TFTP not communicating with my Meraki MX60

I'm not sure who placed that information on the page...I'm referencing an older document: https://github.com/riptidewave93/LEDE-MX60-MX60W/blob/master/README.md

The command you have is nothing like the orginal states - or it was edited:

setenv serverpath; setenv netloadmethod tftpboot; setenv bootargs console=ttyS0,${baudrate} rootfstype=squashfs mtdoops.mtddev=oops; run meraki_load_net meraki_checkpart meraki_bootlinux

Also be advised, it seems your directions require the computer to have the IP 192.168.1.2.

I'm using the version of commands from here: https://openwrt.org/toh/meraki/mx60

I fixed my settings in Tftpd64 and got the following which hits an exception and restarts.

Using ppc_4xx_eth0 device
TFTP from server 192.168.1.101; our IP address is 192.168.1.1
Filename 'buck.bin'.
Load address: 0x192
Loading: #Bad trap at PC: 508, SR: 21000, vector=1300
NIP: 00000508 XER: 20000000 LR: 1FFD8C1C REGS: 1fb57aa0 TRAP: 1300 DEAR: 30B2ABFE
MSR: 00021000 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR: 00

GPR00: 00000003 1FB57B90 1FB57F40 00000000 00000002 0000001F 1FB5B60D 01650CCB
GPR08: EDFB000C 1FB594D0 00005300 FFFFD300 48244728 07FFE300 1FFF1200 30B2A000
GPR16: 1FB57F80 00000000 1FFEE338 1FFEE340 1FFDD522 1FFEE30C 1FFF13D4 FFFF8000
GPR24: FFFFFFFF 00000000 000005EE 00029000 00000008 00000010 1FFF3048 1FB59458
Call backtrace:
1FFD8C1C 1FF9A310 1FF9C2C8 1FFB1DC0 1FFBAA3C 1FFB9F5C 1FFBA270
1FFBA97C 1FFB9F5C 1FFBA168 1FFBD1A4 1FF99BA0 1FF7A710
Exception
▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒
▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒▒        @     ▒H                ▒H                  H                  H
      ▒H             ▒▒   @ @
                       @44@$ @ 4@ 4@4$ 4@h@$ 4@44$ 4 4@4 @   @4@@
                                                      @   @
                                   @    @   @ @
                               @$4@4 4@ H$ @4
 4@4▒

U-Boot 2009.08-00049-g0ee25dd Meraki MX60 (Jul 27 2012 - 14:59:20)

To be clear, you're using this file to boot?

http://downloads.openwrt.org/releases/18.06.4/targets/apm821xx/nand/openwrt-18.06.4-apm821xx-nand-meraki_mx60-initramfs-kernel.bin

To clarify, do not use the TAR file to do first boot or recovery, use the BIN file.

Correct. I'm using that exact file. I actually went and redownloaded it and tried again, and this time Tftpd64 is now connecting, then saying that there was an error and terminating the connection.

This is what happens when I try to run it. It'll start and error out and quit trying.
RujF1zu

And now I'm getting the following error:

=> saveenv
Saving Environment to NAND...
Erasing redundant Nand...
Erasing at 0x1e0000 -- 100% complete.
Writing to redundant Nand... done
=> setenv ipaddr 192.168.1.1
=> setenv bootargs console=ttyS0,$baudrate mtdoops.mtddev=oops;
=> tftpboot 192.168.1.2:$meraki_loadaddr buck.bin;
Using ppc_4xx_eth0 device
TFTP from server 192.168.1.101; our IP address is 192.168.1.1
Filename 'buck.bin'.
Load address: 0x192
Loading: #NIP: 00000508 XER: 20000000 LR: 1FFD8C1C REGS: 1fb57aa0 TRAP: 0700 DEAR: 00100000
MSR: 00021000 EE: 0 PR: 0 FP: 0 ME: 1 IR/DR: 00

