Example not working #76
Comments
|
What exactly is not working? These errors are expected when no pairing data is present. Put the Nuki device in pairing mode and these errors will disappear. |
|
I have a Nuki 2.0 and a Nuki 3.0 but none of them "in binding" mode will change this behavior :-( |
|
You should enable debug logging using the Arduino/ESP-IDF settings in whatever build system you are using to compile and define appropriate debug vars. Might be better to start from a working implementation such as nuki hub (complex) or esphome_nuki_lock (basic) |
|
"nuki hub" is much too complex, would need me to strip down nearly all beside the NUKI plain connect part. "esphome_nuki_lock" is even a big framework (behind ESPHome) , that's why i directly choose NukiBleEsp32 as start for a own project. I simply downloaded the zip from here, extracted it, opened it in VSCode/PlatfomIO, build the example and uploaded it (to different ESP32 Wroom Boards with 4MB Flash) with the result i showed above. As quickcheck i downloaded and build and uploaded "nuki hub" now only to check if the locks and the ESP32 Hardware is working as expected. I never used debugging on a ESP32 (beside Serial) so no clue at the moment to to debug this problem. Not sure how to continue now. |
|
Here are two serial output snippets from debug build: (cut from long serial output) Only difference: If the lock is not in binding mode then it looks like that I marked the line with ************************** at the line end
|
|
I now installed a fresh VSCode with fresh installed PlatformIO . |
|
Below is a serial output from the powerup, a Nuki V3 was in bind mode, so the typical I tracked down the message "No service data found" in sourcecode "NimBLEAdvertisedDevice.cpp" I still think that something is wrong or missing in the example code. |
|
Finally found this older issue ( #38 ) That gave the hint to a specific branch "pair-and-unpair-refactor" ( https://github.com/I-Connect/NukiBleEsp32/tree/pair-and-unpair-refactor ) I tried it and this one immediatly worked out of the box !!! |
|
Good NEWS !!! by comparing line by line and having a lot additional log outputs i tracked down the binding problem to only a couple of lines where the UUIDs are defined. If i change this to: then binding even works now in the "main" branch, no need to go for the "pair-and-unpair-refactor" anymore Please note, both version will compile without errors, but with the not working one binding itself will not work at all, see my comments above. |
|
I have no explanation why this makes a difference, both should create identical NimBLEUUID objects. Could you add some debug output inside NukiBle.cpp for testing, to see if there's a difference. The NimBLEUUID class has a toString() method you can use. Both ways should have identical output. |
|
No, as far i could see, both create different NimBLEUUID objects. I tried to output the UUID in a readable way, but .toString does not work for me, with it i only get non printable characters. So i have done my own conversion to format the data for outputting it as single chars and hex, byte by byte. e.g. here is a byte by byte representation of pairingServiceUUID in NukiBle.cpp when using directly "NimBLEUUID("a92ee100-5501-11e4-916c-0800200c9a66")," at "NukiLock::NukiLock(const std::string& deviceName, const uint32_t deviceId)" and here are when using "keyturnerPairingServiceUUID" btw: |
|
For a test i changed the following: in Nukilock.cpp in Nukilock.h and in NukiSmartlockTest.h (-> main.cpp) Even if this is not a real solution and only a prove of concept, it will work then. For me it looks like as if at the time when the object NukiLock" is created from "NukiBle" ( see "NukiLock.cpp NukiLock.h") the NimBLEUUID object with the constant UUIDs are not already evaluated. |
|
Strange, I can't really make sense of it, I'll have to think about it. |
When i try the example code (without any Modification) i only get the following error at the serial monitor:
rst:0x1 (POWERON_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT)
configsip: 0, SPIWP:0xee
clk_drv:0x00,q_drv:0x00,d_drv:0x00,cs0_drv:0x00,hd_drv:0x00,wp_drv:0x00
mode:DIO, clock div:2
load:0x3fff0030,len:1184
load:0x40078000,len:13232
load:0x40080400,len:3028
entry 0x400805e4
[ 287][E][Preferences.cpp:503] getBytesLength(): nvs_get_blob len fail: bleAddress NOT_FOUND
[ 287][E][NukiBle.cpp:939] retrieveCredentials(): Error getting data from NVS
[ 296][E][Preferences.cpp:503] getBytesLength(): nvs_get_blob len fail: bleAddress NOT_FOUND
[ 299][E][NukiBle.cpp:939] retrieveCredentials(): Error getting data from NVS
[ 807][E][Preferences.cpp:503] getBytesLength(): nvs_get_blob len fail: bleAddress NOT_FOUND
[ 807][E][NukiBle.cpp:939] retrieveCredentials(): Error getting data from NVS
[ 1311][E][Preferences.cpp:503] getBytesLength(): nvs_get_blob len fail: bleAddress NOT_FOUND
[ 1312][E][NukiBle.cpp:939] retrieveCredentials(): Error getting data from NVS
[ 1815][E][Preferences.cpp:503] getBytesLength(): nvs_get_blob len fail: bleAddress NOT_FOUND
[ 1816][E][NukiBle.cpp:939] retrieveCredentials(): Error getting data from NVS
[ 2319][E][Preferences.cpp:503] getBytesLength(): nvs_get_blob len fail: bleAddress NOT_FOUND
[ 2320][E][NukiBle.cpp:939] retrieveCredentials(): Error getting data from NVS
...
Any Hint where to search ?
The text was updated successfully, but these errors were encountered: