Sleep Wake Failure with High Sierra 10.13.2

Dear IdrisSeabright,

Thank you for your response. I fully appreciate the proprietary aspects of this forum and I try to respect rules, terms and conditions, use agreements and all. If you (and anyone else, who has the power to decide the fate of this comment) could kindly bear with me for a few more paragraphs, I shall do my best to present my position concisely and in the spirit of this discussion.

Firstly, few people will be surprised, if I say that I have not read the full terms and conditions or the use agreement. That is because many honest people will recognize how just about anyone nowadays (guiltily or not) has answered numerous times 'yes' to the question 'have you read and agreed with the terms and conditions' in order to proceed to the use of or involvement with some conditional offering. In some cases this is critical, in most this is just a way to move forward to more useful and meaningful activities.

I may not have read the TC but I can assure you that I have read more than 95pc of the entries to this and a parallel discussion on the same subject, available at https://discussions.apple.com/thread/8135104?start=105&tstart=0

While we all may enjoy friendly and relaxing discussions or some suggestions about useful tips and simple solutions as available throughout the numerous discussions on this site, the two threads I am referring to, happen to address a significant (critical it seems for some) performance issue with the HS OS. Although going through the usual customer support channels would be the recommended course of action, it doesn’t take long, reading through the threads, to see that not a single person has mentioned any success from going in this direction. Apple customer support staff have not managed any substantial help or resolution to an apparently persistent and impeding problem. Where do we go from here?

Just like with many community wide problems, people take every possible opportunity to press for a resolution, especially to problems, which are not a matter of convenience or ornamentation, but seriously affect their work and even cause damages. Well, the Apple Support Communities is one such direction.

It is unfortunate that we will never see a response from Apple here, but even with a minimal involvement, they may wish to realise that occasionally it might be beneficial to all concerned to go out of one’s way if this will help significantly a great number of people, especially if there is a possibility to prevent exasperating matters exacerbate.

At this stage I see two logical possibilities. Either Apple knows this is not a macOS High Sierra problem, but would maintain its involvement so minimal that would not do as little as to nudge a higher enough level member into even suggesting it, consequently leaving hundreds of users wondering how come their utterly diverse circumstances cause apparently identical malfunction. Or, Apple knows it is a macOS HS issue and would not acknowledge it, which is a lot more serious matter.

Whichever the case may be, hundreds of Apple users, at least those showing on this and the thread I mentioned above, quite likely many more, have been affected over an extended period of time and this forum becomes as good as any for them to express their expectations from Apple to help resolve the issue.

Here is the daily recurring reason for my plea:

Even with Power Nap turned on, the problem continues. Much less frequent than with it off. Roughly 105 hours since the last spontaneous restart.

Date/Time: 2018-02-27 22:13:14 -0600

OS Version: Mac OS X 10.13.3 (Build 17D102)

Architecture: x86_64h

Report Version: 26

Data Source: Stackshots

Event: Sleep Wake Failure

Duration: 0.10s

Steps: 9

Hardware model: iMac14,2

Active cpus: 4

Time Awake Since Boot: 380000s

Nope, not resolved. I have two 21.5 iMacs on a home network, both running Beta 5, one has the sleep wake problem consistently, the other does not. Both machines have very similar set-ups but they are not identical. One is late 2014 - this has the sleep wake problem, the other is late 2015 and does not have the problem. So far I haven't found whatever difference is causing the issue. None of the "fixes" have helped for me. Given that 1) many people have the problem but many people don't, 2) fixes seem to work for some but not all, 3) machines differ, and 4) installed software must differ a lot across the board, 5) problems started for many people with a change of OS ( the introduction of High Sierra), I believe the issue is probably far more complicated than we think and therefore there is no miracle cure. I'm sure Apple are working on it, and we'll get a result one day!

After updating to 10.13.4 yesterday, I decided to try disabling Power Nap again so that my iMac would actually sleep through the night. When I woke this morning, it did not respond to pressing the keyboard to wake. I had to briefly press the power switch to wake it. It woke, but with the gray screen where it reloads the contents of RAM back from the hard drive. It seems that sometime overnight, it decided that it was low on "battery power" and needed to save RAM contents. This on an iMac always connected to AC power. Sigh.

Hi, Bluetooth made no difference to either of my iMacs in previous tests, whether it was on or off had no effect. However I'm running 10.13.5 beta1 since 5th April, and I've not had any sleep wake issues since. I'm not drawing any conclusions - just making and reporting observations!

Unfortunately, there is no sign of resolution for this dreadful SWF problem. The saga goes on at the expense of so many users. Looks like it has other manifestations, too:

Apple's magical quality engineering strikes again: You may want to hold off that macOS High Sierra update... • The Regis…

With no disrespect to the enormity of the issues associated with such unimaginably complex things like modern computers, it is the disregard of the difficulties experienced by so many people over such a long time that adds insult to the injury.

In case this comment goes through, just as an update on my experience - while some situations with the iMac crashing on the dreaded Sleep Wake Failure appear similar, there is no clearly identifiable pattern of running applications or connected devices.

