Repository navigation
|
Hi, it seems I either misconfigured something, or the 2.5.1 build is really chatty. But so was the 2.3.3 I was running before. I was realising my switches lamps were flickering quite a lot synchronously on all ports. And now I finally decided to have a look what's causing it. Wireshark capture leads to my energyme home. 2026-10-07T11:16:51.945Z 9070698753d0[299101]: [DEBUG][Core1] src/ade7953.cpp[_printMeterValues]: Wärmepumpe Steuerung (0): 234.3 V | 0.215 A || 32.7 W | 10.9 VAR | 50.5 VA | 64.7% || 55350.501 Wh <- | 0.000 Wh -> | 37593.077 VARh <- | 1888.364 VARh -> | 91240.700 VAh Question is: Why? And how do I prevent it from doing this? |
Replies: 2 comments 2 replies
|
Some investigation on the source code of the project revealed a python script in source/utils/udp_log_listener.py. This uses a http call to configure the functionality. Looking at my device
Now I only need to fund out how to convince my device to shut up 😝 |
|
Claude's suggestion did the trick: However, how is this "standard" behaviour? Why flood the customers network with multicast syslog UDP packets. Also here Claude had a possible explanation:
So this sounds like a plausible explanation of my cause. Maybe it affects someone else, hence I'm documenting it extensively. If I may recommend something more: Why not introduce a dedicated loglevel configuration for the syslog UDP? A UDP level that persists like Print/Save and is exposed through the /api/v1/logs-level handler would make sense in my opinion. One thing to watch: that callback level also feeds the MQTT log callback. So either it becomes one shared "callback level", or UDP gets its own check inside _callbackUdp. |
Claude's suggestion did the trick:
However, how is this "standard" behaviour? Why flood the customers network with multicast syslog UDP packets. Also here Claude had a possible explanation: