Apple just force-upgraded me to Sonoma

TidBITS Talk
62

I had 2 machines force upgrade me. A MacBook Pro and a Mac Pro 2019. In both cases, I don’t recall what I did to trigger it.

I didn’t get any notification on either Mac, but I did get a handful of restarts and progress bars before Sonoma came up.

To roll back my MacBook, I had to format the drive and install Ventura using Internet recovery. Since it was a clone of my desktop I connected it with a Thunderbolt cable and used Migration assistant since that was much faster that using the backup on my NAS.

63

3 posts were split to a new topic: Buttons that appear only when moused over

66

There is one method that will get you time to research. Enter Recovery.

I have been thinking about another solution. I can not recommend this, because I do not know for sure what will happen, but if this happens to me I will try to check if the system installer show up as a disk if I follow this Startup Disk And then if it does, I will choose the disk I have my old system on. If it does not work, I will do a hard shutdown at once. Never if it has already done one of the reboots the installer always does.

67

I found this script that you can use to check OS downloads via Jamf. https://community.jamf.com/t5/jamf-pro/monitor-software-update-download-status/td-p/263400

#!/bin/sh
    
    path_updates=$(ls "/System/Library/AssetsV2/com_apple_MobileAsset_MacSoftwareUpdate" | grep asset)
    if [  "${path_updates}" ]; then
    	path_updates=("/System/Library/AssetsV2/com_apple_MobileAsset_MacSoftwareUpdate"/$path_updates)
    	echo "Update Path is $path_updates"
    	download_version=$(/usr/libexec/PlistBuddy -c "Print :MobileAssetProperties:OSVersion" "${path_updates}/Info.plist")
    	echo "Version Downloaded is $download_version"
    else
    	echo "No updates downloaded"
    fi

I checked and got the following output:

Update Path is /System/Library/AssetsV2/com_apple_MobileAsset_MacSoftwareUpdate/f1a0644b170af1921730ab729dd1cccf6dad96b8.asset
Version Downloaded is 14.2.1

2 Likes
68

One more data point. My wife works from home on a company-supplied MacBook Pro (Intel). Last week she was surprised on startup to find she was being updated from Ventura to Sonoma. Her MBP runs mostly corporate software via a VPN. Updates and other IT support is generally handled by her company’s IT department. She assumed the Sonoma surprise was requested by corporate IT support.

If so, perhaps there are reasons not yet widely known to do so. Perhaps a major security breach has been patched in Sonoma but is not easily fixed in older Systems. This could explain the urgent forced updates, which many of us have agreed to for smaller security patches. It could also explain the lack of discussion of the forced updates, since such discussion would reveal the vulnerability of older Systems.

If not, and the update was as much a surprise to IT support as to the rest of us, this seems to confirm a major bug. My guess is unexpected major modifications of managed corporate software are not welcome by corporate IT.

Unfortunately for this discussion, my wife chooses to remain a frictionless cog in the corporate machinery, invisible outside her assigned duties. So long as her IT works, she will not trouble support by asking about issues such as this, so we do not have the information to distinguish between the two possibilities outlined above.

Perhaps other readers who are currently active in corporate IT management could comment. Even something like “NDAs do not allow any discussion, but all will be revealed in good time” would reassure our faith in Apple. The alternative, “Surprised us too that software on our Mac clients was randomly changed” would confirm for us that pushing unwanted major updates is yet another serious bug in macOS, apparently newly introduced after the announced big push to fix bugs.

69

Just to make sure everyone knows—you can ALWAYS call Apple Support, regardless of how recently you purchased a device or your AppleCare status. It’s free phone support for all users and has been for many years now.

I’d be a little surprised if it was mentioned and not in the slightest bit worried if Apple said nothing about it. Adoption rate of Sonoma has no effect on Apple’s financials, so the only reason Apple would trot out a stat like that would be to distract from less-positive numbers.

