- OvmsDev - Open Vehicles

Web frontend
by Michael Balzer 30 Jan '18

30 Jan '18

A first usable web frontend version is now in the hub. There are two new mongoose config options to enable serving from file space and list directory contents. I recommend enabling them, if you want to exclude access to the filesystem, you can do so by configuration params "http.server": // "http.server"Instances: // Name Default Function // enable.files yes Enable file serving from docroot // enable.dirlist yes Enable directory listings // docroot /sd File server document root // auth.domain ovms Default auth domain (digest realm) // auth.file .htpasswd Per directory auth file (Note: no inheritance from parent dir!) // auth.global yes Use global auth for files (user "admin", module password) File access is not needed for the web frontend though, just an option to be able to access i.e. CAN logs. The docroot can be set to "/store", but I don't recommend that. The config path is protected, but this relies on mongoose delivering correct paths -- a bug would open access to the config. If you want to serve from /store, use a sub directory like "/store/www" as the docroot. Global and web session authentication is bound to the module password (config set password module "…"). The frontend is AJAX based using Bootstrap & jQuery, with currently a very basic (functional) design. I've implemented a first home & status screen (just command outputs), a simple shell and config pages for the password and the basic vehicle setup (just type & id for now). I'll add config pages for wifi, modem, server & ota during the next days. The frontend is usable on a small screen, just the shell could need full screen support -- I'll have a look at that later on. The framework provides a page registration and page context utilities to enable custom / vehicle specific extensions with few effort. I'll add an example / template for this. Feedback is welcome. Regards, Michael -- Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal Fon 02333 / 833 5735 * Handy 0176 / 206 989 26

3 11

Um... what just broke?
by Greg D. 25 Jan '18

25 Jan '18

Hi folks, I just noticed I was one commit behind, so did a fetch and make app-flash. No joy... What broke? (and, how to fix?) Greg greg@linux-0rpb:~/greg/ovms/Open-Vehicle-Monitoring-System-3-master/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3> make app-flash Flashing app to serial port /dev/ttyUSB0, offset 0x10000... esptool.py v2.1 Traceback (most recent call last): File "/home/greg/esp/esp-idf/components/esptool_py/esptool/esptool.py", line 2524, in <module> _main() File "/home/greg/esp/esp-idf/components/esptool_py/esptool/esptool.py", line 2517, in _main main() File "/home/greg/esp/esp-idf/components/esptool_py/esptool/esptool.py", line 2246, in main esp = chip_class(args.port, initial_baud) File "/home/greg/esp/esp-idf/components/esptool_py/esptool/esptool.py", line 177, in __init__ self._port = serial.serial_for_url(port) File "/usr/lib/python2.7/site-packages/serial/__init__.py", line 88, in serial_for_url instance.open() File "/usr/lib/python2.7/site-packages/serial/serialposix.py", line 268, in open raise SerialException(msg.errno, "could not open port {}: {}".format(self._port, msg)) serial.serialutil.SerialException: [Errno 2] could not open port /dev/ttyUSB0: [Errno 2] No such file or directory: '/dev/ttyUSB0' make: *** [/home/greg/esp/esp-idf/components/esptool_py/Makefile.projbuild:62: app-flash] Error 1 greg@linux-0rpb:~/greg/ovms/Open-Vehicle-Monitoring-System-3-master/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3>

1 1

What is the format of ms_v_charge_timerstart?
by Geir Øyvind Vælidalo 23 Jan '18

23 Jan '18

Hi, What is the format of ms_v_charge_timerstart? I can’t find any info about it anywhere... Best regards, Geir

3 5

SPIFFS or FAT for /store
by Mark Webb-Johnson 23 Jan '18

23 Jan '18

ESP IDF 2.1 only had FAT support as a filesystem. We used this for the /store vfs mount, for config and script storage. ESP IDF 3.x now has SPIFFS as an option. SPIFFS is obviously more suited to embedded flash. What do people think about making the switch to SPIFFS for /store now? Doing so would probably not be hard (just a few lines of code change in the mounting of that /store partition), but would lose all config and script data in that partition (developers would have to disable vfs protection in menuconfig, then scp out first, to backup). Doing it later is not really an option. Thoughts? Mark

5 8

Transition to new esp-idf
by Stephen Casner 22 Jan '18

22 Jan '18

We expect production OVMS to be based on release/v3.1 of esp-idf in order to incorporate the support for SPIRAM (PSRAM). I'd like to set a target for all develpers to move to the master branch of openvehicles/esp-idf that contains the finished version of my modifications to support the "module memory" command. At the same time I need to then merge the corresponding changes from my temporary branch in openvehicles/Open-Vehicle-Monitoring-System-3 to the master branch there since the API between the two has changed. In other words, we need a flag day. -- Steve

