nickrout, I’m not sure what you are asking.
I’m trying to see if I have too many devices and solve the problem of latency or non-function in my zigbee network as stated.
nickrout, I’m not sure what you are asking.
I’m trying to see if I have too many devices and solve the problem of latency or non-function in my zigbee network as stated.
Which Tube’s coordinator do you have?
Sorry probably an idiom translation issue
The issue is not likely to be a ZHA capacity issue unless you have a particularly wek computer. A capacity issue is far more likely to come from a co-ordinator + router limitation.
Thanks tom_I, I have a CC2652P2 Based Zigbee to Ethernet Serial Coordinator
Plus I have 3 Router/Repeaters
That coordinator supports: 50 Direct Children, 100 normal routes and 200 source routes
So, assuming your routers are connected directly to your coordinator, you can have no more than 47 other devices connected directly to your coordinator. All the rest must go through the routers.
Do you have source routing set up?
Your router only supports 100 normal routes and you have 176 devices.
Thanks tom_I I do not have “source routing” set up as far as I know. I have attempted to bind groups to certain routers, but can’t quite tell if it is working? I googled to see what source routing is in ZHA. But can’t find much. Can you give me some info about that. Sounds like that might be just what I need. Thanks so much for your help
I don’t know anything about ZHA. I use Zigbee2MQTT.
I also have a small network, so don’t really know that much about it. Hopefully someone else can help.
Did things ever come back? Capacity wise with a mix of routers that number of devices should not be an issue. I know others running with 150+ without issues.
I’ll re-iterate it’s hard to know what’s going on with out some debug logs. Use pastebin or similar to post.
tube0013, It is intermittent, and even when it does come up often the message to turn the light on or off or change settings does not ever get delivered, and sometimes its a long time before it changes. I will get some logs. What is the best way to collect the logs? I don’t see anywhere in the ZHA integration that shows logs, there are some in Configuration | Logs. Is that what you are referring to? Also there is the Home-Assistant log file in Visual Studio Code, which I think is the same. Thanks for sticking with me. This is a big zigbee network, and its my first time with zigbee.
instructions for enabling debug logging is here:
https://www.home-assistant.io/integrations/zha/#debug-logging
it will all go into your homeassistant.log it will be pretty voluminous as well.
Thank you tube0013, Here are my logs:
https://zerobin.net/?81c489e4e7e5519d#TE5lh/l/u8SGV1XnHPY7zxe1He0K7h7eFjtUBhW3fM0=
One of the common ones I see in my log is this:
Logger: zigpy.util
Source: /usr/local/lib/python3.9/site-packages/zigpy/util.py:68
First occurred: 1:56:56 PM (66 occurrences)
Last logged: 3:04:28 PM
Error calling listener: Data is too short to contain 4 bytes
I actually think my issues are related to Python 3.9. Seems that is where most of my errors are.
I haven’t really had a chance to look at this much. it’s best too to limit the log to around the times of the issues. also posting about the problems in the HA discord #zigbee channel may not be a bad idea to get some more eyes on it, as I don’t think this is necessarily an issue with hardware which is where I’d want to keep this thread focused on.
I will let you know if I see anything obvious though.
I’m checking with the zigpy-znp dev on what these are:
[zigpy_znp.zigbee.application] Max concurrency reached, delaying requests (63 enqueued)
Thanks for looking at that.
Mini update.
Have you seen these at Digikey? Not sure what type of pricing you were getting direct from the manugacturer…
Or these from SeedStudio (but ship from China, I believe)
I have, and even though they are pictured without headers, they will come pre-installed, as the Manufacturer doesn’t make them without. I was getting some pulled from the line before the headers were put on, but it limited me to only getting them when they do a production run, which isn’t happening in the near future.
Ahh, understood… Figured you wouldn’t have missed them…