Hi everyone,

 

I'm facing an issue with Armbian installation on Rock 5B Plus.

 

Hardware

• Board: Radxa ROCK 5B+ (RK3588)
• Storage:
  NVMe #1 (system): 256 GB, in M.2 slot 1
  NVMe #2 (backup): 128 GB, in M.2 slot 2
• Boot: MTD flash + NVMe
• Power supply: both POE+ HAT 25W and official power supply 30W

Software

• Armbian: 25.11.2 Desktop ubuntu
• Kernel:  6.1.115-vendor-rk35xx
• Boot configured via  armbian-install  → “Boot from MTD Flash, system on NVMe” 

Initial install steps

1. Wrote Armbian image to microSD and booted from SD.
2. Used  armbian-install  to install: Boot from MTD Flash, system on NVMe #1
3. Powered off, removed SD, left only NVMe #1 installed.
4. Board boots fine from SPI+NVMe #1: blue LED, OS comes up, Green LED, everything OK.
5. Power off the board and plug NVME #2 (preformatted ext4 with gparted and empty, the disk works because I tried the procedure the other way around, using the second disk as the main one)
6. Power on. Solid blue light, no OS, no HDMI output (I don't have UART)

lsblk -f with microSD plugged:

mtdblock0

mmcblk1
└─mmcblk1p1 ext4   1.0   armbi_root   2e67f233-22a9-46c9-90a7-8f0e0f2d3154   21.7G    22% /var/log.hdd
                                                                                         /

zram0                                                                     [SWAP]
zram1                                                          36M    16% /var/log
zram2

nvme1n1
├─nvme1n1p1 ext4   1.0                cacf66ca-ea9b-4dc4-b7b3-185f18e7c64f    100G     6% /mnt/main
└─nvme1n1p2 ext4   1.0   data         a7d6675c-3ca6-4a87-85cf-7d91856fd79c

nvme0n1
└─nvme0n1p1 ext4   1.0   armbi_backup 5aadc790-715a-4f48-bbbc-ce4bfd2717bd  110.8G     0% /mnt/backup

/boot/armbianEnv.txt looks correct

verbosity=1
bootlogo=true
console=both
extraargs=cma=256M
overlay_prefix=rockchip-rk3588
overlays=panthor-gpu
fdtfile=rockchip/rk3588-rock-5b-plus.dtb
rootdev=UUID=cacf66ca-ea9b-4dc4-b7b3-185f18e7c64f
rootfstype=ext4
usbstoragequirks=0x2537:0x1066:u,0x2537:0x1068:u

/etc/fstab looks correct

# <file system>					<mount point>	<type>	<options>							<dump>	<pass>
tmpfs						/tmp		tmpfs	defaults,nosuid							0	0
UUID=cacf66ca-ea9b-4dc4-b7b3-185f18e7c64f	/		ext4	defaults,noatime,commit=120,errors=remount-ro,x-gvfs-hide	0	1
UUID=5aadc790-715a-4f48-bbbc-ce4bfd2717bd	/mnt/backup	ext4	defaults,nofail  0  2 <--- also tried without this line

I've also tried by "forcing" the mount of NMVE #2 by adding it to /etc/fstab of NVME #1 but nothing changes.

 

Could you please help me understand if this is an issue related to my configuration or could be a bug?

Thanks

Without uart hard to tell.

Hi @Werner,

I bought an UART USB device and followed your video here https://www.youtube.com/watch?v=UpVMO7gbnYM to read boot logs.

I'm not a huge expert but looks like there is a problem finding /boot when two nvme are connected.

 

Please find attached the following logs:

• uart-bootcorrectnvme.log -> one nvme plugged (the one with os installed in slot 1) - works!
• uart-bootdualnvme.log -> two nvme plugged (the one with os installed in slot 1) - fails! 
• uart-bootmicrosd.log -> two nvme plugged + microsd plugged - works!

 

Tests have been executed with official 30W Radxa Power adapter ( same result with POE+ 25W hat)

 

Please, let me know if there is anything else I can provide.

Thanks

uart-bootcorrectnvme.log uart-bootdualnvme.log uart-bootmicrosd.log

Hi @Werner,

do you have any news on this bug? I had to copy the /boot on the second nvme to make this setup work and to be honest it shouldn't be like that

None I am aware of. neither do I have this board nor do I maintain it. 

I maintain it and can take a look this week. Just need to find a spare nvme somewhere

meco any update on this? dual NVME also does not work on non vendor kernels as well as second HDMI port.