"Invalid B-tree node size (4, 0)"

[toke lahti]

8TB (almost full) Time Machine disk, hfs+.

Is there any other way to fix B-tree(s?, a)ttribute, c)atalog, e)xtents?) than to buy $120 DiscWarrior and even that might be dead end?

This was the last of my "old" TM backups, one was physically detoriated, the second also corrupted, so that copying it to a new drive wasn't possible.

Looks like TM did not succeed to be "long term backup", it is (or has been) only for backup of what-you-have-now.

I had ZERO problems (yep, had used TM from Snow Leopard) with TM backups before I jumped from Mojave to Monterey.
Now those old backups are gone, I think.
Not fun.
Silver lining is, that I accidentally restored one install to my mini2012 earlier this year.

Once again, I'm wondering, can maybe a bad usb cable cause this...

 

[toke lahti]

TokesMini2018:~ ext-toke$ sudo fsck_hfs -r -d /dev/disk6s2
Password:
journal_replay(/dev/disk6s2) returned 0
** /dev/rdisk6s2
    Using cacheBlockSize=32K cacheTotalBlock=81920 cacheSize=2621440K.
   Executing fsck_hfs (version hfs-627.100.6).
** Checking Journaled HFS Plus volume.
   Invalid B-tree node size
(4, 0)
** The volume   could not be verified completely.
    volume check failed with error 7
    volume type is pure HFS+
    primary MDB is at block 0 0x00
    alternate MDB is at block 0 0x00
    primary VHB is at block 2 0x02
    alternate VHB is at block 15627381342 0x3a376ea5e
    sector size = 512 0x200
    VolumeObject flags = 0x07
    total sectors for volume = 15627381344 0x3a376ea60
    total sectors for embedded volume = 0 0x00
    CheckHFS returned -1317, fsmodified = 0

I can't find info about TWO exit codes ((4, 0)).
Where does the zero point?

 

[toke lahti]

Does anybody know why

sudo fsck /dev/disk1s1

does not work in Ventura?
How has it changed?

 

[Fishrrman]

Yet another Fishrrman "stupid question™":
Do you really, really need 8tb worth of very old tm backups?

 

[toke lahti]

Does anybody know why

sudo fsck /dev/disk1s1

does not work in Ventura?
How has it changed?

Yet another Fishrrman "stupid question™":
Do you really, really need 8tb worth of very old tm backups?

I wouldn't die for them.
But they are nice to have.
I'd say preserving digital history is quite handy skill to have.
It is for me something like making a bed at morning is to some people.
Confidence that you can do something.

So, I'm quite pissed at Apple.
13 years of Time Machine copies, no problems what so ever.
And because of that I started to trust TM more and more.
Then jumped from Mojave to Monterey and all old backups have started corrupting.

 

[f54da]

fsck for hfs is open source, so if you really want to know what that return code means you can check, albeit it's probably not practically useful. DiskWarrior will likely fix it, since it's only a catalog error and diskwarrior's entire raison detre is to basically rebuild the entire catalog.

 

[toke lahti]

Anybody has a extra license to DiskWarrior not needed anymore and selling for discount?

 

[toke lahti]

fsck for hfs is open source, so if you really want to know what that return code means you can check, albeit it's probably not practically useful. DiskWarrior will likely fix it, since it's only a catalog error and diskwarrior's entire raison detre is to basically rebuild the entire catalog.

Question was:
fsck-hfs works in my mac.
Fsck does not.

Is this common for Ventura?

 

[toke lahti]

fsck for hfs is open source, so if you really want to know what that return code means you can check, albeit it's probably not practically useful. DiskWarrior will likely fix it, since it's only a catalog error and diskwarrior's entire raison detre is to basically rebuild the entire catalog.

I bought DW and then they say:
"DiskWarrior will repair the directory and allow access to the files. It simply won't be a Time machine backup set. The important part is that the files will be available."

I don't understand what that means.
If there's 60TB of TM backups in that 8TB drive, how can DW make those files available when loosing the TM structure?

 

[gilby101]

I'd say preserving digital history is quite handy skill to have.

Absolutely. Once you have recovered what you can, consider updating those skills.
TM (and CCC, etc.) are backup and recovery tools. Archiving is a different activity. Think Amazon S3 Glacier Deep Archive or, for home/local storage, optical drives.

 

[f54da]

I don't understand what that means.
If there's 60TB of TM backups in that 8TB drive, how can DW make those files available when loosing the TM structure?

The catalog contains information on folder structure. It also contains initial pointers to file contents (with additional pointers for large files stored in extents overflow tree). Time machine backups (pre APFS) deduplicate via use of hardlinks, which HFS just considers a special type of file. When diskwarrior repairs the catalog, it is possible that some folder hierarchy information or hardlinks might get lost.

The net result is that you will still have all the file data, but it will not be directly restorable as a time machine backup. In the worst case things such as file names might not be recoverable.

 

[toke lahti]

The catalog contains information on folder structure. It also contains initial pointers to file contents (with additional pointers for large files stored in extents overflow tree). Time machine backups (pre APFS) deduplicate via use of hardlinks, which HFS just considers a special type of file. When diskwarrior repairs the catalog, it is possible that some folder hierarchy information or hardlinks might get lost.

The net result is that you will still have all the file data, but it will not be directly restorable as a time machine backup. In the worst case things such as file names might not be recoverable.

I tried to restore from DW's "Preview"-window the newest TM backup on the disk.
I got about 20 000 empty folders.

I'll clone the disk to a new one and try with that.
It just takes 2 days for cloning and other 2 days for DW to read through the disk...