70

Because:

I can confirm

Apple is aware and concerned themselves. Last night, I spent ~45" on the phone (a call back at 9 PM) with an Apple Senior Advisor. I told him that my problem was “resolved” but I had no idea how. He said what Engineering really needs is a Feedback report posted while the machine is actually being threatened with the unwanted update, so that Feedback can grab all the logs, and they can get a detailed look at the state of the machine. I had posted a Feedback, but it was after the situation was resolved so my logs were likely not as helpful.

The Senior Advisor was extremely courteous and sympathetic. He emphasized that Apple can’t fix things they don’t know are broken, and encouraged folks to report problems BOTH to Support and through the Feedback Assistant app.

1 Like
71

I had a similar experience yesterday. I’d stuck with Ventura because of worry that some of my apps wouldn’t run under Sonoma. I’ve been having some nagging little problems with my 2019 iMac, so I restarted it yesterday. I figured it would take awhile, so left it to do its thing. I checked on it a couple of times, thinking it was taking longer than usual. When it had finally finished, my OS had been updated to Sonoma. I’ve been a Mac user almost since day one, and this is the first time Apple has done something like this without my approval.

72

Well, I’m glad my iMac is maxed out at MacOS 10.13.6 and my MacBook Pro at 12.7.2. My MacBook Air was maxed out at 10.13.6 but the SSD died so I have to recycle it. I’m going to pull the SSD chips off the motherboard first and smash them to be safe.

1 Like
73

That’s scary. I’ve been holding off upgrading my 2021 MBP M1 because I don’t want to lose my MsgFiler plugin in Apple Mail. I hope that doesn’t happen to me.

74

See Michael Schmitt’s response above which shows a way to to forestall the upgrade:

1 Like
75

I’m running Big Sur on an Intel MacMini and I have not seen this auto-upgrade behavior yet. I’ll keep an eye out for it.

76

That way of downloading the upgrades was so much better. Apple’s default now seems to be that every user is incompetent/computer-illiterate and should just let Apple do the driving. I don’t think that accounts for how screwed up this rogue notification situation is, though.

77

Wow. I mean, just. Wow.

I was complaining recently in another thread about Finder window behavior but that is merely annoying. This is terrifying.

I’ve been using Macs since 1989 and I never thought I would see the day Apple cared so little for their users. If Uncle Steve was alive he would be turning over in his grave.

1 Like
78

I would just like to caution people not to assume that any unexpected behavior is intended by Apple. All indications so far, especially the reported responses of Apple Support to reports of this issue, are that this forced upgrade is a bug and not intended behavior.

I know we all want to think the worst of tech companies, but the fact is that software is incredibly complicated and code is very complex to understand, even when well documented. Bugs can and will happen, and sometimes they will be major, and no reasonable amount of testing will find them all before release, even the big ones.

I do decry the apparent decrease in pre-release bug catching and fixing, and the apparent disinterest in fixing some long-standing bugs (likely because they’re considered too minor to be worth the effort). But that doesn’t translate to Apple “not caring” about users or doing whatever they want because they can. There are definitely areas in which it’s unquestionable that Apple is more interested in the money than in what’s best for users, but those are largely areas in which that behavior was standard during Jobs’ tenure. And Apple doesn’t directly make money from OS upgrades, so the financial incentive to force them is weak at best.

The old axiom of, “Assume stupidity before malice,” continues to apply. In this case, “stupidity” can be expanded to include “bad luck” and “unintended errors”. Just because something bad happens doesn’t mean it was supposed to happen.

3 Likes
79

But does that matter? If you have been force-upgraded and now face the tedious task of trying to clean up the fallout, that task does not become less laborious or annoying just because you subscribe to Apple being stupid rather than malicious. At the end of the day, we will likely never know if they’re more stupid or more malicious. But that’s also entirely irrelevant as long as we don’t know how to prevent force-upgrading or learn of an easy means to reverse it. I don’t believe Apple is evil, but that has tremendously little bearing really.

2 Likes
80

I believe it does.

