Comment by baryphonic | Hacker News Reader

Comment by baryphonic | Hacker News Reader

Parent Full thread

This is interesting. I sat down at my machine on Thursday morning, and it was entirely unresponsive. It had warned me about installing updates the previous evening, but I've done this many times without an issue, and regardless, they are always complete by morning. I waited a minute or two for any sign of life (beyond the fans running), and then held power. I heard the unpleasant sound of the fans spinning up to high RPM and then shutting off rapidly. I waited a few seconds and then held power again. The machine showed a slow progress bar, but then finally I saw "26 minutes remaining" below it. I was pretty surprised that the update was running, but I figured it had just started late or hung, and ran a bit later into the working hours.

This article makes me reevaluate that thought.

Coincidentally, one of my colleagues had warned the team about installing macOS updates, since he was having repeated issues with kernel panics on the original 10.15.4 update.

I hope Apple can get some of this together a little better going forward.

I had a similar issue to yours when I manually ran this update on my 16". It froze on a black screen, after doing the BridgeOS part of the update. I had to force it down. When it started, it continued the update and, thankfully, everything worked.

However, I'm still having kernel panics after the Supplemental update. But interestingly, they only happen when the machine is supposedly sleeping, with the shell closed. I kid you not. I will have it sitting there on my end table in the living room, shut for over an hour, and all of the sudden it will reboot with the lid closed (I know because I have the startup chime enabled). I've never had a Mac kernel panic while supposedly sleeping. This is so f*cked.

Even a supposedly sleeping Mac will wake up from time to time, doing exactly what I don’t know. But I consistently see it in the logs.

It will sync email and other things so that everything is already downloaded when you wake it up. See https://support.apple.com/en-gb/HT204032

And it won’t do CRON because reasons

Bugs me too. Your best bet is launchd which solves the problem, but it is a bit of pain to write job definition property lists. More here: https://www.launchd.info

This is a useful library for plists, in case you're interested - https://docs.python.org/3/library/plistlib.html

> it is a bit of pain to write job definition property lists

http://launched.zerowidth.com/ makes it easier.

Ironically -- I think this bug is fixed in a like 10.15.3 or so. Eitherway a cron I set up over a year ago before I realized this bug has started mailing me since I updated to Catalina.

Ah, thanks. That jogged my memory. I did turn off power nap a long time ago, but my laptop still kept waking up at night anyhow. But I just checked, and now it seems to work as advertised.

Wouldn't it be fantastic, if Apple had an easy way to help users learn, why their mac woke up? Currently that is complex scientific work.

> Mobile Device Management can remotely lock and wipe your Mac.

I appreciate the transparency here. Outside of API documentation, a lot of plain-language information doesn't go in-depth enough about MDM capabilities.

Yes, I know this, but when a Mac does this, it does it only to run routine Apple code, stuff that should not, under any circumstances, cause a kernel panic. And up until Catalina, that it never has, barring binary corruption or defective hardware.

In this case, the kernel panic seems to be related to the Intel Iris video driver.

Same thing was happening to me recently. I turned off Power Nap and it hasn’t happened since.

For added irony, my Hackintosh installed the update without missing a beat.

I have been a Hackintosh user since 2010 and I’ve never had a system break in point release. If you set it up right, it’s not that common.

Major version upgrades are an entirely different matter, although even they sometimes just work with Clover.

I just moved my system from Clover to OpenCore last week. A couple of hours of paying attention and reading and voila.

So did my Mac and my wife's Mac. I don't see the irony. This probably affects a small subsets of Macs [1] and probably still the count of successful updates is much higher for Apple Macs than Hackintoshes. But yes, if you cherry-pick anecdotes, you can get any outcome you want.

[1] Which is still terrible, these things should just not happen. Bad quality control from Apple' side and inexcusable for > 1000 Euro 'premium' computers.

Most hackintoshes are configured not to auto-update too.

Reply on news.ycombinator.com