I have seen so many nodes being bricked trying to update from v0.4.xx to v0.5.0 that many fellow noderunners (including myself) are afraid to update.
Here‘s my proposal:
Why don‘t you write a script that checks all the things that are necessary for a successful update, like DNS/firewall settings, enough disk space on SSD and SD card, fixed IPv4 address, … and saves it into a log file?
That would give us the chance to fix all issues that might occur before (!) the actual update.
But if you decide to just releasing future versions like v0.5.2 you might eventually loose the early adopters still running on v0.4.xx to whom you owe your success…
Thank you @mr.cook! It’s a good idea, and as much as we’d have loved to support it, the main issue is that the problem cannot be fixed in a script. We update Docker in the 0.5 update process, and it turns out that the process of updating Docker is a bit messy and unpredictable (causing these issues). Thankfully, the fix is as simple as reflashing Umbrel OS 0.5 on the microSD card if your Umbrel becomes unresponsive during/after the update, but unfortunately we cannot fix it in a script and release it:(
Thank you for your answer. If the Docker update is the main problem and the solution is simply reflashing the SD card then I will try to update after my vacation.
