯 Critical Expansion Request: Native Pool Import / Export (Disconnect) Wizard

This issue proves that traditional RAID concepts managed by rigid background scripts are failing the users. To solve this, ZimaOS desperately needs to implement a Native Pool Lifecycle Wizard in the GUI, similar to how ZFS handles pools:

  • 【Disconnect / Export Pool】 (Safe Unmount): Allow users to safely export/disconnect a Btrfs/ZFS pool via the Web UI without wiping the hard drives. This spins down the array, cleans up the OS system cache, removes host-specific UUID associations, and prepares the physical disks to be safely pulled out and migrated to another ZimaOS box or any standard Linux system.
  • 【Scan & Import Pool】 (Foreign Array Adoption): A visual scan wizard that detects foreign or migrated storage arrays (whether standard mdadm, native Btrfs, or ZFS pools

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions