Hi, running into this a lot lately. We are managing BYOD networks. Macs on Monterey are having dropped Zooms. When users send us WiFi logs we see WiFi~Debugs each time they report an issue. Not a Roam fail or lack of beacons, but a PRE/POST set of what appears to be process traces. Does anyone have a source for how to parse the driver logs?
Last post on this thread states that hard coding DNS servers resolved the issue.
We see the following a lot:
Ping WAN 1.005 No 17:13:52.089 Could not ping DNS server (17.254.0.22 / 0.00ms / BE / 68 bytes / i=0.100000s / W=1.600000s / t=1.600000s / c=1)**
17.254.0.22 is not one of the DNS servers we provide. How to validate if the disconnects are caused by a check that fails because it’s hitting that invalid DNS server (IP belongs to Apple, according to ARIN)?
One example. User reported issues at 16:48, 17:11 and 17:13. This is a screenshot of the CoreCaptures during that time:

I cannot tell for certain because there is zero documentation available from Apple. But it looks like there is a pattern:
An initial event: "watchdog@BCMWLAN Chip Trap~Type=4" (dont know what this is code for, maybe a wake event)
Followed a few minutes later by a PRE/POST debug pair. Completes in 2-3 seconds
Followed a 10-12 seconds later by another PRE/POST debug pair. Completes in 28-30 seconds.
Pure speculation here. E.g. OSX is constantly doing a check against the
gateway, DNS servers. One of them fails to respond consistently, a
round robin algo aligns it so that sometimes the same endpoint is hit
enough times to trigger a fail. The result is OSX thinking that while
the link is OK, there is no internet on this connection, so start
searching for another network.
I would validate with a pcap if i had something to go on.