Status.lua - unsupported device

[dman776]

Nanostation M2
dump_perl2 build
The "!!! UNSUPPORTED DEVICE !!!" banner is showing on status.lua (but not on the status page - perl based)

[aanon4]

Can you send me the output of the following (a lua script):

require('aredn.hardware')
require('luci.util')
=luci.util.dumptable(aredn.hardware.get_radio())

[dman776]

root@K5DLQ-UNSM2-100:~# lua
Lua 5.1.5 Copyright (C) 1994-2012 Lua.org, PUC-Rio (double int32)

require("aredn.hardware")
require("luci.util")
=luci.util.dumptable(aredn.hardware.get_radio())
/usr/lib/lua/luci/util.lua:144: bad argument #1 to 'pairs' (table expected, got nil)
stack traceback:
[C]: in function 'pairs'
/usr/lib/lua/luci/util.lua:144: in function </usr/lib/lua/luci/util.lua:140>
(tail call): ?
[C]: ?

[dman776]

Also....
root@K5DLQ-UNSM2-100:/usr/lib/lua/aredn# cat /etc/board.json
{
"model": {
"id": "ubnt,nanostation-m",
"name": "Ubiquiti Nanostation M"
},

[aanon4]

The Ubiquiti stuff has some serious mucking about to get the radio info. How about the output for this:

require('aredn.hardware')
=aredn.hardware.get_board_id()

And this:
/usr/local/bin/get_boardid

[aanon4]

Pinging @dman776 for more info.

[dman776]

root@K5DLQ-UNSM2-100:~# lua
Lua 5.1.5 Copyright (C) 1994-2012 Lua.org, PUC-Rio (double int32)

require("aredn.hardware")
=aredn.hardware.get_board_id()
0xe012

root@K5DLQ-UNSM2-100:~# /usr/local/bin/get_boardid
0xe012

[aanon4]

Okay. This one has me confused then as both the old and new board id functions return the same value, and if I force my build to return that value (fixing the return value of aredn.hardware.get_board_id to "0xe012") the board isn't marked as unsupported.

Above you included board.json which looks truncated and not valid json - does it really look like that? That file being invalid might have some unexpected side effects.

[dman776]

That was a partial snippet from board.json Thanks, Darryl

…

On Dec 21, 2021, at 6:18 PM, Tim Wilkinson ***@***.***> wrote:  Okay. This one has me confused then as both the old and new board id functions return the same value, and if I force my build to return that value (fixing the return value of aredn.hardware.get_board_id to "0xe012") the board isn't marked as unsupported. Above you included board.json which looks truncated and not valid json - does it really look like that? That file being invalid might have some unexpected side effects. — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.

[aanon4]

Could you post the whole thing? I'm somewhat stumped as to what's going on :-(

[dman776]

board.json.zip

[aanon4]

So I bought a M2 node but that didn't show this problem, so I guessed it was actually an old M2 node you had. I couldn't source one of those but managed to locally trade a shiny new M5 for an old M5 and that did show the issue! Turns out these older nodes append an extra character return to the information which I needed to strip:

Fixed here: 7fe5372

[aanon4]

Can we close this now?