Since installing Catalina yesterday, multiple crashes from userspace watchdog timeout

I'm waiting till my Holidays in 6 weeks time to purge all the bad thoughts of Cataline. DUe to the disasters i stopped upgarding at 10.15.1 because i had external monitor issues. Then I watched the problems with all subsequent upgrades and decided in times of covid there is no way to trust apple. In 6 weeks i get enough spare time to try and return to mojave away from the mess. This will be my last MAC due to these problems and the fact this no longer 'just works' My MAC was up for update but I just got a great dell XPS 13 in the sales. The MAC unfortunately just can not be trusted to be a daily driver any more.

In honor of the 1st anniversary of this thread, I figured I'd post my 2 cents.

I have a MacPro (Late 2013). In the last month Apple completely replaced both GPUs, the Logic Board, the CPU Riser with the CPU and the central heatsink. So yeah, practically not even the same computer. I'm on 10.15.7 now and this problem that everyone has been having is absolutely persistent. Sometimes I'll go a full day without a crash, sometimes I'll go 5 minutes and that's it. Last night I just couldn't take it anymore, so I did a complete wipe and clean install of Catalina. 20 minutes after I still hadn't installed a single application. I opened up Safari and started downloading 2 application to install (still not installed), and then it happened again.

I am, however, using dual monitors. It's next to impossible for me to use just 1 now.

I've tried all of the basic resets, etc. and nothing. It's definitely not a third party software problem. Not sure why VM became such a huge topic in here, but maybe that just exacerbates the issue for some people.

I'm currently trying to use my computer with just one monitor plugged in and it seems to be working for about 2 hours now. We'll see. But, as I said, 1 monitor just won't cut it for me on a regular basis. Here's hoping Big Sur comes out soon and fixes things.

I've also been testing with a single monitor, and can confirm it seems to make no difference. (Mac Pro on Samsung 4k via DP)

Has anyone seen Watchdog killing WindowServer on MacBook or iMac with no external display connected?

I have two work Macbook Pro's (a 2017 with an i7 and a 2019 with an i9). I run a TB3 dock (CalDigit) with dual monitors, keyboard, mouse, etc...

The i7 runs flawlessly on Catalina 10.15.7. When I run Catalina 10.15.7 on the i9 I get the same watchdog issues described in this long thread. Once I go to 10.14.6 on the i9 it runs flawlessly and I see zero issues.

I only get the watchdog issues when I put the laptop to sleep. All software and (external) hardware is identical for the most part. That tells me it is something configuration wise with the internal Mac pro hardware that is causing these issues. I suspect it's a driver issue with the Radeon Pro Vega 20 OR the i9/chipset but I don't have any proof.

Maybe specifically to you that could be the problem, but overall it seems like it's a larger issue than that. As I (along with many others) have the exact same issue as you, but my configuration is completely different.

MacPro (Late 2013) with a Xeon E5 and dual FirePro D300s

This has been running for over a year now with no real identification of the exact cause and no work around or fix available. A number in Mac Rumours have reprted that replacement logic boards have fixed the issue even on brand new machines out of the box. There appears there will never be a fix for this and no solution from Apple. The obvious answer if it happens too frequently is replace the Mac with an alternative. That seems to be the other work around.

I had the first 16inch MBP and it crashed EVERY time the external display came out of sleep. They didi a logic board change on what was essentially a brand new laptop. It did NOT fix the problem. It was unusable for a year! 10.15.7 eventually made the problem go away. I live in Silicon Valley and offered to bring in a complete setup that reliably and predictably failed so they could diagnose the problem. No interest. I doubt my next computer will be a Mac. Computers just aren't an interesting business to them any more.

I have not an external display, but I continue to think the crash depend by Intel GPU driver. I have an Intel Iris 6200.

I honestly hope in Big Sur, otherwise I will scrap it away, and I will buy an windows pc.... :-(

So after 2 whole weeks of my MacPro (Late 2013) being at the Apple Repair Center I finally got it back.

"We were unable to reproduce your issue(s).

Our technicians performed comprehensive testing and could not reproduce the issue(s) you reported. Your product passed every diagnostic and extended test we performed."

So they didn't do squat.

