**How to avoid overwriting files during MySQL dump from dev to production import in Strapi with S3?**

[keerthi8632]

How to avoid overwriting files during MySQL import in Strapi with S3?

Context:

• CMS: Strapi v4.25.12

• Database: MySQL

• Media: AWS S3 using @strapi/provider-upload-aws-s3

• Deployment: Jenkins CI/CD pipeline

• Issue: During deployment, we import a MySQL dump from the dev environment to production. The dump overwrites the files table, causing loss of production media metadata. While the S3 files remain intact, frontend and admin panel media links break due to missing metadata.

What We Tried:

• Excluding files table via --ignore-table works but loses new dev media added.

• Manual sync is unreliable and not scalable.

Goal:

How can we safely import development data without overwriting files entries while also syncing new media records (if any)?

Ideal Solution: A way to:

• Retain existing production media entries

• Add new files entries from dev without deleting or breaking prod ones

Is there a common practice in Strapi + MySQL + S3 setups for this?

---

Thanks in advance!

Attempted MySQL dump with --ignore-table=files
Command:

mysqldump --ignore-table=mh_website_db.files...
✅ Expectation: Prevent overwriting production media metadata.
❌ Result: Although this preserved existing media, it skipped new media records from the dev environment that we wanted to promote — breaking references to newly added assets.

Tried full dump/import (no ignores)
✅ Expectation: Full environment sync with new content and media from dev.
❌ Result: This overwrote valid production files entries — causing broken media links in the admin UI and frontend due to missing metadata.

Tried manually comparing and inserting missing rows
✅ Expectation: Keep existing production entries and append only new dev media.
❌ Result: Not scalable, hard to automate, error-prone (especially with associations like related_morph),due to primary keys its mismatching in production existing data.