Powewall-Dashboard Power Flow graphic just "scans" #575
Comments
JonMurphy
changed the title
|
It looks like pypowerwall is not running or is unable to connect to your Powerwall. docker logs pypowerwall -fDo you see errors? Also try a full restart: # stop and remove
docker stop pypowerwall
docker rm pypowerwall
# rebuild
./compose-dash.sh up -d
Were you using this address as the endpoint? If so, it looks like you do not have a route to that IP address. The add_route.sh script can help set up route to the TEDAPI endpoint on your new RPi if that is what you are looking to do. |
If I look for just ERROR, there seem to be lots! Is did a ping and I can see the PW3: though it "feels" slow...
will this delete my old data? This has been running since the end of October 2024. |
|
It all suddenly came back though I don't know why.
Let me know if you would like me to look for anything in the logs. EDIT: added current
EDIT2: Just one ERROR below. |
|
Hi @JonMurphy - I'm glad it is back, but you aren't out of the woods yet... :)
No, pypowerwall is just the proxy to get to the Powerwall data. The data is stored in Influxdb and actually in the influxdb directory.
Yes, pypowerwall is having trouble getting data from your Powerwall. This usually indicates a network issues which is confirmed with your ping results dropping 50% of the packets. Some thoughts/suggestions:
|
Got it! I did this:
Done! But still lots of messages in the log. I'll post below.
Yes, Powerwall 3 is connected via wi-fi only. No ethernet. I was told in October 2024 that Powerwall 3's do not connect via ethernet (but this sounded like BS to me). To me this seems OK. And I can move it closer.
I will do this next.
This is what I see now: Here is the log: Since I see Timeout does that mean wi-fi is still slow? Thank you for your guidance! |
Yes, all of these errors are indicating network issues. How are you connecting to 192.168.91.1? Is it via a route or directly to the TEDAPI WiFI access point? If your Powerall has a LAN IP address (WiFi) you could try the add_route method (see https://github.com/jasonacox/Powerwall-Dashboard?tab=readme-ov-file#extended-device-vitals-metrics). Also, you seemed to indicate that it was working but not it is not. Did anything else change on your network or host? I don't know if it would help, but you could also try a cool-down restart: # Stop pypowerwall
docker stop pypowerwall
docker rm pypowerwall
# Test Ping - Make sure you don't see failed packets.
ping 192.168.91.1
# Wait 10m or so just to ensure there isn't a rate limit happening on the Powerwall
# Restart
./compose-dash.sh up -d
# Watch logs
docker logs pypowerwall -f |
My RPi5 with on-board wi-fi enabled connects to the PW3 at 192.168.91.1. PW3 (192.168.91.1) <<-- via WiFi -->> RPi5 onboard wifi (192.168.91.140) RPi5 << -- via ethernet (192.168.65.1/24) -->> desktop computer
I may not be understanding what you are asking. . . The RPI5 onboard Wifi connects directly to the PW3 access point via DHCP. For the RPi5 wi-fi I had to enter the SSID:"TeslaPW_XYZXYZ" and the password from the inside of the cabinet like this: Here is the route I see:
The route already existed when I viewed
It started working at Post 4 (above) This is current screenshot from a few moments ago: It appears to be working but with 1000s of these type of messages: CLICK ME to reveal long log
Nothing had changed when it started working yesterday Post 4 (above). And I do not have a clue what caused it to start working yesterday.
I'll try this now and report back. As a side comment. The PW3 did receive a new firmware update 3 days ago (from 24.36.11 to 24.44.2) on 2025-01-18 00:44:06 CST (Chicago time)
(I never knew there were SO many of these indicators!) |
|
I did the cool-down restart:
Here is the ping (some good, some 10% packet loss) during the And here is the log after the restart |
|
Thanks for the explaination @JonMurphy and the nice screenshot of the alerts! Yes, there are a lot of alerts. That's a good selection of them. On this diagram: PW3 (192.168.91.1) <<-- via WiFi -->> RPi5 onboard wifi (192.168.91.140) RPi5 << -- via ethernet (192.168.65.1/24) -->> desktop computer Is your Dashboard (pypowerwall) running on the RPi5 or the desktop computer? I'm trying to see how many hops it takes to get to the PW3. In any case, your network path from the Dashboard (pypowerwall) to the Powerwall is not great (any packet loss is not good, I woujld prefer to have higher latency rather than packet loss). However, since pypowerwall will recover from timeouts, enough queries are going to be able to get through to service telegraf which feeds InfluxDB and the Grafana dashboard. This may be good enough. Which is fine. Personally, I'm sort of obsessive and compulsive about correctness so I would be driven to get the network path to 100%. My suggestion on that would be to remove the RPi5 WiFi to PW3 link, and instead, let the PW3 connect to your local WiFi network. It will get an address like 192.168.65.100 let's say for example (you can find it using |
the Dashboard runs on the RPi5 (headless). Only a browser on the desktop.
Agree!
The PW3 is already connected via 192.168.65.149 to my house wifi. The PW3 seems to be communicating to both the home wifi and via it own access point via 192.168.91.1.
FYI - this did not work for me. does this need to be run from the docker world?
Done!
Testing it now. It seems to take a few minutes before all of the tracebacks happen. . . EDIT: So far so good after 25 minutes: Hopefully I did not jinx it! EDIT2: Only two more lines. What is the THANK YOU! THANK YOU! THANK YOU! for all of your help! I did not realize that I could directly access the PW3 via my home wi-fi. I thought I could only access it via 192.168.91.1 (or the Tesla server API). Thank you for pointing me in the right direction and for all of your help. I do appreciate it! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hello!
I need your help!
I tired to update my RPi5 for use with an Argon NEO 5 NVMe case and its NVMe drive but I screwed up everything. Currently my TeslaMate stuff seems to work fine but the Powewall-Dashboard Power Flow graphic just does this:
I ran verify.sh and I see this, but I am not sure how to precede.
Can you point me in the right direction?
The text was updated successfully, but these errors were encountered: