I tried to make my Samsung SSD 840 EVO 120GB into a boot drive, but just get a 'TOO-SMALL' Error.

I tried formatting it numerous times, but still the same issue. Any idea into why this is, please?

You are likely to get better informed feedback if you attach your system’s diagnostics (with everything in the one zip file) to your next post in this thread. It is always a good idea when asking questions to supply your diagnostics so we can see details of your system, how you have things configured, and the current syslog.

Here are my diagnostics; baldrick-diagnostics-20261003-2048.zip

The SSD is fine. Unraid sees it as a full 120 GB disk, and that's well above the minimum needed for internal boot.

The problem is the U-Manager Companion plugin. It changes how the Unraid API gets disk information, and a bug in that change reports each disk's size as a few bytes instead of its real size. For example, the wizard probably shows your SSD as "111 B". So every disk fails the size check and gets marked TOO_SMALL, no matter how you format it.

To fix it:

1. Remove U-Manager Companion from the Plugins page.

2. Reboot. This step is required: the plugin's changes stay in memory until you reboot.

3. Make sure the SSD isn't assigned to a pool, then run the Internal Boot setup again.

You can reinstall the plugin after internal boot is set up, but disk sizes reported through the API will stay wrong while it's installed. You should also report the issue to the plugin's author.

Plugin author here, sorry for the trouble. JorgeB's diagnosis is spot on: the companion replaces how the API lists disks (so it doesn't wake spun-down drives), and that code reported sizes as a few bytes instead of the real value, so every disk failed the Internal Boot size check.

This is fixed in U-Manager Companion 2026.10.04.1, released today. Updating the plugin from the Plugins page is enough: the fix applies immediately and no reboot or uninstall is needed. If you already removed it to set up internal boot, you can reinstall it and disk sizes will now be reported correctly.

GitHub issue for reference: https://github.com/jandrop/u-manager-companion/issues/3

Longer term, I'd rather not patch the API for this at all. I've opened a work intent on the Unraid API to fix the spin-up problem upstream (https://github.com/unraid/api/issues/2105, building on https://github.com/unraid/api/issues/2018). If it's accepted and lands, the companion can drop this patch and the API will handle disks on its own.

Thanks @JorgeB for tracking it down.

Thank you both @JorgeB and @Jandrop . I re-installed the Plugin with the new update as i'd previously uninstalled it, stopped the array and the issue was immediately fixed without a reboot. Thanks!

39 minutes ago, Jandrop said:

Plugin author here, sorry for the trouble. JorgeB's diagnosis is spot on: the companion replaces how the API lists disks (so it doesn't wake spun-down drives), and that code reported sizes as a few bytes instead of the real value, so every disk failed the Internal Boot size check.

This is fixed in U-Manager Companion 2026.10.04.1, released today. Updating the plugin from the Plugins page is enough: the fix applies immediately and no reboot or uninstall is needed. If you already removed it to set up internal boot, you can reinstall it and disk sizes will now be reported correctly.

GitHub issue for reference: https://github.com/jandrop/u-manager-companion/issues/3

Longer term, I'd rather not patch the API for this at all. I've opened a work intent on the Unraid API to fix the spin-up problem upstream (https://github.com/unraid/api/issues/2105, building on https://github.com/unraid/api/issues/2018). If it's accepted and lands, the companion can drop this patch and the API will handle disks on its own.

Thanks @JorgeB for tracking it down.

Aow this was quick, I have asked the same few hours ago and jow see the update, well done