Bricked two SqueezeAMPs - how to recover?

[benh]

In fact, just screwing around, I have this one SA that is misbehaving that I can consistently make come online in recovery mode by plugging into serial console. Once online, I can hit the web UI and upload a local copy of a recent firmware. It will go through the upload/upgrade process, proclaim success, and then just...drop offline again. Power cycle, unplug/replug serial console, and it's back online in recovery mode. I don't know why it never actually recovers after the firmware upload.

[sle118]

In fact, just screwing around, I have this one SA that is misbehaving that I can consistently make come online in recovery mode by plugging into serial console. Once online, I can hit the web UI and upload a local copy of a recent firmware. It will go through the upload/upgrade process, proclaim success, and then just...drop offline again. Power cycle, unplug/replug serial console, and it's back online in recovery mode. I don't know why it never actually recovers after the firmware upload.

If you only connect TX and RX plus GND, dis the system come up? If not, then it's the other pins that trigger the success reboot and this could mean that the board somehow gets stuck in flash mode (which the serial adapter gets it out from).

Try only theses pins with a console at 115200/8/N/1 and see what you get

[benh]

(Made some edits to be below after double checking things.)

Connecting only TX, RX and GND does not cause the system to come up.

Connecting only TX, RX, GND, and RST DOES cause the system to come up.

Connecting only TX, RX, GND, and DWL DOES cause the system to come up.

Connecting TX, RX, GND, DWL and RST DOES cause the system to come up.

Console is still garbage:

.5.ßø9..'A.-.5ãøóùìø.'-A9ãèø3'-A9ãùüø.%=''A;øôõø%=.A;ãùìôèó…Iü%ÂÎ…Iüâôô%QøØùëìô≠ø[ueaSuãøcuQø_aagø!3øùùôèõüø;

.5.ßø9..'A.-.5ãøèëõø.'-A9ãùìø3'-A9ãùõø.%=''A;øìóø%=.A;ãùôùüëõ…Iü%ÂÎ…Iüâôõ%møØùëëô≠ ø..-.=%ãø[5.5..-#1ø.!!'ø!3øìóiø!3ø-#.5.#='ø%5%!.

ø3!.øwe}°-#.5.#='ø=''!9=.-!#.…Iü%ÂÎ…Iüâôô%QøØùëèô≠ø[ueaSuãøcuQø_aagø!3øìïïùïø;

.5.ßø9..'A.-.5ãøèüóø.'-A9ãùìø3'-A9ãùùø.%=''A;øìóø%=.A;ãìïïôì…Iü%ÂÎ…Iüâôõ%møØùëçô≠ø 5..A=..A%=-#ãø[5.5.ø.5=.!#ø-.ãøó…Iü%ÂÎ…Iüâôõ%møØùëçô≠ø5..A=..A%=-#ãø[5;!!.ø9!.#.5.ÖùÂÎ…Iü%ÂÎ…Iüâôõ%møØùèüô≠ø5..A=..A%=-#ãøY.=..-#1ø=..A%=-#…Iü%ÂÎ…Iüâôõ%møØùèüô≠ø#..A..-'-.-5.ãøm#-.-='-.-

[sle118]

(Made some edits to be below after double checking things.)

Connecting only TX, RX and GND does not cause the system to come up.

Connecting only TX, RX, GND, and RST DOES cause the system to come up.

Connecting only TX, RX, GND, and DWL DOES cause the system to come up.

Connecting TX, RX, GND, DWL and RST DOES cause the system to come up.

Console is still garbage:

.5.ßø9..'A.-.5ãøóùìø.'-A9ãèø3'-A9ãùüø.%=''A;øôõø%=.A;ãùìôèó…Iü%ÂÎ…Iüâôô%QøØùëìô≠ø[ueaSuãøcuQø_aagø!3øùùôèõüø;

.5.ßø9..'A.-.5ãøèëõø.'-A9ãùìø3'-A9ãùõø.%=''A;øìóø%=.A;ãùôùüëõ…Iü%ÂÎ…Iüâôõ%møØùëëô≠ ø..-.=%ãø[5.5..-#1ø.!!'ø!3øìóiø!3ø-#.5.#='ø%5%!.

