VMs can't boot with Network Mode set to Bridged on Apple M5 Pro machines

IIndySun about 23 hours ago 48 commentsRead Article on github.com

ZH version is available. Content is displayed in original English for accuracy.

⚡ Community Insights

Discussion Sentiment

21% Positive

Analyzed from 665 words in the discussion.

Discussion (48 Comments)Read Original on HackerNews

Aurornis•about 20 hours ago

Classic GitHub issue thread: Someone shows up and tries to explain it’s not a bug and never could have worked. Everyone else points out that it does work on other Mac hardware. Person doubles down and tries to claim that the bug was that it ever worked, and not working is the correct behavior. New Mac OS beta release fixes the bug and makes it work. Nothing related to the repo was changed.

burnte•about 15 hours ago

I loved the last comment where he inferred no one knew the difference between a MAC address and a Mac computer, meanwhile he's fundamentally misunderstanding how networking works and making bizarre statements about wifi that literally everyone disagreed with him on. I bet he's still insisting he was right even though it's all fixed now because he doesn't get how VM network bridging works.

fouc•about 18 hours ago

Also classic, someone finally jumping in with a full and insightful explanation. What a turnaround!

montjoy•about 20 hours ago

This is actually what’s interesting

subarctic•about 17 hours ago

So, are they gonna add official support for this in the wifi standard someday? It sounds like the way that this works is a non-standard workaround

aurelian15•about 17 hours ago

You might be looking for https://en.wikipedia.org/wiki/Wireless_distribution_system

This is a standard way to form bridges in WiFi networks and does not require any ARP proxying. In practise this requires certain OS/driver/firmware/chipset combinations to work though.

eqvinox•about 12 hours ago

WDS is not something a client/STA can ask an AP for, it's a different mode of operation. What you need is 802.11ak GLK (General Link). Unfortunately, that one also isn't really negotiable and never gained much traction.

altairprime•about 13 hours ago

Buried in-thread, both macOS 26.6 and 27 beta have fixed the issue — has anyone studied whether they implemented this or resumed the old proxy arp method?

tosti•about 17 hours ago

> the bug was that it ever worked

https://xkcd.com/1172/

doctorpangloss•about 20 hours ago

remember how there used to be people whose lifestyle was answering Stack Overflow questions? 3 years from now we're going to be looking at charts showing engagement from GitHub Lifestyle people going straight down too.

lloydatkinson•about 18 hours ago

Like that probably still open huge github issue with hundreds of comments, the one about how docker will blow your firewall wide open.

naikrovek•about 11 hours ago

Your assessment is inflammatory, to my eyes.

Everything that person said was factually correct. They weren’t wrong, they simply didn’t have the whole picture.

skygazer•about 11 hours ago

But, they were wrong ... in assuming what they were saying was at all relevant. And they were so sure they knew what they were talking about and that it applied that they dismissed all the ample evidence to the contrary and also dismissed the, they assumed, delusional people pointing out the contradictions. Their initial misunderstanding was understandable given the overloaded terms but their confidence was poorly calibrated to their weakness detecting the misunderstanding.

SlightlyLeftPad•about 6 hours ago

I’ve seen younger engineers (still very senior and capable but with quick tempers) make this mistake time and time again.

The funny part, to me, is that it all could have been avoided by simply saying something to the effect of “the APIs they call in UTM are Apple’s and there are no hacks or workarounds they can do at the UTM level. Any differences between the hardware are just that.”

The actual thread was wildly inefficient but if it hadn’t gone that way no one would have learned anything, including us!

billti•about 19 hours ago

I couldn't get bridged networking to work with my M5 machine and VMs running on Parallels. This thread here would seem to indicate it's a known issue (with no updates or resolution :/ ) - https://forum.parallels.com/threads/m5-macbook-air-windows-1...

arjie•about 19 hours ago

The OP has a thing that solves it for many: update to a beta version.

ArchOversight•about 19 hours ago

The linked Github also states that updating to 26.6 works.

juancn•about 19 hours ago

I love when things work out in the end!

(issue is fixed in the latest MacOS 26 and 27 betas)

TheqO•about 20 hours ago

The M5 Pro and M5 Max support Wi-Fi 7 and Bluetooth 6 via Apple's N1 wireless networking chip, while the base M5 supports Wi-Fi 6E and Bluetooth 5.3.

https://en.wikipedia.org/wiki/Apple_M5

broknbottle•about 11 hours ago

Without looking too deep, I bet their dext wasn’t exposing necessary configuration. I’ve seen this with other third party network hardware and dext, where it was missing certain configuration for bridging to work

whatever1•about 10 hours ago

M5 networking stack is a mess. A lot of corporate deployments had to wait for Apple to patch up things

mcfdoesdev•about 21 hours ago

That's super weird.

kazinator•about 21 hours ago

Remember, if shoes were like software, they would explode while you're trying to tie the laces, because the first of the month landed on a Tuesday.

pjmlp•about 20 hours ago

Unfortunately most consumers put up with it, instead of handling it like shoes, returning them to the shop for a refund, or if surviving the explosion, a lawsuit.

kazinator•about 20 hours ago

Shoes today are garbage that wears through within half a year; consumers are accepting it and coming back for more.

phoghed•about 20 hours ago

Analogy kind of falls apart when you consumers are willing to put up with quite a lot to save a buck. They happily build their own furniture for example.