Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Serious Bug in macOS will Crash after 49 days
- Thread starter boppin
- Start date
- Sort by reaction score
Evidence is that this bug is Tahoe only. You are helping to confirm that.
bogdanw
macrumors G3
No, I have shortcuts ⌥F4 - Shut Down...; ⌘F4 - Restart... 🙂
Yes, I am sure my MBA is shut down before closing the lid. I wait for the LED on the USB hub to turn off before closing the lid.
I have yet to see any evidence. All I see is a claim that all Macs are affected which would include Intel Macs that cannot even run Tahoe to begin with.
See my points in post #27 https://forums.macrumors.com/thread...ill-crash-after-49-days.2480701/post-34537921 which includes a debunk of the main claim that this effects macOS from Catalina onwards.
Also read with reference to https://photon.codes/blog/we-found-a-ticking-time-bomb-in-macos-tcp-networking
Or do you have some evidence that this applies to pre-Tahoe?
It is most definitely "all" Macs. The affected code has existed for over a decade at this point, but because people are using Macs in datacenter with crazy long uptimes, it has become noticed.
Well, I guess it's a good thing then that for cluster purposes on Thunderbolt 5 they've bypassed TCP/IP with RDMA since 26.2......but really, good grief. I guess the software engineers at Apple looked at Windows Registry and went "hold our beers."
Can you provide evidence (not just an assertion)?
From a respected commentator https://daringfireball.net/linked/2026/04/09/macos-crash-49-days the code with the bug is new to Tahoe. He also had a Sequoia Mac running for 91 days.
Whoever found this supposed bug should clarify what versions are affected instead of making broad claims that turn out to be incorrect. They managed to write multiple pages but did not mention what OS they were running at the time. I tested it on my Mac and found no such bug and that's all I can do. Unfortunately I cannot upgrade my Mac to Tahoe as it breaks a crucial workflow so I might not be able to revisit this topic until macOS 27 arrives.
This seems to be the original researcher that found the bug. I have linked the part where they suggest this goes back as far as Catalina, if not earlier. It would be worth pulling the repo for the affected code and seeing when the lines were added. This would tell us exactly when this bug was introduced.
photon.codes
We Found a Ticking Time Bomb in macOS TCP Networking - It Detonates After Exactly 49 Days - Photon Blog
Every Mac has a hidden expiration date. After exactly 49 days, 17 hours, 2 minutes, and 47 seconds of continuous uptime, a 32-bit unsigned integer overflow in Apple's XNU kernel freezes the internal TCP timestamp clock. Once frozen, TIME_WAIT connections never expire, ephemeral ports slowly...
photon.codes
Yes, Photon found the bug.
The code is here https://github.com/apple-oss-distri...1650a4c1ff8ca1ea/bsd/netinet/tcp_subr.c#L3732
As John Gruber has said https://daringfireball.net/linked/2026/04/09/macos-crash-49-days the affected lines were checked in 7 months ago. Hence he suggests this is new with Tahoe. And backs this up with a statement that he had a Sequoia Mac running for 91 days. He says:
"I think this bug is new to Tahoe. If you look at Apple’s open-source XNU kernel code — e.g. lines 3,732 to 3,745 in tcp_subr.c — you can see that the lines assigning the time in milliseconds to a uint32_tvariable were checked in just six months ago, whereas most of the file is five years old. Also, I personally ran my MacBook Pro — at the time, running MacOS 15.7.2 Sequoia — up to 91 days of uptime in January."
Photon refer to Apple Discussions that they say indicate this is an old bug (back to Catalina). But none of them seem to relate to this new bug:
https://discussions.apple.com/thread/250867747?sortBy=rank is an isolated problem about SMB connections with recurrance after a few hours (not 49 days).
https://discussions.apple.com/thread/252991075?sortBy=rank is another isolated problem that recurs after a few hours.
They also refer to this which relates to podman https://github.com/containers/podman/issues/12495. I can't see any relationship between this post and the new bug and, anyway, it recurs after a few days.
In conclusion:
a) there is positive evidence (from John Gruber and code inspection) that this is new with Tahoe, and
b) the supposed evidence (from Photon) of this going back to Catalina is without any foundation.
Any evidence contrary to this would be welcome.
Last edited:
Black_Mage
macrumors 6502a
I think it's weird that people bother to turn off their Mac considering how little power they use when idling. Even my 2011 iMac sipped power. The fact that my Studio consumes only 5 watts idling is mind-boggling.