mint-iso-verify crashes on LMDE 7 with LMDE 7 iso

[sandyscott]

OS: LMDE7

To reproduce:

1. Download the LMDE 7 iso lmde-7-cinnamon-64bit.iso
2. Start mint-iso-verify by either:
   • running mint-iso-verify lmde-7-cinnamon-64bit.iso
   • right-click the .iso in the file manager and choose Verify
3. The tool will start, and think for a bit while calculating the check sum, then the Verify button becomes available
4. Click Verify
5. The tool freezes, and must be killed - usually the GUI recognizes this and offers the option to force-quit or wait.

The ISO is valid, verfied via the second option on the relevant help page: https://linuxmint-installation-guide.readthedocs.io/en/latest/verify.html

This is because the URL of the mint mirror is not correct.

To test it open the browser firefox and insert "https://mirrors.kernel.org/linuxmint"

and you will be redirected to "https://mirrors.edge.kernel.org/linuxmint" which is the correct URL.

You can test it by changing the URL's manually (add edge in the URL so that it will be "mirrors.edge.kernel.org") and perform the verification then.

To make the change permanently:

Open the file: /usr/lib/mintstick/verify.py

replace the line: MINT_MIRROR = "https://mirrors.kernel.org/linuxmint"

with: MINT_MIRROR = "https://mirrors.edge.kernel.org/linuxmint"

The same issue exists when you try to verify the latest Linux Mint 22.3 iso's e.g. linuxmint-22.3-cinnamon-64bit-hwe-6.17.iso

The URL shown are still referring to the previous ISO's:

https://mirrors.edge.kernel.org/linuxmint/stable/22.3/sha256sum.txt
https://mirrors.edge.kernel.org/linuxmint/stable/22.3/sha256sum.txt.gpg

The correct URL would have been:

https://mirrors.edge.kernel.org/linuxmint/hwe/sha256sum.txt
https://mirrors.edge.kernel.org/linuxmint/hwe/sha256sum.txt.gpg

This whole setup here is certainly not working how it should have been.

[sandyscott]

Thanks, that fixes the issue.

Not sure why it would be wrong - I didn't change it. Might be related to the fact that this installation was upgraded from LMDE 6, but that's all I can think of.

Incidentally the URL without "edge" is the one in the docs - when you click on it there's a 301 Redirect. It looks like the script isn't following that.

It's probably not the ideal failure mode for an incorrect URL - I'd have expected it to time out or give some other error (e.g. if it manages to download an invalid file), instead of crashing.