macOS 26.4 Beta: Problems Mounting HFS+ Volumes
I normally don’t write about beta bugs, but I’ve seen lots of people discussing this one and also received customer questions about it. I can reproduce it on my Mac, though not for every HFS+ volume.
HFS external media might fail to mount automatically. (168672160)
Workaround: For macOS only, use CLI tool
diskutil mountto attach the relevant disk device.
This also affects disk images, both creating them and mounting them on macOS 26.4 (even if they were created using another version). I discuss some workarounds for DropDMG here.
Thomas Rohde (Reddit, MacRumors, 2):
Apparently
fsck_hfsis broken
I tried to warn as many people as possible about HFS issues but I did not at the time know it was
fsck.
I have 7 (seven!) external drives that I currently can’t use 🙄
Finally figured out how to properly mount my external HFS drives under macOS 26.4b1!
diskutil mount—suggested as a workaround in the release notes—was SURE not cutting it. I had to usesudo mount -t hfsinstead (specifying the filesystem format and mount path manually). What a pain it was![…]
They are all standard hard drives and I never migrated them to APFS because I didn’t want the loss of performance (plus, I’ve seen some pretty nasty corruptions on APFS that were really REALLY hard to recover from while I have my faithful DiskWarrior for HFS+).
Previously:
5 Comments RSS · Twitter · Mastodon
On the one hand, I just don’t understand, in terms of release cycles, why this — a sudden regression in a minor release — keeps happening. Is their test suite so small, their code base so brittle, their willingness to cherry-pick changes from main so large? Why does 26.4 suddenly change the HFS+ stack?
OTOH, Apple has it in the beta release notes, so that may suggest 1) they were indeed making a change to the disk subsystem, knowing the risk, 2) they intend to fix this by final.
By far the most bemusing thing about this is that it is clear that Apple was aware that there was an issue with the HFS+ filesystem module, but the workaround doesn’t work in many cases.
It is clear from the rpath error that fsck_hfs is some kind of internal debug build, but it makes you wonder how exactly they assemble the OS. I mean, how did that get pulled in, but the other binaries are okay? mount_hfs is fine.
Apple can't even get it right in the release notes. The bug impacts HFS+ volumes, not HFS volumes. Support for HFS volumes was removed in 10.15.
I wonder if they're doing something silly like rewriting fsck_hfs in Swift. You know, since the OS is mostly bug-free and polished.
Maybe this is the special Apple experience: not being able to use external media. Did they pull this build?
