phoscon 2.19.03 (but I had the issue also with previous versions)
homeassistant 2022.12 (but I had the issue also with previous versions)
problem:
I am experiencing a weird problem where sometimes my homeassistant automations don’t start straightaway (i.e. light do not start), but only after a few seconds. It seems that the communication between the devices and phoscon get stucked for a few seconds (sometimes a couple of seconds, sometimes also 30 seconds), and then it get played after it get un-stuck. This has been happening since the last month or so.
Note that when there is this problem, nothing works, not just one light.
Example: I enter a room, nothing happens (even though light should turn on), I go away, from another room I see the light turning on after some time, even if I was in the room 30 seconds ago. So the message is sent eventually, but there’s lag.
what I tried:
enable logs in phoscon, cannot see anything clearly wrong. A few times when the problem happened I ran to look at the logs and I can see that there is “stuff happening”, even though I am unclear of what the log messages mean, I can see that there’s communication with the devices.
update phoscon to latest version (I am having this problem with at least the last 2 versions)
update homeassistant multiple times to latest version
change zigbee channel
change home router to 5ghz only
disable bluetooth and wifi on raspberry
update python to 3.11
move raspberry in different rooms of the house
replace sd card
restart phoscon. When restarted everything works again.
I am not sure what else to try to fix this!
Does anyone have any idea of what to try next?
I am going to post logs as soon as it happens again.
Hope I am not talking too soon, but with 2.18.02 everything works seamlessly since yesterday! I was expecting the problem to have already happened as it was happening multiple times per day.
I’ve seen another thread where people are lamenting similar issues, which I can’t find anymore now it was talking about groups being slow to trigger, I think what I am experiencing is something similar.
I think the pastebin I have posted above is a red herring as, when the problem was happening, most of the times there were no errors at all in the logs.
Can I do anything other than what I’ve already done to help diagnose the issue with 2.19.x? Seems I am stuck with 2.18.02 forever otherwise.
I can confirm after 3 days that the problem is not happening on 2.18.02, and was constant in all 2.19.x.
What should I do to raise this? I don’t think I have a particular setup and I’ve really tried anything that I could to make sure it’s not something specific to my network.
I have dev experience, I could also try to build 2.19.x branches but without some PRs, to isolate what is causing the issue, if this helps?
When tested with 2.20, can you share some new logs and what change in behavior your noticing. I had no issues anymore after going to latest stable. Additionally, a screen of your deCONZ map. A
it was talking about groups being slow to trigger, I think what I am experiencing is something similar.