Generic Matter Bridge Devices Unresponsive

🛎️ Get Help
1

I am running a C8 Pro on platform 2.4.3.177.

I had the strangest thing happen tonight. I added an Aqara FP300 to my Aqara M2 hub which I bring into my C8 Pro via matter and the Generic Matter Bridge driver. Have used this setup for a year, no issue like what just happened. The device showed up in my Aqara app and once I hit initialize on the M2 hub parent device in HE it brought over the motion sensor, temp, and humidity sensors for the FP300 just like it did the two previous occasions for my other two sensors. It fell right in place with the 37 or so devices I was bringing over already. Totaling around 40 devices.

Then when I went to test the sensor, it would show presence in the Aqara app but would not show active in the device in HE. All my other devices were working fine at the time. I shut down and restarted both the M2 and my HE hub and neither fixed the issue.

Note I also bring this M2 into HA via matter and everything even the new sensor was working correctly in HA, so it isn't a network issue.

In having issues in the past, I removed the Aqara M2 hub from C8 Pro and then generated a new matter code and added it back. Which usually cures all. Here is where it gets crazy. Then after bringing the M2 hub back into HE it showed all the devices as it should in the HE device list as child devices. However, none of them worked. Not one single device worked.

After taking the M2 in and out several times they never worked. They would be present but appear as if there was nothing happening with them. None of them would change state when activated. I run a test setup on my beta hub that is a C8-Pro using the Matter Advanced Bridge by @kkossev so I checked it. All the devices were on that hub from the M2 and working exactly as they should even though they were not working on my production hub.

I went into my C8 Pro production hub and changed the driver from the Generic Matter Bridge driver to the Matter Advanced Bridge driver and then discovered all the devices. They all worked exactly as they should with this driver. Keep in mind I did not delete the matter fabric I had established with the HE driver. Then I went back into the parent M2 hub and changed the driver back to the Generic Matter Bridge driver, hit initialize, and all the original devices found from when I first created the matter fabric started working correctly.

I thought there is no way this happened, so I deleted the M2 hub, joined it back with the Generic Matter Bridge Driver, all devices present, but none worked. Switched the driver back to the Matter Advanced Bridge Driver, done discover all, and those devices found worked perfectly. Switched back to the Generic Matter Bridge Driver and the original devices started working correctly again.

I recreated this three times to verify that what I was seeing is correct. Does anyone have any pointers as to what the issue is that the original devices did not work correctly. I did notice I was not getting any log entries from the M2 (Generic Matter Bridge driver) on the original initialize but I did on the second after the devices had been discovered by the Matter Advanced Bridge driver. It is almost like the HE hub was not looking for the device fingerprints and establishing the connect.

1 Like
2

Upon re-reading your description, it seems to me that the Generic Matter Bridge driver attributes and events subscriptions operation is failing. A possible reason could be that adding one more device to your M2 hub has made the subscription list too large. I experienced a similar issue with the community driver...

Could you temporarily remove one or two non-essential Zigbee devices from the Aqara M2 hub, and then reattempt using the Generic Matter Bridge driver again?

3

Based on conversations you and I had in the past I was thinking the same thing that I may have met some type of limit.

When the issue originally occurred the devices that were originally connected continued to work it was just the one that I added to the M2 hub last that did not. Then when I took the M2 hub out and brought the devices back is when none of them worked.

Yes, I can trial this to see if this will work but it will be this weekend before I have the opportunity. To trial it I will also have to take the M2 hub out of HE which means when I bring it back in via matter I will have to rename all the devices and update all the rules with the new devices. Having to actuate them all to be able to determine which one is which is makes all this a two-to-three-hour process so in an effort to keep the WAF up I will wait to the weekend. :laughing: I will be changing the M2 to an M3 hub this weekend and will need to do those activities anyway. I will do it all at once and report back to the group my findings.

1 Like
4

It doesn't look like it was a drive count issue. I deleted the M2 from my beta hub and brought it back in. All the devices worked just fine using the Genericatter Bridge driver.

Today I swapped out the M2 hub with a M3 hub. At first I had some confusion thinking it wasn't working correctly on my production hub again but I determined that may have beeny confusion. So I removed the M3 hub and brought it back into HE again. Every device worked fine using the Generic Matter Bridge driver and I actually have more devices now than when I did when I first had the problem. Reason I switched to the M3.

The only big change to my production hub with the issue was switching to 2.4.4.xxx. something must have been a little sideways in my production hub and the update install, or something must have brought it back in line. That is the only thing I can think of because a reboot didn't fix the issue last time.