How are they supposed to reproduce the issue if they don't actually do a regular startup of your system? Which I was told by the store that they don't. They only start up with their external tools and safe starts, etc.

Looks like I'm screwed. And, so far, I've heard that Big Sur doesn't fix the problem.

Hey everybody! Just wanted to come back with an update on my situation which I have finally resolved in the hopes that maybe my experience can be of benefit to somebody else. Because these problems seem so varied in their root causes, here's the specifics on my system and what I did prior to finally finding the magic bullet for my system (if you want to skip to the resolution, I've marked it out below):

Late 2012 21" iMac 3.1 GHz core i7

8 GB 1600 MHz DDR3 ram (factory installed)

1 TB fusion drive

Nvidia GeForce GT 650M 512 MB

No external monitors or other devices aside from some drives

It never had any problems of any kind all the way through Mojave. I updated it to Catalina 10.15.2 in December 2019 and kept it up to date without any problems into March 2020. Then with the 19E266 build of 10.15.3 which I installed on March 28, 2020 everything fell apart. I was getting 40-50 kernel panics per day and when it wasn't panicking, it was hanging leaving me to look at a beach ball for 5-10 minutes before then panicking. All the reports showed the watchdogD timer as the cause of the panic but, of course as we all know, that's just a countdown timer and something deeper in the system is the true cause.

I fought with it off and on for a few months because I've never met a Mac I couldn't fix before but it never got any better. Finally I gave up, wiped the system clean and did a full clean install of Mojave 10.14.6. Unfortunately, this did not solve the problems fully. It did reduce my kernel panics to about 3-5 per day but the computer was constantly hanging such that it was completely unusable. I could get maybe 10-15 minutes out of it occasionally if I was really careful not to make it angry. Apple suggested the problem was non-standard RAM but the RAM was factory installed. Then it was external devices but I had long prior disconnected all of them. Then the logic board supposedly was the cause at which point I called BS. I also had Linux and Windows 10 installed on this machine and when I would boot into either one, the computer ran perfectly. I ran a full slate of hardware tests and all the hardware checked out okay. I was ready to give up on MacOS.

Here's the resolution part:

Then one day I decided to beat my head against the brick wall again and booted it into MacOS 10.14.6 and started doing some troubleshooting. It immediately began hanging again but as I was working with it, I noticed an odd pattern--all my hangs had the timing and feel of a page file issue--like the computer was dumping stuff from RAM to a page file every time it accessed the disk. I checked the system monitor and my RAM usage was very low but I did see read/write disk spikes every time there was a hang. I started to suspect there might be a disk issue that my hardware tests just weren't finding. So, I took a new 1TB Samsung QVO SSD I had just bought, put it in my drive dock, ran a clean install of Mojave 10.14.6 onto it, changed my boot drive to the docked SSD, booted it up, and... TA-DA! (as my 4 year old would yell), it was perfect! It booted instantly, ran perfectly, and was faster than it has ever been in its entire existence! It has been over a month now since I did that and in that time I have had 0 panics, 0 hangs, rebooted twice just to avoid thunderstorms. Sleep/wake is fine. All external devices are fine. I've added an external display which is fine. Everything just works--the way a Mac is supposed to be!

I know this is merely one anecdotal report of my particular case, but my suspicion is that apple changed some low-level code or EFI stuff that effectively made MacOS unusable with fusion drives. I know they said Catalina was optimized for SSD storage but I didn't realize that meant nonSSD storage was now dead as far as system drives go. Maybe I'm off base and I just got lucky but if your situation sounds at all similar to mine, I'd try doing a clean install on an external SSD and see if it helps you too.

Good luck to everybody--I know this has been an absolutely maddening issue for all of us.

I too have this issue. Started with Catalina, maybe every 3rd or 4th time but it now crashes every time I wake from a lengthy sleep on Big Sur. During the day normal sleeps are fine, like a lunch hour or something but an overnight sleep always causes the Macbook Pro to crash every time it's woken with my external display attached to it.

