iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
The safest archive is a managed system, not a single disk: keep at least three copies on two different kinds of media, place copies in separate physical locations, create checksums when files enter the archive, check those checksums on a fixed schedule, and keep a documented way to replace a bad copy. A checksum tells you that bytes changed; it does not rebuild the missing bytes, so recovery depends on an intact independent copy.
This routine addresses two different risks. Bit rot and drive failure threaten the integrity or availability of the stored bitstream. Format, software, codec, and hardware obsolescence can leave an unchanged bitstream difficult or impossible to open. Integrity checks and redundant copies address the first risk; format documentation and planned migration address the second.
What bit rot and drive failure mean for an archive
Bit rot is an integrity problem
Bit rot is a gradual or sudden change to stored bits. The change may come from a failing disk sector, controller or cable problems, silent corruption during copying, filesystem errors, or another storage fault. You may not notice it until a file is opened years later. A file that still has its original name and size can nevertheless differ from the version you archived.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Drive failure is an availability problem
A magnetic hard disk, solid-state device, or removable medium can stop responding, become unreadable, or suffer enough damaged sectors that files cannot be recovered normally. National Archives and Records Administration guidance emphasizes that media life expectancy varies and recommends copying content into a current storage environment as soon as possible for both solid-state and magnetic media. No consumer-drive lifespan or failure percentage should be treated as a guarantee.
#1 Best Overall
- Easily store and access 2TB to content on the go with the Seagate Portable Drive, a USB external hard drive
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Format obsolescence is a usability problem
Perfectly preserved bits are not automatically usable in the future. A file can depend on an obsolete format, application, codec, operating system, or hardware interface. The Library of Congress distinguishes bit-level preservation from continued future access, and the NDSA Levels of Digital Preservation call for preserving information needed to use files and planning for format changes where appropriate.
The minimum architecture: independent copies in different places
Library of Congress Digital Preservation Practices, updated December 27, 2023, states: “Save three copies of digital files on at least two different kinds of storage media made by different manufacturers.” Its personal-preservation FAQ also recommends at least two media types in separate physical locations. Treat those as design requirements, not as a promise that any particular product will last.
- Three copies: maintain the working archive plus two recoverable copies.
- Two media types: combine categories such as an external hard drive and cloud storage rather than relying on identical devices.
- Different manufacturers where practical: this reduces dependence on one model or production fault.
- Different locations: keep at least one copy outside the building containing the primary archive. Fire, theft, flood, and power events can destroy collocated devices together.
A single external drive is therefore one component, not an archive plan. A mirrored array or other redundant arrangement can help a system stay online when one device fails, but it is not an independently stored backup if the same fire, theft, malware event, controller fault, or administrative mistake can affect every member of the array. Keep a separate, disconnected or separately administered copy.
Choose complementary storage rather than one “perfect” medium
The following comparison focuses on preservation functions. Actual durability, service limits, pricing, and features vary by device, provider, region, and plan; the cited preservation guidance does not guarantee a particular service or drive.
| Approach | Copy independence | Location separation | Fixity checking | Recovery characteristics | Refresh effort |
|---|---|---|---|---|---|
| External hard drive kept with the computer | Independent device, but vulnerable to the same local incident | None unless moved elsewhere | Easy to scan when connected | Fast for large local restores | Requires periodic replacement and a connection test |
| Second external drive stored off-site | Independent device; use a different manufacturer where practical | Strong physical separation if the site is genuinely separate | Scan when retrieved or through a controlled rotation | Depends on access to the off-site location | Requires transport, inventory, and refresh tracking |
| Cloud storage | Can be independent of local hardware; verify what redundancy and versioning the service actually provides | Usually provides geographic separation, but confirm the service terms | Possible through provider tools or your own manifest when files can be read back | Convenient remote restore; internet access and account access are dependencies | Provider-managed infrastructure does not remove the need to verify access and maintain a recovery copy |
| Mirrored or redundant array | Often not independent: members may share a case, power, controller, and administrator | Normally none | May detect some faults, but do not assume it performs archival fixity checks | Can improve uptime after a single-device failure | Replacement and rebuilds still require monitoring and a separate backup |
External hard drives and cloud storage are complementary choices. For example, a local copy can make a large restore practical while an off-site copy protects against a local disaster. Whichever combination you use, document where each copy is, who can access it, and how you will read it back.
Rank #2
- Easily store and access 5TB of content on the go with the Seagate portable drive, a USB external hard Drive
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Build the archive at ingest
Fixity information is most useful when it is created before corruption can occur. The Library of Congress recommends establishing fixity as early as possible and logging it at both file and aggregate levels. An aggregate can be a folder, disk image, collection, or other defined set of files.
- Define the collection. Decide which files belong together, assign a stable collection name, and record the source, acquisition date, creator, rights information, and any original folder structure that matters.
- Use stable paths and names. Avoid changing names or moving files after checksums are recorded unless the change is logged. If a rename is necessary, preserve the old and new path in the log.
