David Cassany Viladomat

> having a mutable one that can be also changed in runtime (for example, debugging, but also situations where recovery is not meant to be ro) Correct, I did not...

> at the same time changing _slightly_ how we would load a squashfs if found with the new layout proposed above in the initrd/grub components. This is what I would...

In addition, I also realized that we could have the _mutable_ recovery system by setting a proper persistent overlayfs, so we could eventually support mutable recovery with persistency and a...

Indeed a big chunk of the changes proposed here are done. Just the final step of my suggestion would be missing: > My suggestion is to use the common Live...

Would be nice to have some sort of config file to name all these devices (COS_LIVE, COS_OEM, COS_PERSISTENT, etc.) the problem I see is that we need to access these...

Well, this is related to #160 indeed. It is possible to configure it if you rebuild the `live/grub` packages with the modified label. Bootloader config needs to know the label,...

Yes, using the docker image sounds like a reasonable effective and simple solution. I have also been wondering if we aren't abusing of the `manifest.yaml` file in elemental-toolkit. I mean,...

> Couple of things, first, I don't think we are using that package list anymore as we use the build.go facility to identify packages missing and build those so it...

I would not only consider the dracut emergency shell. I would consider any emergency stage of systemd, which includes the dracut emergency shell, but not only. After switching root and...

Just for the records on #1074 I am also seeing non consistent failures in installer tests due to connection timeouts. Specially it happens quite often for EFI setups on reboot...