6

The new M1 Macbook Air with latest and greatest Big Sur 11.0.1, 11.1 and now 11.2 has kernel panics with high regularity.

Any suggestions on a remedy, or things to try? Much thanks.

panic(cpu 4 caller 0xfffffe0024233d9c): Sleep transition timed out after 35 seconds while calling Quiesce/Sleep action callbacks. Thread 0x73.
Debugger message: panic
Memory ID: 0xff
OS release type: User
OS version: 20C69
Kernel version: Darwin Kernel Version 20.2.0: Wed Dec  2 20:40:21 PST 2020; root:xnu-7195.60.75~1/RELEASE_ARM64_T8101
Fileset Kernelcache UUID: 3E6AA74DF723BCB886499A5AAB34FA34
Kernel UUID: 48F71DB3-6C91-3E62-9576-3A1DCEF2B536
iBoot version: iBoot-6723.61.3
secure boot?: YES
Paniclog version: 13

an iPhone has been connected, which may be related, when disconnecting it there are fewer kernel panics and the panic message is now slightly different

panic(cpu 2 caller 0xfffffe00274efd9c): Sleep transition timed out after 35 seconds while calling Quiesce/Sleep action callbacks. Thread 0x73.
Debugger message: panic
Memory ID: 0xff
OS release type: User
OS version: 20C69

Running the hardware tools in safemode reports everything as fine--no hardware issues.

jimmont
  • 1,272
  • It seems like there's some device it cannot get to go to sleep. Do you have anything connected to the MacBook Air? – jksoegaard Dec 15 '20 at 13:14
  • It has an iphone plugged in providing cellular networking and the power cable with one of those aftermarket magnetic adapters @jksoegaard (let me know if I should add this to the answer) – jimmont Dec 15 '20 at 13:38
  • Yes, I would add it to the question – jksoegaard Dec 15 '20 at 13:45
  • 1
    Same problem here with M1 Mac Mini and Big Sur 11.2 Main monitor is LG 4K and external monitor is Samsung 4K. panic(cpu 1 caller 0xfffffe001bfc0088): Sleep transition timed out after 35 seconds while calling power state change callbacks. Suspected bundle: com.apple.driver.AppleTypeCPhy. Thread 0xb486a. Debugger message: panic – Rich Feb 07 '21 at 15:20
  • In my case it reliably panics if I have a drive connected through a hub with charging capabilities. Is the phone by chance connected through a hub? – bdecaf Feb 14 '21 at 09:11
  • @Rich: I'm seeing the problem with an Apple Thunderbolt Display. It looks like your panic has something to do with USB-C. In my case, suspected bundle: com.apple.driver.AppleThunderboltNHI – odysseus Feb 14 '21 at 13:26
  • @bdecaf: Yyeses, hubs do seem to be involved. – odysseus Feb 14 '21 at 13:30
  • @jimmont same problem here with MBA M1 and LG 32 4K monitor.
    • If I let the monitor pluggued and on, the monitor will go dark after the mac goes to sleep, but then the monitor wake up for a few seconds every 5-10 minutes and turno off again. Very annoying !
    • If I let the monitor pluggued and put it off, when the MBA goes out of its sleep it indicated it had a kernec panic during the night...
    • If I unplug the monitor, when the MBA goes out of its sleep it indicated it had a kernec panic during the night...

    -I need a last try : unplug, use the mac for few min and then put it to sleep

     – alex Sep 09 '21 at 05:00
  • I had the same problem with same kind of configuration. I made a lot of try and discover that was due to my bluetooth trackpad (old model). If a let it ON, the MBA will crash during the night. I I turn it OFF, the MBA will be OK after the night :)

    Yesterday I updated to macOS Monterey 12.0.1 and it didn't crash during the night despite I didnt turned Off the trackpad I hope the problem is solved !

     – alex Nov 27 '21 at 02:37

5 Answers5

1

I've got the same problem with new M1 Macbook Air macOS 11.2. There's a pattern: LG 4k external display, 2 external drives. After successful ejects, I disconnect devices, close it and reopen 60 seconds later. Voila: the error appears. It hasn't happened unless recently connected to external devices.

V8Tunes
  • 11
  • Great information. External devices - especially thunderbolt 3 or USB 4 have the ability to inject bad data into the kernel and crash a system or induce a panic. – bmike Feb 14 '21 at 13:05
  • Are you seeing this message in the panic log? "Sleep transition timed out after 35 seconds"? That's key to diagnosing this problem. – odysseus Feb 14 '21 at 13:18
  • Yes. Same message. Apple replaced that M1 MacBook Air due to hardware issues (hardware diagnosis repeatedly showed trackpad issues though it seemed to work fine. This new one has been fine but the last two nights it happened again. "panic(cpu 1 caller 0xfffffe002454cc04): Sleep transition timed out after 35 seconds while calling power state change callbacks. Suspected bundle: com.apple.driver.AppleT8103PCIeC. Thread 0x4a18f." – V8Tunes Sep 24 '21 at 04:37
0

I believe I'm getting the same problem/fault.