- Compute a cryptographic checksum for every file. Use one algorithm consistently. SHA-256 is a practical choice; the Library of Congress describes comparing MD5 or SHA values against manifests. A checksum is an identifier for the bytes, not a backup and not a repair code.
- Write a manifest. Store the algorithm, checksum, relative path, file size, timestamp, and collection identifier in a plain-text or structured file. Keep a checksum for the manifest itself, and record when and on which system it was generated.
- Verify each copy after transfer. Do not assume that a completed copy operation means every byte is readable. Recalculate checksums on the destination and compare them with the ingest manifest.
- Replicate only from a verified source. Create the second and third copies from the verified set. Record destination, device or service identifier, location, date, and verification result for each replication.
- Protect the records. Keep the manifest and logs with the archive, but also place a copy of them in a separate location. If the only manifest is on the failed disk, you lose the baseline needed to detect change.
Example checksum commands
On Linux or macOS, a single-file check can be recorded with:
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutesha256sum 'photos/2024/IMG_0001.jpg'To verify a manifest created in the standard sha256sum format:
sha256sum --check archive-manifest.sha256On Windows, the built-in command is:
certutil -hashfile "D:Archivephotos2024IMG_0001.jpg" SHA256These commands calculate values; they do not copy, recover, or repair files. For a large collection, use a script or checksum application that can recurse through directories, export a manifest, and report unreadable files. Test the tool on a small sample first, and preserve the tool’s algorithm and path conventions in your documentation.
Schedule fixity checks and retain an audit trail
The NDSA Levels of Digital Preservation (2013, Version 1) describe four progressive levels across five functional areas. Their relevant practices include checking fixity at fixed intervals, maintaining logs, detecting corrupt data, and, at more advanced levels, having the ability to replace or repair corrupted data. The document does not establish one universal interval for every personal archive, so choose a cadence based on how often data changes, how costly a loss would be, and how quickly you can obtain a clean copy.
Rank #3
- Easily store and access 1TB to content on the go with the Seagate Portable Drive, a USB external hard drive.Specific uses: Personal
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop. Reformatting may be required for Mac
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
A workable schedule
| Event | What to do | Record |
|---|---|---|
| At ingest | Hash every file, verify the first destination, then verify each additional copy | Manifest, algorithm, source, destinations, dates, operator, results |
| After any move or migration | Run a full comparison against the baseline or a newly approved baseline | Old and new device identifiers, changed paths, pass/fail report |
| Fixed periodic check | Read every file that the schedule covers and compare its checksum with the manifest | Start and end times, software, unreadable files, mismatches, action taken |
| After a suspected fault | Stop writing to the suspect device, preserve diagnostic information, and compare the affected files with another copy | Symptoms, device state, recovery source, replacement checksum |
For a practical starting policy, many personal archives choose a more frequent check for frequently accessed or irreplaceable collections and a less frequent check for stable, offline media. The important property is that the interval is written down, actually performed, and adjusted when experience or risk changes. A check that is never logged cannot demonstrate that the archive was monitored.
What a useful log contains
- Collection and manifest version.
- Checksum algorithm and software used.
- Device, volume, cloud account, or location identifier.
- Check date, operator, and scope (all files or a defined subset).
- Number of files checked, unreadable files, checksum mismatches, and missing files.
- Decision: accepted, restored from another copy, quarantined, or awaiting investigation.
- New checksum and reason whenever a file is intentionally changed or migrated.
Repair a mismatch without making it worse
When a checksum differs, do not overwrite the suspect file immediately. The mismatch proves that the current bytes differ from the recorded baseline; it does not prove which copy is correct until you compare the available evidence.
- Quarantine the suspect copy. Stop synchronization or automated jobs that could propagate the changed bytes to every copy. Mount the device read-only where your operating system and workflow allow it.
- Confirm the observation. Re-read the file and recompute its checksum. Check whether the path, filename, and algorithm in the manifest are correct. A deliberate edit, rename, or line-ending conversion can create a legitimate difference.
- Compare independent copies. Check the same file against the baseline on the other archive copies. If two verified copies agree and one differs, treat the differing copy as damaged or unintentionally changed.
- Restore to a new destination. Copy the known-good file to replacement media or a new path rather than repairing in place. Recalculate its checksum and compare it with the trusted value.
- Update the affected copy and document the action. Replace the bad file, verify the complete destination, and record the old result, source of recovery, new result, and date.
- Investigate the device. Look for read errors, filesystem warnings, cable or enclosure problems, and repeated mismatches. Retire a device that cannot be trusted instead of repeatedly rebuilding it without a cause.
If every copy has the same wrong checksum, redundancy has preserved the wrong state. That is why independent copies, versioned or offline history where feasible, and timely detection matter. Checksums can identify the problem; only an intact earlier copy or another authoritative source can supply replacement bytes.
