Unable to add monitored repeaters for multiple integration entities

[dotdavid]

I am unable to add multiple integration entities (meshcore devices, one connected via USB and one via TCP) and have more than one of them able to monitor repeaters.

It works fine if there are two integration entries, and if either one of them is the only one monitoring a repeater station. However if both integration entities are monitoring a repeater station it seems that one of them loses all of its monitored entities in the UI except in the "Managed Monitored Devices" radio button option in the main MeshCore integration page that you get by clicking the gear icon next to each integration entity.

Steps to Reproduce

1. Add two integration entities

2. Add a monitored repeater to one of the entities

3. Add a monitored repeater to the other entity. The first monitored repeater disappears from the UI.

4. The "invisible" monitored repeater still shows in the "Managed Monitored Devices" page

I tried looking at the code but I'm a bit baffled. I assume it's something to do with database keys or something and I do get messages saying "Platform meshcore does not generate unique IDs" in the log for the tracked repeaters. But looking at the code I can't really understand why their IDs wouldn't be unique.

Can anyone reproduce this? There's nothing obvious in the logs that I can see that might be related.

[awolden]

There is likely a key collision somewhere. I will have to find another companion node to test it out. Out of curiosity what is the real use case for multiple companion nodes?

[dotdavid]

At the moment the mesh here is quite fragmented. At home I have no-one to talk to, but at work it's basically meshcore ground-zero for the UK (Andy Kirby lives nearby and there's a lot going on there). I've put a raspberry pi and a two seeed Xiao S3 WIO devices at work (one as a client, and one as a repeater) and connected the client to my HomeAssistant via TailScale so I can read messages when I'm away and monitor the remote repeater too.

It generally works really well (though the Python API does occasionally time out accessing data from the remote device, with the resultant random errors in the log as a result) apart from this issue. The timeouts are really just because the API expects local devices I think with less latency and its probably more the Python CLI than the HA integration, but I haven't really looked into that as it doesn't really impact anything apart from the logs.

I think over the next months/years my home area will join the wider mesh or I might get a different job and have to remove all my work stuff, but this is an awesome integration and I really like how easy it makes it to do this kind of thing :-)

[mannkind]

Heh, I just came here to report the same issue.

I’m trying to connect to the second companion through a vpn in order to monitor a very remote repeater.

The connection to both companions works fine. It only seems to be a problem with the monitored device; its entities all become “Entity not found” on my dashboards once that second companion is added

[dotdavid]

@mannkind glad to see it's not just me. I think adding second companions may be fine though - in my experience it's something that seems to happen when you add monitored devices to that second companion.

[mannkind]

Initially I thought so too, but after starting over — thinking I did something wrong — it seems to happen with only one companion having a monitored device.

ScreenRecording_10-10-2025.23-40-46_1.mov