Got everything setup. 4 sids connected to 4 different VLANs. All seems well but one thing I noticed now is that on the overview screen I get one of two errors (no status is shown)
Maximum call stack size exceeded
or
too much recursion
Any clue whats causing this??
luci error please create an issue there for bisect.
Does the install process listed in the wiki actually work?
Amazon dropped one of these units off to me yesterday, and today I've been pulling my hair out trying to get openwrt running on it. I've managed to get SSH running on it by restoring the pre-made config file, but I've tried a hundred times to complete the rest of the steps of Part 1 Option A, and it always boots into factory firmware (IP address 192.168.216.1).
I have a 32 GB USB disk, formatted as fat32, copied on the initrd (various ones including 23.05.0, 23.05.2, and snapshot), and ran the two commands to start it (verified that the filename is correct), yet it always just reboots into factory. The disk mounts just fine in the factory firmware upon insertion. I've even gone so far as to find something running windows (as painful as that is to deal with) to reformat the disk with.
Any hints?
Can anybody offer a disk image I can put onto my USB disk with 'dd' to eliminate that part of the equation?
Is it possible that there is something changed in the shipping firmware to break the process?
Factory firmware shows;
Product ID: DL-WRX36
Hardware Version: REV_MP0_01_MA0
Firmware Version installed: 1.10.01.245
admin@DL-WRX36:~# ls /mnt/UNTITLED/
openwrt-23.05.2-ipq807x-generic-dynalink_dl-wrx36-initramfs-uImage.itb
openwrt-ipq807x-generic-dynalink_dl-wrx36-initramfs-uImage.itb
openwrt-qualcommax-ipq807x-dynalink_dl-wrx36-initramfs-uImage.itb
admin@DL-WRX36:~# fw_printenv
bootdelay=2
baudrate=115200
bootcmd=usb start && fatload usb 0:1 0x44000000 openwrt-qualcommax-ipq807x-dynalink_dl-wrx36-initramfs-uImage.itb && bootm 0x44000000; bootipq
admin@DL-WRX36:~#
That bootcmd variable seems to persist through multiple reboots.
My guess is that
usb 0:1 (or something similar). In style ofAppreciate the reply.
Per another post way above, it seems that 32GB should be fine;
It is a single partition. I've tried both with msdos and gpt partition tables. No difference.
Could you clear me up on the "usb 0:1" thing? I've got;
[ 383.624410] usb 2-1: new SuperSpeed USB device number 2 using xhci-hcd
[ 383.665007] usb-storage 2-1:1.0: USB Mass Storage device detected
[ 383.665285] scsi host0: usb-storage 2-1:1.0
[ 384.743259] scsi 0:0:0:0: Direct-Access Kingston DataTraveler 3.0 1.00 P Q: 0 ANSI: 6
[ 384.745193] sd 0:0:0:0: [sda] 60437492 512-byte logical blocks: (30.9 GB/28. 8 GiB)
[ 384.751308] sd 0:0:0:0: [sda] Write Protect is off
[ 384.758391] sd 0:0:0:0: [sda] Mode Sense: 4f 00 00 00
[ 384.758970] sd 0:0:0:0: [sda] Write cache: disabled, read cache: enabled, do esn't support DPO or FUA
[ 384.767388] sda: sda1
[ 384.774564] sd 0:0:0:0: [sda] Attached SCSI removable disk
I've also successfully tried a system downgrade to version 1.10.01.201 by editing /etc/system_version.info to 1.10.01.200 and then installing the .201 version linked in this thread. No change.
