How to Move a Failing 320GB Debian HDD to a 240GB SSD Without Losing Important Files
TL;DR:
To safely migrate a failing 320GB Debian HDD to a smaller 240GB SSD, you must first image the entire original drive to a sufficiently large third storage device using GNU ddrescue instead of attempting a direct clone or partition shrink.
● Halt all server workloads immediately and avoid running fsck repairs, live resize2fs operations, or deleting Docker caches on the failing Seagate Momentus drive to prevent further degradation from write operations.
● Install a fresh Debian environment on the 240GB SSD and manually restore validated databases, application volumes, and system configurations from your safe image rather than attempting to preserve the damaged boot environment.
● If a third drive with at least 320GB capacity is unavailable, prioritize copying only the most critical intact data directly to healthy storage, but power off the disk for professional lab recovery if it exhibits physical clicking or repeatedly drops offline.
Ask AI for a summary
ChatGPT
Perplexity
Gemini
Claude
Grok
Do not try to clone a failing 320GB Debian boot disk directly onto a 240GB SSD or shrink the original ext4 partition as your first recovery step. A smaller replacement cannot hold an unchanged whole-disk image, and resizing an unstable source adds writes when you should be preserving its remaining readable data.
In this Debian server recovery question on Reddit, a 320GB Seagate Momentus 5400.6 laptop HDD still boots but causes intermittent system failures. Linux Disks reports an ext4 file system that needs repair; the user owns only a roughly 240GB SSD and wonders whether they can remove expendable Docker data and copy partitions over. The actual used space and drive health are not established from the text alone.
Separate Rescuing Data From Making The New SSD Bootable
Stop normal server workloads and prevent new logs, containers, package updates or database writes to the failing disk. Do not run fsck repairs against the only copy or attempt a live resize of the unhealthy file system.
With adequate spare healthy storage, the preferred first step is to image the full 320GB source disk to a file or disk large enough to hold it, keeping the mapfile. The GNU ddrescue manual explains how it prioritizes readable areas and resumes a controlled imaging job. A 240GB SSD alone is too small to serve as a full-size destination for a 320GB raw image.
Decide By Drive State
- The HDD is physically unstable, clicks or repeatedly drops offline: Turn it off and consider a qualified recovery lab, especially if the server holds irreplaceable data.
- The HDD still reads steadily and an adequate third drive is available: Image first, then extract important configs, application data and user files from the image rather than repeatedly reading the original.
- No full-size rescue destination is available: Prioritize intact backups and the most valuable accessible data onto healthy storage, accepting that further reads carry risk. Do not spend the last readable hours pruning Docker caches on the failing disk.
Build A New Debian Installation On The Smaller SSD
Once the important files are protected and you have confirmed they fit, install Debian on the 240GB SSD and restore the needed system configuration, application volumes and data from the safe copy. Verify databases and application-specific backups, not just filenames. This avoids making the recovery attempt depend on preserving the old partition geometry or an already damaged boot environment.
The official resize2fs documentation permits shrinking an unmounted ext4 file system, but it does not resize partitions and assumes the underlying data can be managed safely. If resizing is later necessary for a controlled migration, do it only on a verified backup or working copy, never as the first experiment on the failing original disk.
Recoverit or another file-recovery application is not a substitute for saving or imaging a deteriorating system disk. The next objective is to preserve critical files once; getting a fresh Debian installation to boot comes afterward.