Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Upgrade scripts failed and bricked a system #20

Open
Jookia opened this issue May 12, 2020 · 4 comments
Open

Upgrade scripts failed and bricked a system #20

Jookia opened this issue May 12, 2020 · 4 comments

Comments

@Jookia
Copy link
Collaborator

Jookia commented May 12, 2020

technomancy in IRC ran the upgrade scripts on an existing buster install.
The senoko script failed to run since i2c-get wasn't installed.
The kernel upgrade script removed the old zimage and added a new uEnv.txt before confirming the actual buster install worked.
It also could be that u-boot-tools wasn't installed so zImage.buster couldn't be made.
I've updated the scripts with a hack to install these unconditionally for now.

What actually needs to be done:

  • [] Make sure the senoko package depends on i2c-tools
  • [] Make sure flash-kernel depends on u-boot-tools or we install it
  • [] Put back old kernel and remove uEnv.txt on upgrade failure
@anon8675309
Copy link

I believe I have also run into this problem (or something very similar) after running ./upgrade_5_BETA_KERNEL.sh and rebooting. The HDMI output shows a picture of Tux and the text: U-Boot 2014.10-rc3-00039-gc5efead (Oct 17 2014 - 19:41:27).

This is from a fresh flash of an SD card with the original Novena image, and then running each of the upgrade scripts in order.

The version of upgrade_4_apt.sh that I used did install i2c-tools and u-boot-tools and I pulled the SD card and looked at the /boot partition and saw that it did create the zImage.buster file.

I do not have a zImage file, but it looks like eEnv.txt is instructing something (uBoot?) to boot to the zImage.buster kernel. As a test, I moved eEnv.txt out of the way and renamed zImage.buster to zImage. This resulted in no output via HDMI and nothing output on the serial console.

Next I tried moving the kernel and dtb files out of the way and putting the recovery images in their place and I was able to see the uboot messages, it started the kernel, and I was able to SSH in.

I can keep this SD card around for a couple days in case it'd be helpful at troubleshooting what went wrong. Just let me know what to do and I can report back the results.

@Jookia
Copy link
Collaborator Author

Jookia commented Nov 9, 2022

Honestly I can't see myself fixing this but instead making a new SD card image for people. The effort for testing upgrades like this is extremely high :\

Edit: I'm very happy for other people to do this but when I last did this it was a nightmare of having to load old images to SD and run the scripts manually.

Could you paste the uEnv.txt or ... eEnv.txt? Maybe I can see something obvious. Or yeah, upload the image.

@anon8675309
Copy link

uEnv.txt
zImage.buster.zip

Testing upgrades is indeed very time consuming. Several hours for a full test. I'm willing to do the testing though. I figure we're going to have to figure out these issues one way or another and end-to-end testing seems like it'll get the most reliable results.

I'd also be happy to help with making a new "factory" image, and/or do testing for that. If you are going to make a new image, please watch out for #34. I am doing further testing right now to try to find a solution that result in me being able to use X11 on at least Buster, if not Bullseye.

@Jookia
Copy link
Collaborator Author

Jookia commented Nov 10, 2022 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants