Ota upload can not be completed about lws-esp32-factory

[lws-team]

You have some more info than "cpu crashes", like a backtrace?

[nodi-andy]

yes. The specific error is "4: lws_spa_destroy: destroy spa 0x3ffd66d0<\r><\n>
Guru Meditation Error of type LoadProhibited occurred on core 0. Exception was unhandled.<\r><\n>"

and here is the core dump:

4: writing 0x2c5e49... 0x2c6e48<\r><\n>
4: LWS_H2_PPS_UPDATE_WINDOW: sid 1: add 65536<\r><\n>
4: LWS_H2_PPS_UPDATE_WINDOW: sid 0: add 65536<\r><\n>
4: writing 0x2c6e48... 0x2c7e47<\r><\n>
4: heap :113648 (+776)<\r><\n>
4: ah excessive hold: wsi 0x3ffdfda4<\r><\n>
peer address: 192.168.178.21<\r><\n>
ah rxpos 0, rxlen 0, pos 317<\r><\n>
4: post = /otaform<\r><\n>
4: origin: = https://192.168.178.51<\r><\n>
4: accept: = /<\r><\n>
4: accept-encoding: = gzip, deflate, br<\r><\n>
4: accept-language: = de,en;q=0.8,tr;q=0.6<\r><\n>
4: content-length: = 2249798<\r><\n>
4: content-type: = multipart/form-data; boundary=----WebKitFormBoundaryteDppABNkHLZf8fR<\r><\n>
4: referer: = https://192.168.178.51/<\r><\n>
4: :authority = 192.168.178.51<\r><\n>
4: :method = POST<\r><\n>
4: :path = /otaform<\r><\n>
4: :scheme = https<\r><\n>
4: user-agent: = Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36<\r><\n>
4: lws_spa_destroy: destroy spa 0x3ffd66d0<\r><\n>
Guru Meditation Error of type LoadProhibited occurred on core 0. Exception was unhandled.<\r><\n>
Register dump:<\r><\n>
PC : 0x4011d66c PS : 0x00060f30 A0 : 0x801282e1 A1 : 0x3ffc2710 <\r><\n>
A2 : 0x99400099 A3 : 0x993ffd99 A4 : 0x00000000 A5 : 0x3ffdfda4 <\r><\n>
A6 : 0x00000009 A7 : 0x3ffd4478 A8 : 0x8011d722 A9 : 0x3ffc26d0 <\r><\n>
A10 : 0x40120cd0 A11 : 0xffffffff A12 : 0x00000005 A13 : 0x00008f29 <\r><\n>
A14 : 0x3ffdffa4 A15 : 0x3ffdfda4 SAR : 0x0000000c EXCCAUSE: 0x0000001c <\r><\n>
EXCVADDR: 0x99400157 LBEG : 0x400014fd LEND : 0x4000150d LCOUNT : 0xfffffff5 <\r><\n>
<\r><\n>
Backtrace: 0x4011d66c:0x3ffc2710 0x401282de:0x3ffc2870 0x40128379:0x3ffc2ae0 0x4011dc0d:0x3ffc2b00 0x40100cb0:0x3ffc2b20 0x400d0e68:0x3ffc2b60<\r><\n>
<\r><\n>
Entering gdb stub now.<\r><\n>
$T0b#e6

i have to also mention about randomness, that sometimes there is not a total crash but upload starts from beginning but ends altough with a crash

[lws-team]

Please retry with current -factory... notice it's built against a specific esp-idf commit mentioned in the README.

[lws-team]

I can complete uploads just fine with current stuff. If you still have a problem, let me know.

[nodi-andy]

I cant still upload. May be it is about my network connection but here is my script, if you find any error and may be you can use the script for documentation

"bash..."
export IDF_PATH=/home/andy/esp/esp-idf
printenv IDF_PATH
export PATH=$PATH:/home/andy/esp32/xtensa-esp32-elf/bin
printenv PATH
git clone https://github.com/espressif/esp-idf
cd esp-idf
git reset --hard 0c50b65a34cd6b3954f7435193411a88adb49cb0
cd ..
git clone https://github.com/warmcat/lws-esp32-factory
cd lws-esp32-factory/
git submodule init
git submodule update --recursive
make menuconfig
make
cd ..
git clone https://github.com/warmcat/lws-esp32-test-server-demos
cd lws-esp32-test-server-demos
git submodule init
git submodule update --recursive
make menuconfig
make

[lws-team]

I dunno if it's the cause of the different behaviour, but esp-idf has submodules too. After forcing the checkout with git reset --hard, you must run something like git submodule update --recursive there too to align its submodules to the older checkout.