This started happening to me today. It was happening while Windows 10 was running on Parallels. I would leave the computer for a while and come back and find the computer restarted. I received this error. I'm running new Mac Pro 2020 with Apple screen and have not had any problems until today.

anic(cpu 2 caller 0xffffff8009953a13): userspace watchdog timeout: no successful checkins from com.apple.WindowServer in 120 seconds

service: com.apple.logd, total successful checkins since load (1570 seconds ago): 158, last successful checkin: 0 seconds ago

service: com.apple.WindowServer, total successful checkins since load (1540 seconds ago): 142, last successful checkin: 120 seconds ago

service: com.apple.remoted, total successful checkins since load (1570 seconds ago): 156, last successful checkin: 0 seconds ago

Backtrace (CPU 2), Frame : Return Address

0xffffffe5e674b670 : 0xffffff80066bc66d

0xffffffe5e674b6c0 : 0xffffff80067ff073

0xffffffe5e674b700 : 0xffffff80067ef6aa

0xffffffe5e674b750 : 0xffffff8006661a2f

0xffffffe5e674b770 : 0xffffff80066bbf0d

0xffffffe5e674b890 : 0xffffff80066bc1f8

0xffffffe5e674b900 : 0xffffff8006ebee84

0xffffffe5e674b970 : 0xffffff8009953a13

0xffffffe5e674b980 : 0xffffff80099536ba

0xffffffe5e674b9a0 : 0xffffff8006e460ee

0xffffffe5e674b9f0 : 0xffffff8009952b0a

0xffffffe5e674bb20 : 0xffffff8006e502bb

0xffffffe5e674bc80 : 0xffffff80067aaa61

0xffffffe5e674bd90 : 0xffffff80066c1d77

0xffffffe5e674be00 : 0xffffff80066985d5

0xffffffe5e674be60 : 0xffffff80066afb82

0xffffffe5e674bef0 : 0xffffff80067d3823

0xffffffe5e674bfa0 : 0xffffff8006662216

   Kernel Extensions in backtrace:

     com.apple.driver.watchdog(1.0)[7948A279-A8B8-3650-AFBF-B1E3EB68942A]@0xffffff8009952000->0xffffff8009953fff

Process name corresponding to current thread: watchdogd

Boot args: chunklist-security-epoch=0 -chunklist-no-rev2-dev chunklist-security-epoch=0 -chunklist-no-rev2-dev

Mac OS version:

20B29

Kernel version:

Darwin Kernel Version 20.1.0: Sat Oct 31 00:07:11 PDT 2020; root:xnu-7195.50.7~2/RELEASE_X86_64

Kernel UUID: 84C6DC45-6B02-335F-9439-5D2A9BC385A4

KernelCache slide: 0x0000000006400000

KernelCache base: 0xffffff8006600000

Kernel slide:   0x0000000006410000

Kernel text base: 0xffffff8006610000

__HIB text base: 0xffffff8006500000

System model name: MacPro7,1 (Mac-27AD2F918AE68F61)

System shutdown begun: NO

Hibernation exit count: 0

Interesting theory. I think you are right about some low-level code or EFI code changes, but my Mac Pro already runs APPLE SSD SM0512G. It ran fine with Mojave. As soon as I upgraded to Catalina (first version) last year, it was hosed and has never been able to operate normally since then. Big Sur has not helped either. I'm sure there are many of us running SSDs that still have the userspace watchdog timeouts. The sad thing about this whole situation is that I have no doubt that Apple knows what is wrong. It might be a strategic decision to just let the apple customers with userspace watchdog issues continue to suffer until we just go away. We only account for less that 10% of Apple's total sales.

For me the problem didn't go away until Big Sur!

But I'm running a Mac Pro (Mid 2010) upgraded with RX 5700 XT graphics. This isn't supported to run Catalina nor Big Sur, (but it works with some tinkering…). Still, In Big Sur I haven't seen any of the Kernel Panics that I had in Catalina when waking from sleep, all the way from version 10.15.2 to 10.15.7.

So, this problem sure seem to have different causes; can seemingly be both hardware and software related.

”Watchdog timeout” kicking in seems like a symptom of the cause.

Sorry for all that still have problems. :(

Now that the new M1 macs have the exact same issue maybe some thing will happen. I've seen 4 people now return there M1s because of the watchdog failure. If this is a hardware problem then Apple carried over to the new architecture. Its pretty clear that after a year and new hardware this is a software issue with Macos.