New iMac Pro with clean install (obviously) of macOS 11 (from today 11.2.1) three times in 60 days I have come back to an unattended (and so, presumably sleep-induced) ungraceful restart with the login (not wake) screen. But - in my view, significantly - 57 times I have left the machine on and come back to everything normal.

Twice I have had a k/p on shutdown - also (I believe) related to some sort of timeout… external device(s)? Not iPhone.

Happy to share logs. But don't know my way around (system?) logs well enough to know which, where and what to look for.

kextstat reveals only Apple kexts installed.

The only other thing I do think might be relevant is the fact that my Time Machine icon in the Finder sidebar often spins for several hours (!) after a TM backup - and the logs show multiple lines like these:

Feb  8 16:51:44 iMac-Pro com.apple.xpc.launchd[1] (com.apple.mdworker.shared.05000000-0500-0000-0000-000000000000[20917]): Service exited due to SIGKILL | sent by mds[115]
Feb  8 16:51:59 iMac-Pro com.apple.xpc.launchd[1] (com.apple.mdworker.shared.02000000-0400-0000-0000-000000000000[20924]): Service exited due to SIGKILL | sent by mds[115]
Feb  8 16:52:56 iMac-Pro com.apple.xpc.launchd[1] (com.apple.mdworker.shared.02000000-0500-0000-0000-000000000000[20941]): Service exited due to SIGKILL | sent by mds[115]
Feb  8 16:53:32 iMac-Pro com.apple.xpc.launchd[1] (com.apple.mdworker.shared.0A000000-0300-0000-0000-000000000000[20902]): Service exited due to SIGKILL | sent by mds[115]

often by the hundred; which make me think more than ever of failed time outs.

TIA for any guidance :-)

MarkSealey
  • 9
0

I want to stress that this is a thread for panics on M1 devices with a very specific error: "Sleep transition timed out after 35 seconds."

@jksoegaard: disconnecting devices is an interesting proposed solution, but it begs the following question:

Why should a response delay in an external device cause a kernel panic?

More importantly, in my case, I'm fairly sure that the device involved is a Thunderbolt Display, so it isn't an option just to disconnect it.

odysseus
  • 975
  • When a panic is induced by a device injecting bad data, you disconnecting while troubleshooting, but then you have to work with the vendor to fix the underlying issue. Is it a cable, a controller board on the display, or something Apple patches when their engineers analyze the sysdiagnose logs. – bmike Feb 14 '21 at 13:27
  • In the case of the Thunderbolt Display, I can't really disconnect while troubleshooting, since it's the display that causes the issue. On the other hand, I've never seen the panic occur when I'm using the MBP by itself; it always occurs when I connect the MBP to the TB display and try to wake up the MBP. And I do have an open case with Apple and have submitted sysdiagnose logs but couldn't get the panic to occur while on the phone ). But I’m fed up with this M1 (no problems with a 2018 MBP 15" in the exact same configuration). – odysseus Feb 14 '21 at 13:32
  • I think I linked to how to set up kernel debugging on this site. If so, I’ll update my other answer. You can debug this as a developer and pinpoint the cause. Do you have support from Apple to send the sysdiagnose logs after you make it crash? – bmike Feb 14 '21 at 13:33
  • Please include a link on how to set up kernel debugging. However, the Thunderbolt Display is a Thunderbolt 1 device (connected via Thunderbolt 2/3 to 1 adapter). – odysseus Feb 14 '21 at 13:50
  • I’m on a phone so it’s hard to do research. I will once I get back home. This involves the kernel debug kit (KDK)and there’s nice write up how to set things up from Apple. – bmike Feb 14 '21 at 13:59
  • @odysseus This should not have been an answer, but rather a comment to my answer. I don't get pinged this way, and it is a chance that I noticed this question. I haven't said that a response delay should cause a kernel panic - ofcourse it shouldn't. I just said that you can probably avoid the kernel panic that way. You can report the bug to Apple, and there's a chance that it is fixed in one of the next updates. – jksoegaard Feb 14 '21 at 14:32
  • It would be a good idea to update the question with the full kernel panic, so that we have more information to work with. Also update the question with the info about the Thunderbolt Display. As always it is a good idea to unplug everything and find out whether the kernel panics stop or they keep being there. That's just a matter of debugging, it does not mean that I'm recommending that you never plug those devices in again. By doing this we can hopefully try to pinpoint that the panics occur when a specific device is plugged in. Then you can write a good bug report for Apple for example. – jksoegaard Feb 14 '21 at 14:34
  • @jksoegaard You’re right. For some reason, I couldn’t post a comment to your answer. And yes, I don’t get the kernel panics if I don’t connect the Thunderbolt Display to my M1 MBP. – odysseus Feb 14 '21 at 17:35
0

I have just had the same issue with a HP Zbook Thunderbolt3 adaptor with power pass through (ie one TB connection to MBA M1). Connected to this I have:

