I have about 70 Thread/Matter devices supported by two Apple TV 4Ks. If I reboot HA, all 70 Thread/Matter devices come online in roughly 10 minutes. However, when I do staggered tvOS updates (e.g. to tvOS 18.6 RC), HA is VERY slow to get all 70 devices back online. The stagger between tvOS updates/reboot is multiple hours. About 25 minutes after the second ATV rebooted I still have 9 offline devices in HA. I thought with dual OTBRs and IPv6 routing, that as long as one ATV was continuously up that rebooting the other should be fairly seamless. And worst case, I'd hope for a full recover in 10 minutes or less. Is there maybe something going on with stale routes or other issues causing such an extended outage? Or is this behavior normal for Thread 1.3 and the HA matter controller?
#Expected OTBR (ATV 4K) failover time with HA? Extended device recovery times.
plush wren
tight schooner
I have a similar setup. recently removed the HA otbr, because it caused way more issues than it helped. The thread network and access from both Apple Home and HA has been more robust, consistent and fast without my ZBT-1 and HA OTBR running.
Your issue may also be affected by this, but apples slow primary home hub election process once the current Master/primary reboots is another part of why this is happening. This process also speeds up when I took the ZBT-1 out of the mix, but still is not fast.