ø3!.øwe}°-#.5.#='ø=''!9=.-!#.…Iü%ÂÎ…Iüâôô%QøØùëèô≠ø[ueaSuãøcuQø_aagø!3øìïïùïø;

.5.ßø9..'A.-.5ãøèüóø.'-A9ãùìø3'-A9ãùùø.%=''A;øìóø%=.A;ãìïïôì…Iü%ÂÎ…Iüâôõ%møØùëçô≠ø 5..A=..A%=-#ãø[5.5.ø.5=.!#ø-.ãøó…Iü%ÂÎ…Iüâôõ%møØùëçô≠ø5..A=..A%=-#ãø[5;!!.ø9!.#.5.ÖùÂÎ…Iü%ÂÎ…Iüâôõ%møØùèüô≠ø5..A=..A%=-#ãøY.=..-#1ø=..A%=-#…Iü%ÂÎ…Iüâôõ%møØùèüô≠ø#..A..-'-.-5.ãøm#-.-='-.-

Note that the first bootloader that the ESP32 runs has a different baud rate than 115200 and may cause some garbled content. But the log you are showing seems to indicate that the system's baud rate for console is completely different. This is intriguing to say the least. Philippe is "out of reach" at the moment and if there's anyone that might have a clue it's him (he designed the boards... Several iterations of them).

[benh]

It is interesting. I can't help but think that I have a serial adapter issue of some kind. I think I'll try a different USB-Serial dongle and make up a different adapter.

[philippe_44]

It is interesting. I can't help but think that I have a serial adapter issue of some kind. I think I'll try a different USB-Serial dongle and make up a different adapter.

Just jumping-in but AFAIU, the SA only works when you plug-in the serial adapter, right? If this is the case, it means that you have burnt (likely) the 3.3V DC/DC converter that is on board and you are powering the 3.3V chips from the serial connector

[benh]

It powers up without anything attached to the serial connections. The status led illuminates green and it will blink green twice quickly about every 3 seconds. The red led behind the barrel jack also lights up.

[philippe_44]

It powers up without anything attached to the serial connections. The status led illuminates green and it will blink green twice quickly about every 3 seconds. The red led behind the barrel jack also lights up.

Then it's working normally as the slow blinking means WiFi connection is working in normal mode (not recovery). I'm confused.
[edit]: oh, sorry - twice quickly?

[benh]

Yes, twice quickly.

[philippe_44]

Yes, twice quickly.

Looks like a boot loop. So if you have fully erased the SA with the espressif tools, have a proper 12V power supply, have re-downloaded a full binary set, then best is that you send them back to me. Would be surprising but there might be an issue with HW.

[benh]

I have connected via serial console, which caused the SA to boot into recovery mode and be available via wifi. I used the web UI to upload a squeezelite binary package, such as squeezelite-esp32-master-v4.3-SqueezeAmp-16-21254. That's successful, but it never comes back from that reboot.

I'll try a different 12v power supply. I've also got a different serial adapter coming because I am still suspicious about the fact that the serial console is all garbage. I am giving my current serial adapter some side eye, despite it working fine in other contexts.

[philippe_44]

I have connected via serial console, which caused the SA to boot into recovery mode and be available via wifi. I used the web UI to upload a squeezelite binary package, such as squeezelite-esp32-master-v4.3-SqueezeAmp-16-21254. That's successful, but it never comes back from that reboot.

I'll try a different 12v power supply. I've also got a different serial adapter coming because I am still suspicious about the fact that the serial console is all garbage. I am giving my current serial adapter some side eye, despite it working fine in other contexts.

