Hi,
I am currently testing a system based on yocto dunfell + swupdate + 
EFIBootguard and I am concerned by the way the kernel is stored in the same 
partition of the configuration.
I am using 2 config partitions for EFIBootguard, and also a double copy 
strategy for rootfs. At start, partition 0 contains the kernel associated 
with the installed rootfs.
If I use swupdate to update some parts of the system (not related to rootfs 
or kernel), this creates a new configuration on partition 1, but the kernel 
file position is still on partition 0.
Should then I need to update both the kernel and the rootfs, this would 
switch back to having a new active configuration on partition 0, but the 
new kernel should be written on configuration partition 1 to preserve the 
currently running kernel (resident on partition 0) as a fail safe in case 
of a rollback.
This brings to a situation where cfg in use is in a partition,  but the 
kernel is on another one.
It also makes updated more complex: due to the double copy strategy, not 
only I have to detect the inactive rootfs partition, but also the inactive 
configuration partition where to store the kernel. I am really uneasy about 
this.
Sadly EFIBootguard cannot boot the kernel stored on ext4 /boot rootfs 
directlry, so I am 
looking for an alterlative strategy. At the moment this involves 2 
partitions dedicated to storing the kernel file, without saving it in the 
configuration partitons.
What do you think?

Thanks!

-- 
You received this message because you are subscribed to the Google Groups "EFI 
Boot Guard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/efibootguard-dev/cba82c2b-3d41-4c45-bad8-4e7ab7729de9n%40googlegroups.com.

Reply via email to