The usb start was properly done in my first test too, it's just a copy/paste mistake. The fatload usb command would have failed otherwise.
Octeon ubnt_e300(ram)# usb start
Octeon ubnt_e300(ram)# fatload usb 0:1 0x20000000 openwrt-octeon-ubnt_edgerouter-4-initramfs-kernel.bin
Octeon ubnt_e300(ram)# bootoctlinux 0 numcores=4 endbootargs mem=0
The issue really was in the initramfs image that failed to see usb devices.
Thanks for the tip. No matter what I do I always end here after "bootoctlinux 0 numcores=4 endbootargs mem=0"
[ 0.000000] Linux version 5.4.89 (builder@buildhost) (gcc version 8.4.0 (OpenWrt GCC 8.4.0 r15541-e87c0d934c)) #0 SMP Tue Jan 19 12:01:03 2021
[ 0.000000] Skipping L2 locking due to reduced L2 cache size
[ 0.000000] CVMSEG size: 0 cache lines (0 bytes)
[ 0.000000] printk: bootconsole [early0] enabled
[ 0.000000] CPU0 revision is: 000d9602 (Cavium Octeon III)
[ 0.000000] Checking for the multiply/shift bug... no.
[ 0.000000] Checking for the daddiu bug... no.
[ 0.000000] Wasting 278528 bytes for tracking 4352 unused pages
[ 0.000000] Initrd not found or empty - disabling initrd
[ 0.000000] Using appended Device Tree.
[ 0.000000] software IO TLB: mapped [mem 0x0330e000-0x0334e000] (0MB)
[ 0.000000] Primary instruction cache 78kB, virtually tagged, 39 way, 16 sets, linesize 128 bytes.
[ 0.000000] Primary data cache 32kB, 32-way, 8 sets, linesize 128 bytes.
[ 0.000000] Zone ranges:
[ 0.000000] DMA32 [mem 0x0000000001100000-0x00000000efffffff]
[ 0.000000] Normal empty
[ 0.000000] Movable zone start for each node
[ 0.000000] Early memory node ranges
[ 0.000000] node 0: [mem 0x0000000001100000-0x000000000324bfff]
[ 0.000000] node 0: [mem 0x0000000003300000-0x000000000fafffff]
[ 0.000000] node 0: [mem 0x0000000020000000-0x000000004ebfffff]
[ 0.000000] Zeroed struct page in unavailable ranges: 10932 pages
[ 0.000000] Initmem setup node 0 [mem 0x0000000001100000-0x000000004ebfffff]
[ 0.000000] percpu: Embedded 18 pages/cpu s35872 r8192 d29664 u73728
[ 0.000000] Built 1 zonelists, mobility grouping on. Total pages: 247286
[ 0.000000] Kernel command line: root=/dev/mmcblk0p2 rootfstype=squashfs,ext4 rootwait console=ttyS0,115200
[ 0.000000] Dentry cache hash table entries: 131072 (order: 8, 1048576 bytes, linear)
[ 0.000000] Inode-cache hash table entries: 65536 (order: 7, 524288 bytes, linear)
With a blinking cursor. I can't type or reconnect. Any idea?
Kind regards,
carstino
I flashed my original firmware v.2 & u-boot 3.0 to OpenWRT 21.02 snapshot today. I first tried the original method but it failed so I then tried your method with positive result.
Used console cable connection.
Actually it works so smooth I would have no problem seeing this method as formal install method if the other one doesn’t work.
The only thing I changed was to shorten the filenames a lot because PuTTY didn’t manage well with the longest lines of text.
I used the instructions above, had to copy image on EdgeOS to eMMC (uboot is version 3 on my router). After installing snapshot version (USB+terminal cable), I "upgraded" it to May 30 rc2 version using Sysupgrade. I'm more general purpose user than developer and more of GUI than CLI guy, but without these instructions it would have been impossible to get it up and running.
For me, reasons for choosing OpenWRT and EdgeRouter was 4G/5G wireless internet connection. I do have ZTE MC801A modem which has powerful hardware, but quite sad firmware. My ISP has CG-NAT on 4G, public IPv4 costs extra, public IPv6 is free, /64 prefix without PD. So I need something that can handle this IPv6 and is well documented, and I'll use ZTE as bridge. I can do basic things with Cisco IOS, but IPv6/64 without PD seems to be such new thing, that there aren't too many examples on Cisco side, therefore using OpenWRT. For hardware side, EdgeRouter 4 seems very powerful (I do have C897 and I'm used to snappy hardware in metal enclosure!) and is reasonably priced.
Thanks for damex and other people contributing for ER4-version!
soon after device was implemented - there happened to be a bug (not directly related to edgerouter 4) regarding not having a usb storage available for initramfs. it should be fixed by now and if it is still not - please report back so it could be checked even further.