What Was Happening
Two retirements at once. A MacBook Pro that was being upgraded and an older laptop that was being decommissioned. The complication was the Samsung external hard drive that lived on the older laptop — it would mount cleanly on that machine and behave oddly on every other computer the customer had tried.
The destination was a single Lacie external drive the customer supplied, ready to plug into the next setup.
The interesting part of this job was the Samsung drive. Drives that “only work on one specific machine” almost always have a filesystem or partitioning compatibility issue rather than a hardware fault — and the fix is to image them via the machine they do work on, then deal with the contents on whatever platform handles them best afterwards.
Our Diagnosis
The order mattered. The Samsung drive had to come off first while the older laptop was still working — once that machine was decommissioned the drive would be much harder to read. The MacBook Pro could happen second because it was healthy and accessible.
Samsung HDD first:
- Connected the Samsung drive to the older laptop and confirmed it mounted as expected on that machine. The drive showed its full partition structure and all the user files were visible.
- Identified the filesystem. The drive was formatted with a version of HFS+ (Apple’s older filesystem) that had been written by an early macOS version. Newer macOS versions can read it but sometimes refuse to mount it depending on partition table quirks and the journaling state.
- Used the older laptop to make a complete image of the drive — bit-for-bit copy of the entire partition structure, not just a file copy. The image landed on dedicated dedicated recovery storage, untouched by anything but the imaging tool.
- Mounted the image read-only on a separate machine that could handle the older HFS+ variant cleanly (a specific macOS combination known to work with these legacy partitions).
- Verified the image by browsing the file structure and opening sample files. Everything intact.
- Extracted user content to a working folder on dedicated recovery storage.
MacBook Pro second:
- Booted the MacBook Pro in Target Disk Mode and connected to a separate machine via Thunderbolt. This gave us read access to the internal drive without booting the MacBook Pro itself, eliminating the small risk of macOS writing to the drive during a normal boot.
- Imaged the internal drive as a precaution before extracting user files.
- Extracted Documents, Desktop, Downloads, Pictures, Mail and a list of other user folders the customer had flagged as important.
- Verified extraction by opening representative files in their expected applications.
Merge and deliver:
- Organised everything onto the customer’s Lacie destination drive — a top-level structure with Samsung-drive content and MacBook-Pro content kept separate so the customer knew where everything had come from.
- Cross-checked against the customer’s list of “things I cannot lose” — every item present.
- Final verification by opening sample files directly from the Lacie drive to confirm portability.
The Result
Both sources recovered onto a single, well-organised destination drive. The customer could safely decommission the older laptop and the legacy Samsung drive — the data was now on a current external drive that would work with their next MacBook Pro and any other current machine.
Why This Happens
Several common causes, all software-side rather than hardware-side:
- Older filesystem variants. macOS has supported a series of filesystems over the years (HFS, HFS+, journaled HFS+, APFS), and Windows has its own progression (FAT32, NTFS, exFAT, ReFS). Older variants written by older OSes can confuse newer ones.
- Partition table format. MBR (older) vs GPT (newer). Drives partitioned with MBR on older systems sometimes confuse newer macOS or Linux that expect GPT.
- Journaling state. A journaled filesystem that wasn’t unmounted cleanly on the last machine that used it will sometimes refuse to mount cleanly on other machines until it’s re-mounted and unmounted properly on the original.
- Older driver requirements. Some external drive enclosures had vendor-specific drivers that no longer ship with current operating systems.
- Bus-power quirks. A USB drive that draws more power than the spec allows can fail to enumerate on some hosts but work on others — usually a hardware quirk but easily mistaken for a filesystem issue.
In every case, the right move is to image the drive via the host that does mount it cleanly, then work with the image on whichever platform is most convenient afterwards.
What to do with old external drives before retiring the host
If you have an external drive that only mounts on a specific older laptop:
- Image the drive before retiring the laptop. Once the laptop is gone, the drive becomes much harder to access.
- Don’t try to “convert” the filesystem in place. Conversion tools sometimes work, but a botched conversion can leave you with neither the old format nor the new one. Image first, then decide.
- Copy out, then reformat. Once you have the data safely elsewhere, reformat the drive to a current filesystem (exFAT for cross-platform compatibility, APFS for macOS-only, NTFS for Windows-only). The drive itself is usually fine; the format on it is the limiter.
- Don’t trust a drive that has weird mounting behaviour. Even if you got it working, treat the data on it as at-risk and get it onto something more stable.
How to keep your backup chain healthy
- One copy of important data is not a backup. Two copies on different media is a minimum. Three copies (with at least one off-site) is the proper standard.
- Use a current filesystem. Drives you depend on should be formatted with something widely supported now, not something that was current ten years ago.
- Rotate backup drives. A backup drive that lives plugged into the source machine is vulnerable to the same risks as the source.
- Test the backup occasionally. A backup you’ve never restored from is a backup that might not work.
Local Help in Putney SW15
The earlier we see a failing drive, the more we can usually recover.
If your drive is making unusual noises, disappearing from the operating system intermittently, or throwing read errors — stop using it and call us.
Workshop diagnosis on 020 7610 0500 or via the contact form.