I spend hours talking to Apple Customer Support over this issue, and they are totally clueless. They will always try to get you to reinstall your OS, which I have done twice now, and although the problem disappears for one to two weeks, the problem always comes back. I took it to a Genious Bar, and the "genius" representative suggested the same. My advice is to live with it until Apple catches up to the issue, is less frustrating than dealing with their customer support.

I've recently being having issues with my late 2012 iMac and this sleep wake failure issues. My iMac is on macOS 10.13.4 and this issues has only been occurring over the last month.

My iMac has also been going into Safe Sleep mode as well when in sleep for over 4hrs which from reading an Apple support document this is normal with all Macs made before 2013 and it should have always been going in to Safe Sleep mode since day one but it has never done this, as I say it only started have these issues with sleep/wake in the last month.

I have been in touch with Apple but they were not much help, all they advised me to do is carry out all the basic stuff like resetting, refreshing and starting my iMac in safe mode and then to wipe and install my macOS again. All this has been carried out, twice, with no change.

I keep finding my iMac has had to restart itself and when I look at the fault report it shows me it just says a sleep wake failure.

I'm currently trying the tips in the last post, hoping this may have an effect.

I set hibernatemode to 3, after which my system kept running fine for quite a few days. Then, a new build of the beta arrived, and I started seeing the problem again. Just now, I went to see how things are set: hibernatemode was back to 0! I set it back to 3, and now we wait...

Just found this thread after suffering for 2+ months.

Constant crashes during sleep on both a brand new MacBook Pro with Touch Bar and a brand new iMac.

Both running 10.13.4.

Just saw that 10.13.5 is available. Going to give it a shot now ...

A couple of days ago I updated my iMac to macOS 10.13.5 and today has been the first real day that I have been able to try my Mac since the update. I switched it on this morning at 8am and then put it to sleep and then this afternoon I woke up my iMac and for the first time everything worked without any issues. My iMac still wakes from safe sleep mode but this I can deal with but after it did it's thing, everything all worked OK and I did not get any sleep wake failure error message.

Currently I do not allow any bluetooth device to be used to wake my Mac up from sleep, I will re-actavite this feature soon and try it again. I'm not yet holding my breath that it might be fixed but at the moment it's defiantly looking slightly better.

I'll post again once I have switched on the bluetooth feature.

So much for my desperate hopes the SWF saga would be finally over. Unfortunately, I did not have to wait for iDeMi's reply. After a month with the .5 beta without any problems, the dreaded double-chime of the SWF reboot returned. This is only a quick recap for the sake of continuity.

My hopes were quietly propped up by the lack of the usual daily notifications from the two SWF threads. Since people generally write more about problems than they do about the absence of such, it seemed there is a faint chance the solution might have been found, at last. Looks like it hasn't.

It should be pointed out that there definitely is a difference after the .5beta update in my case, at least. It has been a great relief that for a month my work was free of the daily SWF reboots, which I had mercilessly experienced for months before that. However, as I already mentioned in my first post .5beta update comment, it did not seem very convincing that a complete solution has been found, because the sleep state appeared different and not going to a hibernate-like mode.

For the sake of completeness, it has to be clarified that during all this time I always left the iMac to follow through the power management set period of display and computer sleep (5/10 min). Yesterday for the first time I used the 'apple menu' sleep command, which I used to do in the long distant past, before the SWF plague. It could have been coincidental, but having used the 'sleep' command combined with the passing of the arriving of 'autopoweroff' state, which most certainly invokes hibernation, appears to have caused this horrible SWF again, after otherwise uneventful and so much hopeful month.

Such a disappointment!

Pleased to read that things have started to look promising, Stitchy. Glad that some of my efforts have been of help. Thank you for your kind mention. Your comment is the closest to my understanding of the problem.

We may not be out of the woods yet, as I think that the 10.13.5 version may have only modified the 'sleep' state, so that hibernation now takes place only when autopoweroff is effected (typically after about 3-4 hours). This may only reduce the probability of processes clashing, as there is more chance for most completing, which could have been an issue with an earlier hibernation effect.

All this, of course, is still hypothesising before time or apple confirms the results. We hope.

This is really bizarre. I have not experienced a single SWF for two weeks after updating to 10.13.5. This is in complete contrast to the SWF reboots I had every day for months before that. Extended comment follows ...

Logan, only a small addition to jtonn's comment, the SWF problem discussed in this thread involves arbitrary, unintended and complete system reboots. Assuming this not being the situation in your case, it could be related to my previous comment, suggesting that the 10.13.5 also modifies the pmset parameters default values (possibly in an attempt to address the darn SWF). It is unlikely that people have kept records of what the settings were before their updates, but if someone has this could help corroborate my guesses.

If it turns out that an automatic and unannounced change of a setting is used to avoid rather than eliminate the SWF issue, in some cases it may not work and in others, like Logan's for example, it could even make things worse.

Requiring specific parameter value limitations, in order to avoid underperformance or malfunction, at the least should be officially documented, if not completely removed, as parameter ranges should work for all allowed vaulues for all users without product performance deterioration.