[Bug] After a fresh install of TH `v2.11-beta1+fall2024`, backend has an install failure and fails

[tcarmelveilleux]

Describe the bug

After running the install steps from Matter TH manual, TH backend has a start failure and TH frontend shows internal errors.

Steps to reproduce the behavior

• Fresh-installed a raspi4 on 22.04LTS server
• Cloned TH repos per instructions at tag v2.11-beta1+fall2024
• Ran ./scripts/pi-setup/auto-install.sh to install everything

First time device failed to download some pieces and rebooted:

Then I re-ran ./scripts/pi-setup/auto-install.sh on next boot, and it succeeded to do everything right, and I selected "1)" to reboot.

After reboot, the TH is still not working

Expected behavior

• Running ./scripts/pi-setup/auto-install.sh on the correct check-out against a fresh Ubuntu 22.04LTS install results in a working TH.

Log files

• th_doctor_logs_tennessee.txt
• th_frontend_logs.txt
• th_backend_logs.txt

PICS file

No response

Screenshots

No response

Environment

• Firefox
• TH v2.11-beta1+fall2024
• Raspi 4 4GB, Ubuntu 22.04LTS Server

Additional Information

No response

[tcarmelveilleux]

This got resolved after I did:

git clone -b v2.11-beta1+fall2024  https://github.com/project-chip/certification-tool.git certification-tool-2
cd certification-tool-2
git submodule init
git submodule update --recursive
./scripts/pi-setup/auto-install.sh

....

However, the initial install should have succeeded. It appears maybe something got partially downloaded that left the whole system in an indeterminate state.

There should be a "clean sweep" script that can be run to delete all the docker containers and images, and allow to restart fresh like whatever the side-effects of cloning a separate version of the repos did.

[hiltonlima]

Hello @tcarmelveilleux, Thanks for the feedback.

We agree with the script's suggestion. We will make it available as soon as possible.