Lennart Poettering
Lennart Poettering
This one confuses me. the commit msgs suggest this had an effect on regular initrd→host transitions, but i don't see how that could be, because regular initrds should not be...
lgtm. i'd even merge it as is, but the rework @yuwata/@YHNdnzj asked for sounds good to.
This is actually the common case in environments where the initrd is generically built: it's not parameterized and hence comes up with a different machine ID each time. This is...
hmm, this is a really weird usecase. I am not really sure we should hide this kind of error, it seems really really specific to me, in particular as it's...
> assembles the key via a combination of TPM2 + FIDO2 devices btw, this is somethin i eventually want to support natively in systemd, i.e. that we validate a challenge...
> 🔐 Please enter passphrase for disk Linux LUKS (test): •••••••• > Set cipher aes, mode xts-plain64, key size 512 bits for device /dev/vda3. > Device test already exists. >...
seems ok, just some nitpicks
Hmm, the intention was actually that programs that drop in .user/.group files would also drop in .membership files. But I guess we never properly documented that. i.e. for efficiency reasons...
Well, the stuff is in /run/ anyway, it's not really a slow disk, but a fast tmpfs, given how slow the rest of the passwd database access is it seems...