Have you re-downloaded not just the squeezelite binary but as well all the recovery, partition and al?

[benh]

That I have not done. Is there a way to do that from the web UI? If not, I'll have to solve the console issue first.

[philippe_44]

That I have not done. Is there a way to do that from the web UI? If not, I'll have to solve the console issue first.

No, but you can use sle118,'s web installer. Otherwise, use espressif tools and I think in 1st post of the squeezesp32 we explained what binaries should be downloaded

[benh]

I was correct that something is amiss with my trusty serial adapter. I got one of the FTDI adapters and I have a non-garbage console now, which makes all of this much easier to diagnose.

As far as I can tell, the problematic unit is in a reboot loop, I *suspect* that the reason is that I misconfigured an argument to the squeezelite command and that's making it throw an exception:

****************************
Your terminal application does not support escape sequences.
Line editing and history features are disabled.
On Windows, try using Putty instead.
****************************
W (3423) console: Processing autoexec commands while network manager active.  Wifi related commands will be ignored.
I (3433) console: autoexec is set to perform auto-process
I (3443) console: Running command autoexec1 = squeezelite -b 500:2000 -d all=info -C 30 -W -n Attic -o I2S 192.168.10.40
W (3463) config: Value not found for key autoexec2
squeezelite-esp32> I (3463) squeezelite_cmd: Calling squeezelite

Error: command line argument error

Squeezelite v1.0-1250-16, Copyright 2012-2015 Adrian Smith, 2015-2019 Ralph Irving. See -t for license terms
Usage: squeezelite [options]
  -s <server>[:<port>]  Connect to specified server, otherwise uses autodiscovery to find server
  -a <f>                Specify sample format (16|24|32) of output file when using -o - to output samples to stdout (interleaved little endian only)
  -b <stream>:<output>  Specify internal Stream and Output buffer sizes in Kbytes
  -c <codec1>,<codec2>  Restrict codecs to those specified, otherwise load all available codecs; known codecs: flac,pcm,mp3,ogg (mad,mpg for specific mp3 codec)
                        Codecs reported to LMS in order listed, allowing codec priority refinement.
  -C <timeout>          Close output device when idle after timeout seconds, default is to keep it open while player is 'on'
  -d <log>=<level>      Set logging level, logs: all|slimproto|stream|decode|output, level: info|debug|sdebug
  -e <codec1>,<codec2>  Explicitly exclude native support of one or more codecs; known codecs: flac,pcm,mp3,ogg (mad,mpg for specific mp3 codec)
  -f <logfile>          Write debug to logfile
  -m <mac addr>         Set mac address, format: ab:cd:ef:12:34:56
  -M <modelname>        Set the squeezelite player model name sent to the server (default: SqueezeESP32)
  -n <name>             Set the player name
  -N <filename>         Store player name in filename to allow server defined name changes to be shared between servers (not supported with -n)
  -W                    Read wave and aiff format from header, ignore server parameters
  -r <rates>[:<delay>]  Sample rates supported, allows output to be off when squeezelite is started; rates = <maxrate>|<minrate>-<maxrate>|<rate1>,<rate2>,<rate3>; delay = optional delay switching rates in ms
  -R -u [params]        Resample, params = (b|l|m)[:i],
                         b = basic linear interpolation, l = 13 taps, m = 21 taps, i = interpolate filter coefficients
  -Z <rate>             Report rate to server in helo as the maximum sample rate we can support
  -t                    License terms
  -?                    Display this help text

Build options: EMBEDDED LOOPBACK RESAMPLE16 NO_FAAD LINKALL


assert failed: 0x400d560e

Backtrace:0x40081a9d:0x3f8075b0 0x40097c45:0x3f8075d0 0x4009c2e5:0x3f8075f0 0x400d560e:0x3f807630 0x400dde97:0x3f807650 0x400eeca4:0x3f807980


ELF file SHA256: 7ad37d97020ce9a5

Rebooting...​