Solved CLOCK_WATCHDOG_TIMEOUT errors on ASUS PRIME x570-Pro

hi all,

i'm trying to install a new machine, have been for over a week now, due to this error. at random times the machine will freeze up. 9 times out of 10 i also get a BSOD -- CLOCK_WATCHDOG_TIMEOUT error. in addition, when i was installing drivers from the included cd, the install program would continually say the drivers weren't installed, even after installing them a couple of times over.

here's what i've tried so far:
- update bios to latest [v.3402]
- download fresh mobo drivers from asus internet site and update all
- swap out video card [was AMD SAPPHIRE HD-7750, now ASUS ENGT520SL]
- run onboard disk test of NVME drive [no errors]
- run windows memory test, complete suite, 10 full passes [let it run all day long -- no errors]
- run sfc /scannow [no errors]

NONE of this has helped at all.
i'm getting pretty frustrated here. i've never had this many problems w/any mobo, and i've put together systems based on intel, asrock, gigabyte, msi... first time w/asus, though. have to say i'm not impressed, despite their good rep...

i'd like to emphasize that i'm using the standard factory settings and haven't tweaked anything ; no overclocking, no performance 'improvements'.

w/all of these problems, is it safe to conclude i've got a bad board, or is there something else i should try ? it would really suck if i have to return this board and start over.

my config --
- CPU : Ryzen 9 3900X
- RAM : Hynix CJR Extreme Performance DDR4 3600 CL18
- GPU : ASUS ENGT520SL
- SATA: Saumsung 850 EVO
- NVMe: Sabrent Rocket 4.0 1Tb
- Peri: TI IEEE-1394 PCIe 1x

O/S: Windows 10 Enterprise x64


many thanks for any help.

I built a similar system (using different components from different manufacturers) in Summer 2020 (see my system specs, or the CPU-Z banner in my signature), and I can say this:

1. Ignore the drivers on the CD, download them from the manufacturers' websites directly for the latest versions.

If you need to, do this on a separate computer.

2. Also, there may be a bit (or a lot) of setup required in the BIOS / UEFI menu prior to getting all the components working. For example, out of box, the 570 doesn't support 3600 MHz RAM unless you enable it by XMP profiles, or else if you assign manual timings / settings for your RAM (and CPU / related settings, like multipliers) via the overclocking menu. In particular, this needs to be done before you actually install Windows, although it can be done after as well, usually with no effects to the Windows installation. But stranger things have happened, so I suggest that you get your BIOS settings dialed in and then cleanly install Windows again to be sure.

But seriously, the big thing to look at is the most up to date drivers from the manufacturers themselves (so from ASUS for your motherboard and GPU, etc.) and use those on a completely clean install of Windows on your system - do not use the CD drivers at all. (You may find that the online versions are the same as the ones on the CD - still, use the online versions, for posterity's sake).

Also, for further reference, a few more details would help:

1) I'm assuming you're installing Windows to the NVMe drive;
2) I'm assuming that you've updated the BIOS / UEFI on the motherboard to the latest from the website already;
3) I'm assuming that you're installing Enterprise for a specific reason, and that it is the latest stable release (or else a LTS release).

If adjusting the settings in BIOS for XMP as well as a completely clean install still trips the BSOD, then it may be worth looking into the BSOD forum to have it analyzed directly for a better clue as to what is actually tripping on your computer.

thanks for the reply, @johngalt.

yes, i d/l'd and installed the latest bios, also the latest drivers from the web site. and yes, you're right, the ones from the web site were actually much more recent [like 6 months or so].

re- the 3600mhz ram, for the moment i'll be happy if it runs even with lower performance, as long as it runs. i was told that the ram would be under-utilized without performance tweaks, but that the system should still run fine, and that i could fiddle around with the timings in bios if i wanted to get some extra performance. is that incorrect?

thanks for the tip on the bsod forum, i just noticed that. i'll follow your suggestion and post there too.

