How to tell your X archive is complete and intact: a hashing walkthrough
Why archives come out incomplete
Many people think that once the download finishes and the zip lands on disk, the job is done. Archives can break at three points: the platform omits recent entries while packaging, the download is cut short by a network drop, or files corrupt during extraction. Our article on missing recent tweets notes that the most common gap happens at packaging. Verification is the insurance you add afterward, and it does not depend on "it felt finished" but on a repeatable digital fingerprint.
What a hash is, and why it helps
A hash is a fixed-length string computed from file content. Change one byte and the hash becomes completely different. It is a fingerprint for the file: matching hashes let you trust that two files are identical. Eyeballing size or name is far less reliable, because same-named files can differ and similarly sized files can differ by a single byte.
Practice: get and compare hashes in three steps
- After the first backup, run a hash command on the archive and store the output in a manifest
- After the second backup, hash the same named file again
- Put the two hashes side by side; identical means unchanged, different means investigate
Which hash algorithm to pick
For everyday use SHA-256 is enough. It resists collisions better than MD5 and is the default on most systems. Unless you have a compatibility requirement, there is no reason to fall back to MD5. Make sure both runs use the same algorithm, because mixing them means they will never match.
How to keep the comparison tidy
Do not keep it in your head. Make a text manifest where each line is date, filename, and hash. Append a new line each backup and the history stays clear. Here is a sketch:
| Date | File | Hash prefix |
|---|---|---|
| 2026-09 | archive-2026-09.zip | 9f3c... |
| 2026-10 | archive-2026-10.zip | 1b7a... |
Check the media files too
The images and videos in an archive are usually the largest and the easiest to chip during a dropped download. If you only hash the whole zip, small damage can be averaged out. A safer move is to hash the extracted media directory again, or sample-check a few large files by size and hash.
What to do when you find corruption
If the hash changed but the size did not, suspect corruption first. The order: keep the bad copy, do not overwrite it, request a fresh archive, hash the new download immediately, and compare the new and old packages. To build the full monthly routine, see our guide to automating archive backup.
Return to the homepage for more hands-on digital footprint guides.
Frequently Asked Questions
What exactly is a hash value?
A hash is a fixed-length string computed from file content. Change even one byte and the hash changes. It acts like a fingerprint that confirms two files are identical. The same file yields the same hash on any computer, which makes it good for comparison over time.
If the archive is corrupted, do I lose old tweets?
The damaged copy is the one on your side; the platform still has the data. Request a fresh archive to get a complete package, which is exactly why two backups matter. Even if one copy breaks, the other still works.
Do ordinary users need to learn hashing?
You do not need to master it. Run one command, read the output, and compare it with last time. You do not need to understand the math to benefit from the comparison. Most systems ship a hash command, so no extra software is needed.
The size is normal but the hash changed, is that OK?
Possibly. The platform may have changed the archive format or backfilled content, which changes the hash. As long as the size also changed, it is likely content update; unchanged size deserves a closer look. If two runs both show unchanged size but different hashes, re-download once.
Check your own X/Twitter footprint
Free on-device scan. Your archive never leaves your computer.
Start Free CheckRelated Reads
Designing a Resumable Deletion Job: Checkpoints, Idempotency and Safe Restarts
Any deletion job handling tens of thousands of posts will get interrupted, whether by a dropped connection, a sleeping laptop or a reclaimed process. What separates a job that resumes from one that starts over is checkpointing, idempotency and state recovery. This covers the design and a minimal implementation skeleton.
Local vs Cloud Parsing: Which Way to Process Your X Archive Is Actually Private
The first step to cleaning old X tweets is downloading your data archive (ZIP), but how you read that archive decides your privacy boundary. This piece compares local parsing with cloud parsing and gives a workable technical approach.
Turn X archive backup into a monthly automated task: scripts, scheduling, and verification
Manual X archive downloads get forgotten. This guide lays out a practical automated backup: schedule a script to pull your X data archive, store it locally and in the cloud, and verify integrity so a complete copy is always on hand when you want to clean up old tweets.