-
-
Notifications
You must be signed in to change notification settings - Fork 80
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Latest version of controller code only supports V2 of influxdb #47
Comments
Hello, if you are running v1.8 or newer of Influx, it should support the forward compatibility interface. https://docs.influxdata.com/influxdb/v1.8/tools/api/#apiv2write-http-endpoint In InfluxDB 2.0 uses API Tokens to access the platform and all its capabilities. InfluxDB v1.x uses a username and password combination when accessing the HTTP APIs. Use the Token schema to provide your InfluxDB 1.x username and password separated by a colon (:). For example: username:password. |
When I get my esp32 version of the controller up and running, I'll have to try this. |
If someone can test this and verify, then we can close this (hopefully) non-issue |
Has anyone tested this? |
Not me.. I don't have an ESP32 controller yet.. |
hello, just tested it with influxdb 1.8 on a windows10 machine. So assuming that's what it's meant to log, it's working :-) BTW, BMS is connected via CAN to Victron (raspberry pi 3B+ running VenusOS 2.80Large) and through that I also get bank voltage/current/watts and all warnings from the system as sent to Victron. Wonder if they should also be sent to be logged directly from the bms influxDB interface? cheers |
still works without crashes in diyBMS 10h later. Excuse the regular jumping about it's 4 lifepo4 modules connected to one half dead 18560 cell each with some tiny cables, when it starts balancing all hell breaks loose. |
hm, influxDB works fine, but now controller started rebooting. Was away from the pc, will try to note down how often/when it reboots. Sadly cannot easily pick it up/record it from influx... V. |
looks like circa 2h between controller reboots. |
Hi, there was a bug on the influxdb which was fixed a few months. Are you running the latest version? I really could do with the latest version being applied and then the serial port log files being captured when the reset happens. |
yes Stuart, I'm aware of that, running the latest version (loaded it 4-5days ago):
for now I'm trying to log reboots (not easy when not working on my desk though...), changed to a powerbrick for the ESP32 (plan to get it running from the pcb power input tomorrow if I have time) since yesterday I had (at least recorded maybe a few more I missed): 19:15 I'd also be happier if we could sent to influx on set intervals (like 30s or 1m) - don't want all this traffic from the RUT240 onboard nor the data charges... Shall I put up a new issue for that? cheers V. |
Yes please raise a new issue for the influx timing. |
OK, will do, now how do I log serial port log? any pointers? |
Probably just connecting a laptop to the USB port on the ESP32 - then use a serial terminal like PUTTY to log everything to a file. Then just wait until something goes wrong! More info about it here... |
lol, was hoping for something more advanced than waiting looking at the putty screen tbh 😁 anyway, only took 50mins before hearing the relay clicking (ie rebooting!)
|
Thanks
outputs...
|
way above my paygrade Stuart, if you want anything else let me know. |
not sure how helpful (definitely not conclusive!) changed the 8sec to 60, system survived 3h before rebooting but was also playing with overcharging the cells to check the behaviour so not quite sure. |
Dumb question... Did you try a different ESP32 yet? I have had an instance (on some other project) where one module behaved poorly.. I was scratching my head for a while and I tried a different ESP32 and it worked fine. So I gave the bad module an early, forced retirement! :-) |
not dumb at all 😀 luckily I have a second ESP32 board ready and programmed on my desk - plan is to have one board running on the boat with the rpi3 and a second one running at home for testing with the second PEAK CAN on a r4 V. |
issues threads re influx are getting slightly messy and mixed up (imho) not sure how to report so listing as concise as I can here my last three days observations:
so influxdb activation affects venus can bus as well as causing system reboots. cheers V. |
typical...
ignore all the touchscreen events, was just checking thing was still working... following a raspberry pi reboot I get lots of :
Only solution seems to be to reboot ESP32. |
I'm wondering if this is an issue with the raspberry pi version of cerbo? With a real device I've left it running for days without issue. I generally don't use the influxdb interface though, only the MQTT one. |
@stuartpittaway an update: was only using CANL and CANH from the P-CAN to the diyBMS. Added GND to pin3 (of the P-CAN). Now I can even unplug the USB and replug it (have to go to the rpi settings and re-edit can(2) settings) or unplug and replug on the diyBMS side with rpi picking it up straight away. So seems that CANBUS issues were mine. Influx issues most likely remain will continue testing and will revert to sending every 10secs to get to the problems faster... Will also try without the CANBUS connection for good measure. Waiting for a GX device sometime next week from a friend for further testing, but looks like there is something going on in influxDB code/comms. So updated list of comments:
Hope that helps. cheers |
that was quick rebooting again:
|
This is a sign of the LwIP stack running out of buffers for TX/RX activities. Increasing the data publish rate from the default 8sec to 60sec should reduce the occurrence of this condition. There are a few other possibilities but it would need more in-depth testing. |
thanks, well done that and went a bit further. the only warnings I get now are these:
See if it makes it overnight at 60sec intervals 🤞 V. |
happy new year to all and an update 😁 redid all my test setup at 3S2P (3new and 3s/h 18650cells that is and a .7A engineroom light 12V), better wiring all around and a 2.7A powersupply to the terminals of the board. I'm posting the backtrace from 3-4 of such reboots if it's going to be of any help: after 1h charging:
after 90min charging:
after 2:15h discharging:
after 1h charging:
tcp_out, tcpip, tcp and AsyncTCP feature on most of them... tbh I'm not too concerned about system rebooting (OK, obvs would be nice if it didn't!) since it's a minimal disruption to the operation of the whole BMS and as long as it cleanly reboots and all systems are fine afterwards (seems to be the case in the 2weeks of testing here) cheers V. |
fwiw, 60sec interval in influxDB v1.8, venus CANBUS connection working fine, latest s/w, uptime 6days and counting, charging/discharging 3S all the time. V. |
this issue is also completely solved as influxDB 1.8 is supported, and with the NOV2021 code I've yet to see a reboot. V. |
Will close, I only really left it open as instructions on how to use it with 1.8. |
There are a lot of us out there that don't want to/ can't upgrade to V2 of influxdb.
It would be really nice if we had the option of staying with the 1.x implementation.
The text was updated successfully, but these errors were encountered: