Nv update engine runs failed on JP5.0.2


I try to run nv update engine on ramdisk based on JP5.0.2.
it reports errrors as follows:
/usr/sbin/nv_update_engine -i bl --payload /ota_work/bl_update_payload --no-reboot
Nvidia A/B-Redundancy Update tool Version 2.0
Got payload file: /ota_work/bl_update_payload
Error: read variable BootChainFwCurrent failed!
Error: installing bootloader updates failed: -5

Could you please help to check what’s wrong ? Thanks.

hello weileng,

this is one of UEFI variables.

may I know all your steps for creating Bootloader update payload (BUP).

Hi Jerry,

it was generted by scripts “l4t_generate_ota_package.sh”. Thanks.

hello weileng,

may I know what’s your use-case,
please see-also Update Engine session for the footnotes,

The current Update Engine only supports updates to slots A and B at the same time.

Hi Jerry,

Thanks for your update. It seems JP5.02 doesn’t support AB redandunce update, but all slots update the same time.

However, for JP5.0.2 In normal rootfs(ubuntu brings up with UI), I can run cmds as follows:
#nvbootctrl get-current-slot
#nvbootctrl dump-slots-info
Current bootloader slot: A
Active bootloader slot: A
num_slots: 2
slot: 0, status: normal
slot: 1, status: normal

If I change ramdisk init scripts run into ‘/bin/bash’, simiar issue was found as follows:
-bash-5.0# nvbootctrl get-current-slot
Error: read variable BootChainFwCurrent failed!
Invalid current slot found: -22(normally this means all slots are corrupted).
-bash-5.0# nvbootctrl dump-slots-info
Error: read variable BootChainFwCurrent failed!
Error: invalid current bootloader slot: return(-22)

could you please help to check what’s UEFI setting should be do to fix such issue? Thanks.

I think the actual point here is not whether there is a UEFI setting or not. But the scenario you are running.

Did you also use ramdisk to run nvbootctrl when you were in jetpack4?
If you don’t use ramdisk to run this, would you hit issue?

Hi Wayne,

Yes, the previous code I did some change on jetpack4 to implement some update on ramdisk, which run nvbootctrl smoothly. However, after I switch to jetpack5/R35.1, such issue was found.

By default we don’t guarantee ramdisk could run this. Are you able to reproduce the issue in file system?

Hi Wayne

After some experiments, it seems due to efivar fs hasn’t been mount.
after mount efivarfs in ramdisk, it can be workable. Thanks
#mount -t efivarfs efivarfs /sys/firmware/efi/efivars

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.