  • an Anker USB-C adaptor that is connected to an HDMI display and microSD card
  • a Seagate HDD for Timemachine connected via USB

The ZBook HP TB3 adaptor is great in that it was very cheap secondhand, works with M1, despite being 'unrecognised' (didn't work with Intel MBPro). However I have an issue where after going to sleep the screens won't turn on unless the TB3 adaptor is unpowered OR the TB3 disconnected from the MBA (and disks not properly ejected in both cases). Caps lock key will show light when pressed when this happens but just has black screen. However just now this didn't work and I had to reboot and I got the same 35s panic error with Suspected bundle: com.apple.driver.AppleTypeCPhy. I think the solution may therefore to be to eject the disks on sleep (or not to go to sleep obvs) but we'll see. Hope this may give some insight to someone!

baronmax
  • 1
-2

The panic message indicates that it cannot fully transition to sleep mode within 35 seconds - meaning that the operating system is waiting for something (a device).

In your case, you indicate in the comments that the only thing connected to the Mac is an iPhone.

Unplug the iPhone from the Mac, and this should solve the sleep mode problem.

jksoegaard
  • 77,783
  • this helps somewhat, there appear to remain various sleep related kernel-panic causing issues with the new hardware+software, I'll continue to attempt to identify what happens and if the message and circumstances are consistent--detaching the iphone changed and seems to have improved the situation – jimmont Dec 17 '20 at 09:03
  • If you have kernel panics without the iPhone attached, please add the message to your question. That would indicate a hardware problem with the M1 - most probably needs to be sent back to Apple for logic board replacement. – jksoegaard Dec 17 '20 at 09:18
  • 1
    Check your pmset log too. You can do this by entering pmset -g log in terminal. This log may show what’s holding up sleep. Also check pmset -g assertionsLog – AVelj Dec 17 '20 at 11:10
  • thanks @AVelj looking also at https://en.wikipedia.org/wiki/Pmset what in particular should I look at to understand and also share here relevant to the problem? I can google something or follow links and edit the question with more related details. – jimmont Dec 18 '20 at 10:26
  • 1
    @jksoegaard I don't think returning the hardware as relevant because: these sleep + panic issues have consistently been problems for both intel and now apple-silicon, and across major and minor operating system versions, if something else indicates a hardware issue then hardware diagnostics (and other related indicators) will show this – jimmont Dec 18 '20 at 10:28
  • Panic issues at sleep is not a consistent problem for macOS on either Intel or Apple Silicon. It is not something that is ordinary and to be expected - it is the indication of a major problem with drivers or hardware. If you haven't got any external hardware connected at all, and haven't loaded any third party kernel extensions, then the problem is either with the Apple drivers or the hardware. As it is unlikely that you should have Apple driver problems across multiple versions that noone else is seeing - it is likely a hardware issue. – jksoegaard Dec 18 '20 at 10:45
  • And no, you cannot expect hardware diagnostics to come up with something for an error such as this. So you cannot rule out hardware issues on that account. In any case, Apple Support can help you with their software if the problem isn't caused by hardware. – jksoegaard Dec 18 '20 at 10:46
  • @jimmont have a look at the manual page for pmset. To be honest I don't know all the technical jargon myself, but pmset logs tells you everything about power (like shutdown, reboot, battery charging, sleep, etc). So if you can see what happens immediately after you attempt to put the mac to sleep, you can see what's holding up sleep – AVelj Dec 18 '20 at 11:18
  • @jimmont, also as jksoegarard said, hardware diagnostic tests may not pick this up (either software or hardware). In my instance, the hardware diagnostic test didn't pick up a failed logic board (the most important component of the Mac!), but a surefire way to know if it is hardware related is to completely wipe the drive and install the operating system from scratch. This way, if you get kernel panics immediately after reinstalling the system, you can rule out software issues. – AVelj Dec 18 '20 at 11:22
  • 1
    I just observed the exact same type of kernel panic on an M1 MacBook Pro with a CalDigit Thunderbolt 3 Dock attached. I just got this MBP, but I used the Thunderbolt 3 Dock with a 2018 MacBook Pro for a few weeks without incident under 10.15.7, and for a few days with Big Sur 11.1. – odysseus Dec 19 '20 at 03:33
  • 1
    This is not an isolated issue with one person's hardware. There are a number of people, including me on an Intel 2020 MacBook Pro 16", that continue to experience kernel panics in Big Sur. I've gone the the full procedure several times, having most recently erased my drive and re-installed Big Sur for the THIRD time, and then reverted back to Catalina after a kernel panic while working the next morning. I'm a software developer and 25+ year Mac user and I have never had such difficulty with a new Mac OS. I'll wait for a few .x versions of Big Sur or fix of the problem before trying again. – Les Brown Dec 20 '20 at 15:20
  • I think you’re mixing two different problems together. I’m talking about one specific kernel panic in one specific situation. There are many, many different types of kernel panics - so you cannot just generalize them. – jksoegaard Dec 20 '20 at 19:27