It shouldn't be - it should default to something like 2133 / 2166 MHz without the XMP profile being enabled, pr perhaps even 2400 MHz - but it should be perfectly usable no matter what, just not at optimal usage due to the lack of native support for anything >3200 MHz by most X570 boards. But, at least in my case, it was a simple matter of going into the BIOS and enabling the XMP profile - nothing else - no other type of fiddling necessary.

But since you've covered all of the bases, yeah, BSOD is a good place to go to have those gurus really get down to the nitty-gritty and figure out where this issue is stemming from.

GL!

hi all,

posting here following the suggestion of one of the forum members when i posted this problem under the drivers and hardware section. hope i'm not making an error for double-posting.

anyway, i'm trying to install a new machine, have been for over a week now, due to motherboard problems. at random times the machine will freeze up. 9 times out of 10 i also get a BSOD -- CLOCK_WATCHDOG_TIMEOUT error. i'd like to emphasize that i'm using the standard factory settings and haven't tweaked anything ; no overclocking, no performance 'improvements'.

in addition, when i was installing drivers from the included cd, the install program would continually say the drivers weren't installed, even after installing them a couple of times over.

here's what i've tried so far:
- update bios to latest [now v.3402]
- download latest mobo drivers from internet and install all
- swap out video card [was AMD SAPPHIRE HD-7750, now ASUS ENGT520SL]
- run onboard disk test of NVME drive. no errors.
- run windows memory test, complete suite, 10 full passes. let it run all day long. no errors.
- run sfc /scannow. no errors.

NONE of this has helped at all.

i'd be really grateful is someone could help me troubleshoot this or at least determine if it's a bad board.

my config --
- CPU : Ryzen 9 3900X
- RAM : Hynix CJR Extreme Performance DDR4 3600 CL18
- GPU : ASUS ENGT520SL
- SATA: Saumsung 850 EVO
- NVMe: Sabrent Rocket 4.0 1Tb
- Peri: TI IEEE-1394 PCIe 1x

i couldn't figure out where to find the "V2 Log Collector output" mentioned in the posting guidelines, but i did find the event viewer, and the following entries about the latest crash :




many thanks for any help.

The V2 log collector is in the resources dropdown at the top of the page.

Download it, extract the files and run the shortcut link. Do not attempt to run the file before extracting everything first.

The output should be a zip file on the desktop. Just attach it to your next post.

One more think you can do to help me is to provide a link to the memory.dmp file which you can find in your C:\Windows folder. Copy this file to somewhere like OneDrive or Dropbox and share a link to it in your next post. I ask you to do this because the Clock_watchdog _error is a type of BSOD crash which requires the full memory dump to be analysed. Most times it can't be done with the mini-dump file which the log collector finds.

wow, thanks for the speedy reply, @philc43 ! and thanks so much for offering to help.

okay, i've run the v2 log collector and uploaded the output to sendgb. you can d/l it here.
i've also uploaded the memory.dmp, you can d/l it here.
as it so happens, the machine crashed again right after i ran the log collector, so the memory.dmp file overwrote the one that existed at the time i ran the log collector, but once again it was exactly the same problem. please let me know if you need me to run everything again anyway.

thanks again for the help.

I've been looking at the memory.dmp file and it shows that your system crashes because it is waiting for a response from a processor that is hung. The hung processor is number 11 (Hex: b).

CLOCK_WATCHDOG_TIMEOUT (101)
An expected clock interrupt was not received on a secondary processor in an
MP system within the allocated interval. This indicates that the specified
processor is hung and not processing interrupts.
Arguments:
Arg1: 0000000000000008, Clock interrupt time out interval in nominal clock ticks.
Arg2: 0000000000000000, 0.
Arg3: ffffd00059283180, The PRCB address of the hung processor.
Arg4: 000000000000000b, The index of the hung processor.