5 46

Boot Loop
by Tom Parker 22 Jan '18

22 Jan '18

The following script reliably puts the ovms into a boot loop: vfs mkdir /store/events vfs mkdir /store/events/system.start vfs rm /store/events/system.start/startmeup vfs append "log level debug" /store/events/system.start/startmeup vfs append "server v2 start" /store/events/system.start/startmeup Press the reset button and it loops with the log below. It can be fixed by "make erase_flash flash" but this obviously erases your config. If you swap the order of the two commands in the script, or if the script only contains on or other of the commands then it doesn't boot loop. ets Jun 8 2016 00:22:57 rst:0x1 (POWERON_RESET),boot:0x1f (SPI_FAST_FLASH_BOOT) configsip: 156795334, 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:0x3fff0018,len:4 load:0x3fff001c,len:4456 load:0x40078000,len:0 ho 12 tail 0 room 4 load:0x40078000,len:12988 entry 0x40078d14 I (168) ovms_main: Set default logging level for * to WARN Welcome to the Open Vehicle Monitoring System (OVMS) - Async Console OVMS > ***ERROR*** A stack overflow in task Housekeeping has been detected. abort() was called at PC 0x4008a824 on core 1 0x4008a824: vApplicationStackOverflowHook at /home/ubuntu/esp/esp-idf/components/esp32/./panic.c:572 Backtrace: 0x4008a70c:0x3ffcea70 0x4008a80b:0x3ffcea90 0x4008a824:0x3ffceab0 0x4008791c:0x3ffcead0 0x4008921c:0x3ffceaf0 0x400891d2:0x632e6769 0x4008a70c: invoke_abort at /home/ubuntu/esp/esp-idf/components/esp32/./panic.c:572 0x4008a80b: abort at /home/ubuntu/esp/esp-idf/components/esp32/./panic.c:572 0x4008a824: vApplicationStackOverflowHook at /home/ubuntu/esp/esp-idf/components/esp32/./panic.c:572 0x4008791c: vTaskSwitchContext at /home/ubuntu/esp/esp-idf/components/freertos/./tasks.c:4837 0x4008921c: _frxt_dispatch at /home/ubuntu/esp/esp-idf/components/freertos/./portasm.S:406 0x400891d2: _frxt_int_exit at /home/ubuntu/esp/esp-idf/components/freertos/./portasm.S:206 Rebooting... ets Jun 8 2016 00:22:57

2 1

Server v3, proof-of-concept
by Mark Webb-Johnson 22 Jan '18

22 Jan '18

I’ve committed a proof-of-concept for the server v3. Trying to keep this to just simple pure MQTT, so it can operate with any standard MQTT server. Of course, we’ll need our own api conventions on top of MQTT, but the basic metrics should work with any standard client/server. Screenshot below is from MQTTInspector on iPad, and I’m using a mosquito server. Don’t try this over cellular, as it currently sends metrics in real time as they change. For the final implementation, we’ll poll them in a similar way to v2 and only send changed ones once every N minutes. Regards, Mark

2 5

OVMS v3.0 modem active antenna modification:
by Mark Webb-Johnson 20 Jan '18

20 Jan '18

Here is the modification for active GPS antenna on OVMS v3.0 modems. I’m having a bunch made up pre-soldered and pre-tinned, so just need to be soldered at two points. I will send those out by post when I get them. Regards, Mark.

2 2

DB29 expansion GPIO
by Michael Balzer 20 Jan '18

20 Jan '18

Mark, I'm currently planning the SimpleConsole connection to v3.x. Given the schemes, it seems the DB29 connector does not offer direct connections to the GPIO lines -- is that correct? GEP_1…7 are not the GPIO lines but currently unused expansion pins on the piggy board connectors? So without additional expansion boards only EXP_1 & _2 are available as general purpose IO pins on the connector? Thanks, Michael -- Michael Balzer * Helkenberger Weg 9 * D-58256 Ennepetal Fon 02333 / 833 5735 * Handy 0176 / 206 989 26

2 1

Missing File?
by Tom Parker 20 Jan '18

20 Jan '18

I just updated to the latest IDF and I'm getting this error: CXX build/main/ovms_module.o /vagrant/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3/main/./ovms_module.cpp:40:28: fatal error: esp_heap_debug.h: No such file or directory compilation terminated. /home/ubuntu/esp/esp-idf/make/component_wrapper.mk:273: recipe for target 'ovms_module.o' failed make[1]: *** [ovms_module.o] Error 1 If I delete line 40 of ovms_module.cpp to avoid trying to import this file then everything compiles. Well it compiles if I turn off the mdns server in menuconfig. If I leave that as the sdkconfig.default.hw30 values then I get this error: CXX build/ovms_mdns/src/ovms_mdns.o In file included from /vagrant/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3/components/ovms_mdns/src/ovms_mdns.cpp:36:0: /vagrant/Open-Vehicle-Monitoring-System-3/vehicle/OVMS.V3/components/ovms_mdns/src/ovms_mdns.h:48:5: error: 'mdns_server_t' does not name a type mdns_server_t *m_mdns;

3 4