Refresh aging media before it becomes unreadable
Library of Congress guidance gives two related planning statements: one guide says to replace media every three to five years, while its personal-preservation FAQ says, “Migrate saved copies to a current storage medium about every five years.” These are guidance intervals, not universal service-life statistics. NARA likewise says media life expectancy varies and recommends moving content into a current storage environment as soon as possible for both solid-state and magnetic media.
Start a refresh before the target date, not on it. Obtain new media, verify that it has enough capacity, copy from a known-good source, compare checksums, and keep the old readable copy until the new copy has passed at least one subsequent verification. Refresh every copy, including off-site copies, and update the inventory with purchase date, location, interface, capacity, and retirement date.
Free tools Windows power users keep installed
One-click scans. No signup required.
Rank #4
- Easily store and access 4TB of content on the go with the Seagate Portable Drive, a USB external hard drive.Specific uses: Personal
- Designed to work with Windows or Mac computers, this external hard drive makes backup a snap just drag and drop
- To get set up, connect the portable hard drive to a computer for automatic recognition no software required
- This USB drive provides plug and play simplicity with the included 18 inch USB 3.0 cable
- The available storage capacity may vary.
Refresh triggers that override the calendar
- The device reports read errors, bad sectors, degraded health, or repeated disconnects.
- The interface, power adapter, filesystem, or encryption method is becoming difficult to access.
- The provider is ending a service, changing export behavior, or making account recovery uncertain.
- The media has been stored in unsuitable heat, humidity, dust, vibration, or magnetic conditions.
- You can no longer perform a complete checksum read within the time or hardware resources available to your process.
After migration, perform a full checksum comparison and retain the migration log. A copy that was merely transferred, without verification, is not a completed refresh.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Preserve enough information to use the files later
Alongside the files, preserve a short technical description: formats and extensions, character encoding where relevant, compression or encryption, software and codecs needed for access, hardware or peripherals, dependencies between files, and any passwords or key-recovery procedure that policy permits you to retain. Keep licenses and documentation needed for lawful access.
Monitor formats that are central to the collection. When a format becomes difficult to render, plan a controlled migration to a supported format while retaining the original when feasible. Hash both the original and the migrated derivative, label the relationship between them, and record the software, settings, date, and person or process that performed the migration. A migrated file is a new object; its checksum should not be expected to match the original.
Common failure scenarios and responses
“The backup drive is plugged in all the time.”
Disconnect or otherwise isolate at least one recovery copy when it is not being updated. Continuous attachment exposes the backup to the same power event, malware, accidental deletion, and synchronization error as the primary system. Continue to verify the isolated copy on a planned rotation.
Recommended Free Tools
“The checksum tool reports a mismatch after I edited the file.”
That is an expected result of an intentional change. Preserve the prior version if it has archival value, create a new manifest entry or version record, and mark the old checksum as superseded rather than silently replacing the baseline.
Best Value
- High capacity in a small enclosure – The small, lightweight design offers up to 6TB* capacity, making WD Elements portable hard drives the ideal companion for consumers on the go.
- Plug-and-play expandability
- Vast capacities up to 6TB[1] to store your photos, videos, music, important documents and more
- SuperSpeed USB 3.2 Gen 1 (5Gbps)
“One copy is unreadable, but another passes.”
Quarantine the failing device, restore the file or collection from the passing copy to new media, verify the restored checksums, and replace or retire the failing device. Do not use the unreadable copy as the source for further replication.
“All copies open today, so format preservation is finished.”
Opening files today does not establish future usability. Retain format and dependency documentation, watch for access problems, and perform a controlled migration when continued rendering requires it.
“The cloud provider says data is redundant.”
Provider redundancy may improve service resilience, but it does not replace your own manifest, independent recovery path, or account-access plan. Confirm that you can download representative files, verify their checksums, and recover if the account, provider, or network is unavailable.
A repeatable archive checklist
- There are three current copies, on at least two media types, with at least one copy in another physical location.
- Each file has a recorded checksum and each collection has a manifest checksum.
- Every copy was verified after its last transfer or migration.
- A written schedule assigns the next fixity check and identifies its scope.
- Logs record pass, fail, unreadable, missing, restored, and intentionally changed files.
- The repair procedure names a trusted recovery source and a replacement destination.
- Media purchase dates, interfaces, locations, and planned refresh dates are inventoried.
- Formats, software, codecs, encryption, dependencies, and access requirements are documented.
- At least one recovery copy is protected from routine synchronization and local disasters.
- A sample restore has been performed, not merely assumed possible.
The practical standard
Protecting an archive is a continuing control loop: create verified copies, separate them, record fixity, check it on a schedule, investigate differences before they spread, restore from an independent good copy, and refresh media before access becomes uncertain. Apply a separate format-migration plan so that preserved bytes remain understandable. This layered approach is more dependable than trusting a single drive, a mirror, or an unverified cloud upload.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
