I have not tried firstboot
yet (because I don't have easy ethernet access at the moment), but it is something I would try before repeating the ramboot sequence.
Have you tried the "dry-run" mode with sysupgrade "--test" option?
It might give some clues.
But maybe not, as your console log seems normal, like @borromini already said.
I have ran into similar things a few times in the past dozen years, and usually the debugging requires serial console to see what happens...
...and possibly also debug "echo" statements inserted into /sbin/sysupgrade
, /lib/upgrade/do_stage2
and /lib/upgrade/stage2
depending on your platform's upgrade characteristics, so that you see exactly where the flash fails.
Yes, that was the first thing I tried. Unfortunately, it did not any output at all.
sysupgrade -n
did not work, neither did firstboot
initially - from the CLI it said it could only delete the files as the partition was still mounted. However, I then chose "reset" from LuCI and after that was finally able to flash 21.02.1. (Recovery of the mesh network configuration was a bit of a chore as I needed to install some kmod
packages the get it running, and of course that was not possible without working network connection. But it's all sorted now.)
So it looks like being able to flash the Fritz!Repeater 3000 once does not mean much - I've created a custom build (basically same as stock with a few more built-in packages), but again flashing fails.
Any ideas if there's anything I can try to fix this once and for all (e.g. flashing the same image multiple times after the next reset)? Having to completely reset the device each time before flashing a new firmware is annoying.
This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.