Yes, fixing things after this kind of an issue is a major pain, to say the least. But whether Apple intended this determines whether Apple is likely to try to find a way to help users who have been affected. If this was what Apple wanted to happen, they have little incentive to help victims and no incentive to keep it from happening to others.

On an individual level, having this happen can be devastating. But any issue that affects one person is almost guaranteed to affect several, and if it’s a bug, the more help we give Apple in reporting it, the better the chances of it happening to fewer people. If we all assume that Apple intended this, we’re not going to be giving them the information they need to fix it.

One’s own circumstances are rarely the only consideration. If all you care about is what’s happened to you, then sure, maybe it doesn’t matter whether it was malice, error, or incompetence. But I would hope that most of us care about preventing other people from suffering the same misfortunes we have. And even if all one cares about is oneself, if it was truly an error, there’s a possibility that providing information about it to Apple will lead to not just a fix for the issue itself, but assistance with cleaning up the aftermath.

There is no benefit in assuming evil intent where there is none. Not to you, not to me, not to anyone. It’s ultimately the same principle as “it costs no more to be nice than to be nasty”, except stronger, because at least being nasty may be temporarily emotionally satisfying. Assuming malice just intensifies negative emotion.

5 Likes
81

If Uncle Steve was alive, I sure hope he is not in his grave🤣. (Sorry, once an editor, always an editor)

3 Likes
82

You’re referring to Hanlon’s razor:

Never attribute to malice that which is adequately explained by stupidity.

I’ll just respond by citing Gray’s Law, a variation on Clark’s third law:

Any sufficiently advanced incompetence is indistinguishable from malice

These days, we’re seeing some pretty advanced incompetence in all sectors of our lives.

More specifically in this case, any company pushing out an update like this should be expected to test that update. And at least one of the tests should involve canceling the notification. Either they didn’t run that test, or they did and ignored the result. Incompetence or malice? Impossible to tell from here.

2 Likes
83

Once again demonstrating that the fastest way to get correct information is to post incorrect information. :slight_smile:

In this case, I wasn’t sure of the exact original wording or what it was called. It was easier to paraphrase it than look it up.

I don’t agree with Gray’s Law. I believe that intentions matter. Actions matter also, but Gray’s Law accounts only for actions and not intentions. Intentions are the indicator of whether someone may accept responsibility for their errors: the intentionally evil never will, but the incompetent sometimes will.

Also, as I mentioned in the comment you quote, I’m including bad luck and honest mistakes (as opposed to incompetent error) in this case. All three are separate possibilities that stand in contrast to malice. Neither bad luck nor honest mistakes need be a result of malice or incompetence, even though the consequences can be just as devastating.

Of all the FAANG-level companies, Apple is the one that has, since Jobs returned, routinely and generally consistently demonstrated that they care about users and how what they do affects them. That’s not always the topmost priority, but that’s understandable since they’re a for-profit business with stockholders to satisfy. (We’ll save criticisms of that model for another discussion.) And even with the other FAANG-level companies, I’m usually more inclined to attribute bad actions to apathy than to malice, although I’ll grant that the distinction between the two can be quite thin sometimes. (Not with Meta, though. I believe they’re primarily genuinely evil.)

What this all boils down to is that I really dislike the fact that every time Apple makes a screwup, no matter how minor, someone will inevitably come along and use it to support the idea that Apple is “evil” and doesn’t care about their users, despite all the evidence to the contrary. Do they make mistakes? Absolutely; everyone does. Are those mistakes sometimes catastrophic for some users? Unquestionably, but that’s going to happen sometimes (see “bad luck”).

While that kind of sentiment is par for the course in many forums, I’ve long felt that TidBITS readers and TidBITS-Talk contributors are a little higher on the scale than the general Internet commenter. I see this kind of comment as a doorway to reducing the quality of discourse here to that of the average Reddit community, and I don’t want to see that happen. I’ve been reading TidBITS since its inauguration, and joined TidBITS-Talk when Adam first founded it. I want this community to stay above the trolling, ranting, and baiting I see just about everywhere else. And I’m pretty sure Adam does, too.

6 Likes