Well, I looked at my router, and under OpenWrt it classifies the USB stick as usb-storage 1-1:1.0
My router has the USB failsafe bootcmd that I designed (and which is show also in wiki), but the USB part is the same as in the wiki install section.
root@router5:~# ls /mnt/sda1/op*
/mnt/sda1/openwrt-ipq807x-generic-dynalink_dl-wrx36-initramfs-uImage.itb
root@router5:~# fw_printenv |grep boot
bootcmd=run openwrtusb; run openwrtboot
bootdelay=2
openwrtboot=setenv bootargs console=ttyMSM0,115200n8 ubi.mtd=rootfs rootfstype=squashfs rootwait; ubi part fs; ubi read 0x44000000 kernel; bootm 0x44000000#config@rt5010w-d350-rev0
openwrtusb=usb start && fatload usb 0:1 0x44000000 openwrt-ipq807x-generic-dynalink_dl-wrx36-initramfs-uImage.itb && bootm 0x44000000
root@router5:~# dmesg | tail -n 13
[90565.484175] usb 1-1: new high-speed USB device number 2 using xhci-hcd
[90565.692204] usb-storage 1-1:1.0: USB Mass Storage device detected
[90565.692516] scsi host0: usb-storage 1-1:1.0
[90566.704972] scsi 0:0:0:0: Direct-Access Pretec UltimateGuard 2.00 PQ: 0 ANSI: 2
[90566.706297] sd 0:0:0:0: [sda] 7897088 512-byte logical blocks: (4.04 GB/3.77 GiB)
[90566.712479] sd 0:0:0:0: [sda] Write Protect is off
[90566.719548] sd 0:0:0:0: [sda] Mode Sense: 00 00 00 00
[90566.724623] sd 0:0:0:0: [sda] Asking for cache data failed
[90566.729321] sd 0:0:0:0: [sda] Assuming drive cache: write through
[90566.904949] sda: sda1
[90566.905274] sd 0:0:0:0: [sda] Attached SCSI removable disk
[90566.919314] mtdblock: MTD device 'rootfs' is NAND, please consider using UBI block devices instead.
[90567.112782] mtdblock: MTD device 'rootfs' is NAND, please consider using UBI block devices instead.
I tested with my stick , and it nicely rebooted to the old OpenWrt initramfs version on the stick... OpenWrt SNAPSHOT, r21917-cfb296b79a
I doubt that the firmware upgrade/downgrade would modify the u-boot bootloader. Likely unrelated.
Any chance you could share an image of your usb disk?
Try to partition the USB as MBR (msdos) and format it as FAT32 using Rufus https://rufus.ie/en/
That program doesn't look like it has the option to just format based on the website.
It does both partitioning (MBR or GPT or Super Floppy Disk) and formatting (FAT32, exFAT, NTFS).
I've tried that. No change, still not working.
I used Windows DiskPart to get me up and running, another user recently had the same issue and they used it and it worked for them.
In my case, I update the factory firmware to latest version, then insert the exFAT usb stick accidentally to start the installation as wiki shown, eventually all thing work right.
I gave up on that SSH/USB approach. Wasted too many hours on it. Decided to take Option B w/TFTP instead and worked first try.
I had a similar problem with my USB data disks. The USB sticks that I had created with Windows did not work for me. Only when I created one with FAT16 (max. 4 GB paritition) with mini tool paritition wizzard (free version) did it work.
That is what I did also.
Later I tried it with FAT32 (also using Minitool Partition wizzard) and that also worked for me.
I can imagine if you have a really slow/old USB stick it will time out.
Ill add that I was able to get it to boot off of a USB drive that I formatted to FAT32 with the KDE GUI partitioning tool (on Fedora 39 (KDE spin)) and then copied the file over to.
Ill also say that I tried a couple other ways first and those didnt work.
I wonder if the USB stick has something to do with it....I know that (on windows at least) some of my USB drives when you plug them in dont show up as "removable storage" but just as an attached hard drive. When this happens right clicking on the drive in explorer the right clock menu doesnt give you an "eject" option.
Im not sure what exactly, but something on the USB flash drive is making windows recognize it differently...Perhaps something on the USB flash drive is making u-boot label the drive differently too.
Unless somebody can figure out EXACTLY what is needed to make the USB install approach actually work reliably, the wiki should have a big ugly warning with it saying that it is NOT A RELIABLE METHOD, and that users should be prepared to use the UART/TFTP method instead.
Noticed that too.
I have an expensive high capacity USB3 stick which shows up as scsi attached storage
However I did not have any problem with the USB method but did use a cheaper 32 GB stick