The other day I thought I might try a complete reflash of crimson, so after backing up some items I wanted to keep and/or reapply, I attempted it, but the process errored out immediately every time:
Searching...
uuu (Universal Update Utility) for nxp imx chips -- libpureos/1.2.91+0git6b465-0pureos+librem5.2-1-g5e5fee8
Wait for Known USB Device Appear...
(Here, I hold volume-up, connect cable to Librem 5, and reinsert battery, then release volume-up.)
Error: Unknown Command:FB: reboot
Traceback (most recent call last):
File "/home/xxxxxx/librem5-flash-image/./scripts/librem5-flash-image", line 630, in <module>
sys.exit(main())
^^^^^^
File "/home/xxxxxx/librem5-flash-image/./scripts/librem5-flash-image", line 613, in main
flash_image(uuu_target, args.debug)
File "/home/xxxxxx/librem5-flash-image/./scripts/librem5-flash-image", line 403, in flash_image
subprocess.check_call(['uuu', uuu_target])
File "/usr/lib/python3.12/subprocess.py", line 413, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['uuu', './Images/flash_librem5r4.lst']' returned non-zero exit status 255.
Anybody have any idea what might be wrong? I have the latest version of uuu and all the scripts were freshly dowloaded.
I just did a successful reflash from a different computer which I hadn’t used for flashing before, with newly downloaded Librem 5 scripts, and the USB-C-to-USB-C Librem 5 cable.
It all worked without error.
Before, I must have experienced some silly, hitherto unseen, and inexplicable “operator” error…
Interesting that when I was preparing to flash, I even followed the instructions to install uuu even though I knew it was already installed, and apt confirmed that I already had the “latest version.”
And also output from which uuu (as @‍dos already requested) - because if your setup is really really old, you might have built uuu from sources (or scrounged a binary from somewhere) and then left it in a directory in PATH that is masking the officially installed version.
Lol. No. I had already flashed crimson a couple of times. (As well as upgrading with the tool at least once.)
I might have actually installed that originally from the Librem Mint repo, or I might have built it from your guide in the old Tips & Tricks. I don’t really remember anymore.
To be clear though, those two questions are about the host computer, not the Librem 5.
Edit: Yes, I can see the version that I myself built from sources, no longer in the PATH, but still squirreled away, and it identifies as “libpureos/1.2.91+0git6b465-0pureos+librem5.2-1-g5e5fee8”. So I would say that which uuu will point at /usr/local/bin/uuu (and not/usr/bin/uuu), which will in turn explain where that host computer is falling down. If that is correct then
Yes, you can just delete it because, after all, if that is a mistake, you can reinstall that package. However I like to be cautious, particularly when “messing with” someone else’s computer. So probably what I would recommend is: mv followed by testing that uuunow works on that computer followed by rm.
After my successful reflash, I applied updates, then immediately imaged the eMMC. I’ll be able to boot that from the microSD card to restore the OS by connecting to my laptop (if I need a more-or-less default install).
In preparation for duplicating my former crimson setup, I had collected files, app lists, scripts, my DIY apps, and configurations from the old install, and copied them into a dedicated folder on my microSD card.
I’ve now placed those into the new system in the appropriate places, restored all the services and permissions, installed the extra applications I had before (- the ones I still wanted, that is -), adjusted settings, and applied new updates.
Everything’s nice and fresh now, so I imaged this crimson install with all my personalizations included, too.
Now I’m compressing both .img files and backing them up on my network.
I used this as a general checklist, and added in some additional items I wanted to restore.