FAILURE_BUCKET_ID: CLOCK_WATCHDOG_TIMEOUT_IDLE_THREAD_nt!KxWaitForSpinLockAndAcquire



Details of what is happening on Processor 11 is shown below:
11 TID:000c kb kbn kbnL kn knL kpn kPn
# Child-SP RetAddr Call Site
00 ffffd000`592ad730 fffff800`2f30da0b nt!KxWaitForSpinLockAndAcquire+0x20 <--- where the timeout occurred
01 ffffd000`592ad760 fffff800`2f30d54c nt!EtwpDequeueFreeBuffer+0x2ab
02 ffffd000`592ad7c0 fffff800`2f2dcf10 nt!EtwpSwitchBuffer+0x2c
03 ffffd000`592ad820 fffff800`2f2dc908 nt!EtwpLogKernelEvent+0x540
04 ffffd000`592ad8c0 fffff800`2f2dbf10 nt!KiExecuteAllDpcs+0x348
05 ffffd000`592ada10 fffff800`2f3dc97a nt!KiRetireDpcList+0xe0
06 ffffd000`592adc60 00000000`00000000 nt!KiIdleLoop+0x5a


The DPC's for processor 11:

--------------------------------------------------
CPU#11
--------------------------------------------------
Current DPC: Unable to load image \SystemRoot\system32\DRIVERS\nvlddmkm.sys, Win32 error 0n2
nvlddmkm+0x12A950 (Normal DPC)
Debugger Saved IRQL: 0
Cumulative DPC Time Limit: 120.000 seconds
Current Cumulative DPC Time: 0.000 seconds
Single DPC Time Limit: 20.000 seconds
Current Single DPC Time: 0.000 seconds



This information seems to suggest that the graphics driver is the one holding up the processor and causing the bugcheck.

Your graphics driver is dated from 2012 so it is really out of date!!

Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys
Timestamp: Thu Aug 30 16:50:16 2012 (503F8BB8)

Please update this driver and see if the problems go away.

unfortunately that's the latest driver available, i'm afraid.

that being the case, i took out the ASUS gpu and used a cleaner from a company called 'wagnardsoft' to get rid of the drivers, then re-installed the first card i'd put in, a sapphire ultimate hd-7750. even though i was getting these same CLOCK_WATCHDOG_TIMEOUT errors with it before, i'd managed to find some drivers that were just a bit more recent, so i thought that might do it.

unfortunately the drivers for this one just stall at the very first step, as it tries to gather info to see if the system is 'compatible' [so... my guess : not... ;-) ].

nonetheless, i closed the install process, then i searched through the install tree until i found the drivers for the card, then went into the control panel to update drivers manually, navigating to the branch of the install tree i'd found. windows found the drivers and began installing them, but the machine froze up during the install. [really not having much joy w/this machine, lol...]

so you're pretty sure this is a conflict w/the gpu then ? and not that the cpu itself is bad ? if so, i guess i'll go buy a new gpu.

thanks again for all your help on this.

Let me look through the other mini-dump files and see if there is anything I can pick-up from them. Certainly the CPU would be the next item to check and it is a common cause of 0x101 bugchecks.

I've not been able to get any further details from the mini-dumps. If you get more crashes and can save the memory.dmp files as you did before I can look at those for you.

If you can get hold of another CPU for swap testing I would do that before purchasing another graphics card.

well i think i've got this one sorted. it turns out it wasn't a single issue, there were several factors conspiring against me...

went by my local computer shop this morning and bought a more recent video card, but windows still froze up, even w/the latest drivers for that card. the only real possibility by this time was windows itself. i'd tried to update, but the update service was never allowed to finish because of windows freezing up [frustrating cuz i'd noticed one of the updates was a large videocard patch...]. wsus didn't work either [it finished but didn't fix the issue], so i eventually d/l'd a completely recent image from microsoft using the media creation tool and did a completely fresh install. lo and behold, it works. [er... well, touch wood; it's been going w/o a freeze for a few hours now.] funny how up-to-date hardware, drivers, and o/s working together can make or break your Windows Experience®.

thanks so much for your analysis, @philc43, i reckon i never would've thought to go this route if you hadn't pointed a finger at the graphics card. if there were stars on this forum you'd get five from me.

That's great news. I did see that you had been having problems with the updates so that all fits. All credit to you though for going for the fresh install of Windows.

If you click the "Like" button at the bottom left of my analysis post you can select Great Support or Helpful to give your feedback.

Glad you got it working. And, somehow, I'm not surprised at it being the OS. Not surprised at all.