The SSD is visible in Windows and you can open some files, but you cannot save a new document to it. Before you start looking for a command to remove write protection, secure the copies of data you need. A rejected write may be a software restriction, but also a health state reported by the NVMe controller. A single error window will not distinguish between these causes.
This procedure is for a detected SSD from which data can be read stably. It is not a guide for a drive with zero capacity, an inaccessible controller, or repeated disconnections. At the same time, it does not claim that every readable SSD that refuses writes already has an irreversible physical fault.
Why „read-only" alone is not a diagnosis
NVM Express describes health warnings and the ability to report a degraded or read-only state. The specification distinguishes the read-only media state from a change in write protection of a namespace. A namespace is a logical addressable space of the NVMe controller, not simply a regular Windows partition.
In addition, Windows may reject writes due to permissions, application restrictions, or the state of the file system. An attribute or report from the host system is therefore not automatically proof that the controller itself enabled the protection due to the media. Nor does it follow that it can be safely removed with a host command.
Device health data and file content answer different questions. A warning may be a reason for further assessment, not a list of damaged documents. Verifying content, in turn, does not replace an assessment of the drive’s reliability for further operation.
Determine the scope without test writes
Write down the complete message, the affected location, and the application. Did only one document in one folder fail, or several different operations on the same device? Is it the system drive, an external NVMe in an adapter, or secondary storage? Use observations of operations that have already happened before you add another test.
Do not probe the restriction by creating dozens of test files, and do not delete content „to release the protection". If you have an existing health report from a trusted tool, keep the exact details including the model and the messages. Obtaining another report must not mean installing a tool on the affected media or running a long test on an unstable device.
If the computer is managed by a company, involve the administrator. The restriction may be deliberate policy, and the administrator needs to know the impact of a change on both applications and sensitive data. Do not perform a repair in privileged mode just because a regular account refused the write.
Drive not responding, clicking or reporting errors? Stop trying anything else.
What not to try any more:
- do not keep powering the drive on or reconnecting it,
- do not run data recovery software, CHKDSK or formatting on the failing drive,
- do not open or take the drive apart.
How it works with us and what it costs: SSD data recovery.
Initial HDD and SSD diagnostics free of charge — we find out what can be recovered and quote the price in advance; you decide only then. NONSTOP line 775 556 063.
Non-standard cases (drive opened by another service, overwriting/formatting, complex analysis) cost 1 210 Kč including VAT (1 000 Kč excluding VAT) — we tell you in advance.
A crossroads for further work
| Observation | What is not yet proven | Safe response |
|---|---|---|
| A single application’s write fails | The state of the controller itself | Record the scope and verify with the responsible administrator |
| Multiple operations refuse writes | A physical cause or repairability | Protect the data you need, assess the device’s state |
| SMART reports a critical warning | Specific damaged files | Keep the report and verify copies separately |
| Reading ends in errors or the drive disappears | The safety of further home attempts | Stop ordinary copying |
Do not use unverified life-percentage figures as a countdown of remaining time. From a single health log you cannot promise how many files you will still retrieve or how long the drive will last. Likewise, a temporary return of write capability after a restart does not necessarily mean the fault is gone.
Stable reading: retrieve the important copy
If the device remains available and reading shows no errors, prepare a different healthy storage medium. Do not mistake a second folder on the same SSD for an independent copy. For a company system, first agree whether ordinary file copying will capture the state you need; an open database has different conditions than a closed document.
- Write down your priorities and find out what already exists in a verified copy outside the SSD.
- Copy the needed accessible files to the chosen healthy target. Do not install tools on the affected drive or save results to it.
- Verify the specific copies by opening them and comparing the content with the owner’s needs. State what has been verified and what has not.
- If new read errors, disconnections, or unusual delays appear, stop further home attempts. Do not return repeatedly to the same failing file.
- Once the important data is retrieved, hand the state over for assessment and agree on further operation. A retrieved copy is not confirmation that the SSD can again be considered reliable storage.
The expected outcome is a usable verified copy, not an automatically repaired drive. If the conditions for stable reading are not met from the start, do not try to force them with ordinary copying. The method of retrieving the content requires a different assessment.
What does not belong ahead of data protection
Formatting, secure erase, and other erasing operations do not save the content you need. A firmware update, a file system repair write, or an attempt to clear the read-only attribute are not a safe default response to an unclear cause. Every such intervention needs a specific reason, a verified procedure, and a resolved data need.
This does not mean that host write protection has no legitimate administration, or that firmware never helps with an operational problem. It means that a command recommended for one kind of restriction must not be used on an unknown controller state and the only originals in existence. A benchmark or repeated restarts will not safely remove this uncertainty either.
An SSD has no rotating platters or HDD motor. Do not look for confirmation of its read-only state in mechanical clicking. What matters is availability, specific messages, and read results — not a generic list of hard drive sounds.
What to hand over for assessment
Prepare the SSD model, connection type, system, and the exact messages. State whether it is the system drive, what you can still read stably, what you managed to copy, and where reading failed. Add any existing health report, a timeline, and a list of changes, including updates, repairs, and deletions.
State the encryption and the availability of the required key without disclosing the key itself. Agree on a secure method of handover separately. Plan the SSD replacement or reinstallation only in relation to the resolved recovery of the data you need, so that a subsequent service procedure does not erase an original you still need.
Frequently asked questions
Does a rejected write automatically mean a faulty SSD?
No. It may also be a host-side restriction, permissions, or another system state. Telling them apart requires specific evidence; a single message is not a diagnosis of the controller.
Can a read-only command clear the media state reported by an NVMe controller?
That cannot be assumed. A host attribute and the media state reported by the controller are different things. Before changing anything, you need to know what is being changed and how your data is protected.
Does SMART prove which files are intact?
No. It provides information about the device. Content and required completeness are verified on the retrieved file copies, separately from the assessment of the SSD’s health.
Is stable reading the same as a healthy drive?
It is not. It may allow you to retrieve a copy, but it does not guarantee reliable further operation or the integrity of all data. If new errors appear, stop this do-it-yourself branch.
Is zero capacity the same problem as read-only?
These are not the same observed state. This guide assumes a detected device with accessible files. Zero capacity or a non-functional detection requires separate assessment.
If a copy cannot be safely retrieved or the SSD’s state remains unclear, describe its behavior to ITHOPE. Attach the specific messages and an overview of the data already retrieved; the next intervention is chosen based on the cause, not based on a universal unlocking command.
Sources and limits of the procedure
Documentation verified on 7 October 2026. For a different system version, verify the corresponding manufacturer procedure.
Where to find us
ITHOPE is a company based in Brno, Czechia: we do data recovery from disks, RAID and NAS, PC and laptop service and IT for business, all in our own lab at Viniční 3587/216, 615 00 Brno.
Phone +420 775 556 063 (24/7 line). You can bring the device in person, or send it by courier (Zásilkovna, PPL).