TL-R605 router disconnected and forgotten by Omada Software
I've been running Omada software (v 4.2.11) in docker on Unraid and recently added the TL-R605 two weeks ago. Because I'm using a different subnet (192.168.5.0), I followed the process outlined here to adopt it into the controller software. The setup worked exactly as described and had been working fine for a week.
Then the R605 seemingly got disconnected from the software and could not be readopted. The DHCP was also no longer working to assign IP addresses. Restarting the device allowed it to show back up in the software but as "Managed by Other" and it would fail to adopt again. I can reached the webUI with the assigned IP of 192.168.5.6 but I'm unable to login due to the software controller changing the ID and password. So I had factory reset the R605 and set it up again with the same process.
And now it seems to have crashed again one week later with the same issue. Looking at the event logs in the controller, it had initially disconnected, but was able to be readopted automatically before I noticed anything amiss, but then become disconnected again. And same issue again with the DHCP not working. So I have restarted the device and it is working okay to route traffic, but it is showing up as "managed by other" and won't let me change any settings. I don't want to factory reset it just yet in case there's some useful logs I could pull from it, plus it seems like this would likely just happen again in one week.
So what is going on with the adoption of the R605 when on a non-default subnet? It seems like I might have to just use it in standalone mode until the next big update. Very frustrating.
I am wondering if it could possibly be a memory issue with the R605, but because I don't currently have the router adopted in, it won't show me the performance statistics from when it was previously adopted. I'm not sure reseting the router to get it adopted back in will show the previous statistics either. Maybe there's a way to see the statistics in some log files.
I have also just noticed that my two EAP245s become periodically disconnected and readopted as well. And this had also been occuring prior to adding the R605, and using my old Edgerouter X, so it seems like the actual issue might be with my controller software. And for some reason when the same has happened to the R605, it isn't always able to be readopted, causing the issue my post is about. Is it to be expected that the EAPs will periodically become disconnected and need to be readopted? It also happens to both at the same time, and has happened 3 times in one day recently. Should the controller docker be setup differently than the host network in Unraid? Perhaps br0 with it's own internal IP address would fix the issue.