> Evidently, that extra bit of RAM is enough to run dracut properly. Okay, this pretty much confirms for me that this is a "limited memory issue". However, there's also obviously a fairly serious bug in dracut in that it's capable of truncating an initramfs, and metaphorical shrugging its shoulders and carrying on like nothing happened.
I'm going to re-purpose the kernel bug I originally opened (LP: #2157789) to deal with that, as that's one specific thing that can be dealt with cleanly. I'll keep this bug open to cover the more general "piboot-try has fallen back and I don't know why" case as I consolidated a few bugs into this one that probably aren't memory related (some of the bugs related to a Pi4 and I'd be rather surprised if those were memory constrained). I've already opened LP: #2197975 to deal with the tmp.mount issue; that's fixed in stonking and hopefully I can SRU that to resolute shortly. Anyway, thank you very much to both irihapeti and efruendt for all the debugging information -- I think we're getting somewhere with this now! -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/2155094 Title: RPi 3B, Zero 2W, 4B: New boot assets in /boot/firmware/new failed To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+bug/2155094/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