GPR00: 00000004 1FB57B90 1FB57F40 00000000 00000002 0000001F 1FB5B60D 01650B27
GPR08: EEB1000C 1FB594D4 00005300 FFFFD300 48D44728 08000010 1FFF1200 00000000
GPR16: 1FB57F80 00000000 1FFEE338 1FFEE340 1FFDD522 1FFEE30C 1FFF13D4 FFFF8000
GPR24: FFFFFFFF 00000000 00000232 00029000 0000000C 00000018 1FFF3048 1FB59458
** Illegal Instruction **
Call backtrace:
1FFD8C1C 1FF9A310 1FF9C2C8 1FFB1DC0 1FFBAA3C 1FFB9F5C 1FFBA270
1FFBA97C 1FFB9F5C 1FFBA168 1FFBD1A4 1FF99BA0 1FF7A710
Program Check Exception
▒▒▒▒▒▒▒▒▒▒▒
@ ▒ @▒▒▒▒▒▒▒ ▒                              @                 H  H                         H  @
H      H H           @                 H                   H          @  @         H        H
U-Boot 2009.08-00049-g0ee25dd Meraki MX60 (Jul 27 2012 - 14:59:20)

CPU:   AMCC PowerPC  UNKNOWN (PVR=12c41c83) at 800 MHz (PLB=200, OPB=100, EBC=100 MHz)
       Bootstrap Option D - Boot ROM Location NAND wo/ECC 2k page (8 bits), booting from NAND
       32 kB I-Cache 32 kB D-Cache
Board: Buckminster Wireless - Meraki Buckminster Wireless Cloud Managed Router
============================

I'm not sure why you keep ruining the saveenv command. It only needs to be ran once. I'm hoping you didn't accidentally erase anything. Looking at my orginal environment before making changes, I have this:

meraki_loadaddr=800000

You have:

I went and picked up a second MX60W from our pile at work and it's doing the exact same thing.

That looks like the wrong load address. It could be crashing the bootloader by overwriting RAM that the bootloader uses.

Is there any fix for this?

I got mine to work a while ago, now that I have a second one, I am running into this issue. For the last one, I think I threw a "hail mary" and changed the load_addr, but I can't remember what I did. Does anyone know how to change the load address from 0x192 to something correct?

If that's the only thing that got broken, you can do

setenv meraki_loadaddr 800000

and then

printenv

to check/confirm/verify the setting (ideally you also run printenv before in that case). I don't have any MX60(W) so I don't know what a healthy u-boot env looks like on those. I think @lleachii has posted a few on the forums though.

printing the meraki_loadaddr yields 800000, so it didn't get messed up. There's something wrong with the commands in the toh.

I did these commands from another thread (linked below).

`setenv serverpath; setenv netloadmethod tftpboot; setenv bootargs console=ttyS0,${baudrate} rootfstype=squashfs mtdoops.mtddev=oops; run meraki_load_net meraki_checkpart meraki_bootlinux`
https://forum.openwrt.org/t/solved-mx60w-commands-dont-work/12935/5

Then I connected the WAN port and used wget to pull the latest image
http://downloads.openwrt.org/snapshots/targets/apm821xx/nand/openwrt-apm821xx-nand-meraki_mx60-squashfs-sysupgrade.bin

and used

sysupgrade <image>.bin

selected NO for keep config, upon reboot everything is happy as a clam.

I was NOT able to get a recovery partition built as I couldn't decipher how to mess with UBI without bricking the board_config.

Thanks for the reply.

How big is the board_config partition? ubinfo -a in the OpenWrt prompt should tell you. If it's tiny (1 MiB or less) then you probably need to mess with the board_config to begin with, unless you absolutely want to squeeze out every KiB there is. From what I know the MX60 has like a GigaByte of NAND so space is not really a problem there.

1 Like

ubinfo -a found me my recovery partition of ubi0_0 which is what I couldn't find yesterday. Thanks for your help!

These are really nice devices actually, the only issue is the CPU is a little weak for high intensity tasks, but only compared to the new routers with quad core whatevers.

Thanks a bunch, I think someone who knows what they are talking about should mark this one solved and update the wiki commands.

This did it! I was able to get the image installed on the router, but the only issue is I can't access 192.168.1.1 or anything like that. However, I can still restart the router and have it still show have Open Wrt installed.

So I guess the biggest issue for this was that TFTP wasn't communicating because the Firewall on the computer was still in place.

1 Like

Not for me, it was loading into the wrong memory address, nothing to do with the firewall. Something about the wiki commands redirected the loading. The new command caused the load to occur in the correct memory, so it worked.

Now that it’s installed, just install luci through the command line.

1 Like

You got it, you now need to install luci using the console.

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.