9

I have experienced this twice in the last 48 hours, both occurring around the same time. Are the times correlated or coincidence? Can anyone shed some light on this for me?

Relevant lines from system.log:

Oct 16 12:08:20 redacted_hostname syslogd[59]: ASL Sender Statistics
Oct 16 12:08:21 redacted_hostname Dropbox[605]: [1016/120821:WARNING:dns_config_service_posix.cc(306)] Failed to read DnsConfig.
Oct 16 12:08:24 --- last message repeated 1 time ---
Oct 16 12:08:24 redacted_hostname systemstats[70]: assertion failed: 17A405: systemstats + 665801 [FB138264-8F86-3462-9C53-FC084690A767]: 0x7faaa8a01a80
Oct 16 12:14:49 localhost bootlog[0]: BOOT_TIME 1508174089 0

---

Oct 17 12:19:52 redacted_hostname syslogd[59]: ASL Sender Statistics
Oct 17 12:19:53 redacted_hostname Dropbox[493]: [1017/121953:WARNING:dns_config_service_posix.cc(306)] Failed to read DnsConfig.
Oct 17 12:19:56 redacted_hostname systemstats[70]: assertion failed: 17A405: systemstats + 665801 [FB138264-8F86-3462-9C53-FC084690A767]: 0x7fbb6340e520
Oct 17 12:19:57 redacted_hostname netbiosd[3498]: DEPRECATED USE in libdispatch client: Setting timer interval to 0 requests a 1ns timer, did you mean FOREVER (a one-shot timer)?
Oct 17 13:48:19 localhost bootlog[0]: BOOT_TIME 1508266099 0

Sleep Wake Failure report from 16 Oct:

Date/Time:       2017-10-16 12:13:43 -0500
OS Version:      Mac OS X 10.13 (Build 17A405)
Architecture:    x86_64h
Report Version:  26

Data Source:     Stackshots

Event:           Sleep Wake Failure
Duration:        0.08s
Steps:           7

Hardware model:  MacBookPro11,3
Active cpus:     8

Time Awake Since Boot: 96000s

Process:         kernel_task [0]
Path:            /System/Library/Kernels/kernel
Architecture:    x86_64
Version:         Darwin Kernel Version 17.0.0: Thu Aug 24 21:48:19 PDT 2017; root:xnu-4570.1.46~2/RELEASE_X86_64
Task size:       2766.31 MB
CPU Time:        1.468s

<ALL THE TASK DETAILS>

================================================================
UUID: <REDACTED>
Code: 6c611cd2 00000016
Stackshot reason: Watchdog

Sleep Wake Failure report from 17 Oct:

No samples for specified time interval

================================================================

status: 0x40003
Code: 0x6c611cd204000016
Stackshot reason: Watchdog
  • From an engineering standpoint a wake failure simply could be an arbitrary time it takes to resume. If engineering says - let's collect all slow wakes longer than 5 seconds, this could be nothing at all. Is there a specific thing you are concerned about or something that literally doesn't work each time? My experience is you will get the failure always - dozens and dozens of times reliably if there might be a hardware issue. (even then - it's often just software still) – bmike Jan 3 '18 at 17:23
5
+100

Here's how I would pursue this at work by getting some more power logs and then submitting all to Apple Engineering for analysis.

pmset -g stats
pmset -g log
sudo sysdiagnose Watchdog

To amplify on my comment - watchdog simply watches for stuck tasks so a slow wake from sleep error might be nothing based on a desire to capture statistics on any wake that's not blindingly fast so that the fraction of people that submit error logs to Apple can help them aggregate and measure the overall health of this operation.

If your storage is slower than the best SSD or you have a lot of RAM and your memory is not compressed - I would expect wakes in the 5 to 20 second range for recent computers (last 5 years) and much longer for older Mac hardware.

Unless you have a specific failure, or suspect your hard drive / storage is struggling (run Disk Utility or fsck at single user mode boot) only two occurrences of that wouldn't cause me to do anything other than set a reminder to check in 60 days if the wakes continue to throw "error" diagnostics. I would also check every week for updates and patch the OS in case it's something that gets remedied or tightened up.

Basically, watchdog is designed to monitor for performance, it's clearly gotten a signal or guidance to collect some more details for slow wakes. Other than that - without Apple's source code and a session with their engineers publicly explaining why this specific event was triggered, you may have to monitor this and focus on other aspects like specific apps crashing / failure to wake entirely, etc...

  • I'm responding to your comment and answer here: My wakes are typically 2–3s. There have been a couple of past incidents where my laptop did not enter sleep when closing the lid, so this made me question whether there might possibly be a connected underlying issue causing sleep AND wake problems. However, neither issue has occurred since updating to macOS 10.13.2, so at this point, I think you're right that this was an OS bug. – jsejcksn Jan 3 '18 at 21:00
  • Good news @jsejcksn - There’s a huge amount of new tools, logs that don’t get logged to the filesystem, so your question is awesome and i am glad it got a bounty. It will surely help others when they see that message and someone more clever than I might have a far superior answer still. – bmike Jan 4 '18 at 3:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .