Thee technical implementation is complete and functional - the remaining issue is deployment configuration specific to the Emergent hosting platform.
#URGENT: Deployment issue & wasted credits
red steppe
You're absolutely right to ask this - having 2 different Emergent URLs is unusual and is likely the root cause of all our problems.
Why You Have 2 URLs:
contentai-17.emergent.host
This looks like an older deployment or different app instance
The "17" suggests it might be app #17 in Emergent's system
This is where your production site has been calling
Has bugs/old code
viral-ai-content-3.preview.emergentagent.com
This appears to be your current/newer deployment
The "3" suggests it's a newer version/iteration
This is where I've been applying all the fixes
Has all the fixes and works correctly
What Likely Happened:
You initially deployed to contentai-17.emergent.host
Later, you created a new deployment (maybe during development) that became viral-ai-content-3.preview.emergentagent.com
Your custom domain easyviralcontent.com is still pointing to the old deployment (#17)
When you "Save to GitHub" it updates the new deployment (#3), not the old one (#17)
Solutions:
Option 1: Check Your Emergent Dashboard
Look for:
Multiple apps/deployments listed
Which deployment your custom domain easyviralcontent.com is connected to
Update the domain to point to the working deployment (viral-ai-content-3)
Option 2: Consolidate
Delete/deprecate the old contentai-17 deployment
Use only viral-ai-content-3.preview.emergentagent.com
Can you check your Emergent dashboard and see if you have multiple deployments listed? This would explain everything!
I have no idea how to solve this without an admin
waxen spoke
@red steppe firstly, please stop just copying and pasting everything the agent gives you, its a wall of text that is not very helpful
If you hardcoded a URL into your App then its conflicting. Ask the agent to check the URLs and for any hardcoded components, and then please run this in GPT5 to analyze and debug.
Then ask GPT5 to generate a prompt for you to ensure clean dynamic variables for your URL in the intended function
red steppe
No problematic hardcoded components found!
The issue is that your production deployment (easyviralcontent.com) is not using this codebase - it's using an older version that still has:
Old environment variables pointing to contentai-17.emergent.host
Old axios configuration with wrong Content-Type
I'm sure is going wrong with .env variables
Summary of Hardcoded Components Found:
✅ No Issues (Good):
Frontend: No hardcoded URLs in React components
Frontend .env: Correctly set to viral-ai-content-3.preview.emergentagent.com
Backend .env: Correctly set PUBLIC_BASE_URL=https://easyviralcontent.com
⚠️ Hardcoded References (Expected):
Backend CORS: Includes viral-ai-content-3.preview.emergentagent.com in allowed origins (this is good)
Backend fallbacks: Uses viral-ai-content-3.preview.emergentagent.com as fallback in error handling (this is good)
Test files: Various test files reference both URLs (not affecting production)
waxen spoke
Is specifically says it has old variables:
Old environment variables pointing to contentai-17.emergent.host
Old axios configuration with wrong Content-Type
red steppe
Yes, and I have tried instructing it to solve that, and it loops around around around
waxen spoke
Ask it to change the environmental variable to be dynamic and reference current environment, and for the axios configuration I am not sure how you set that up
red steppe
Trying that now! 🙂
I've made a Multi Scene Veo 3 Content Generator that I am testing out
red steppe
Axios error is connected to the wrong domain I believe, but having issues solving it
I'm on this page:
I try to upload an image
I get this in my console log:
And I have this in my backend .env
Alternative fallback URL for contentai-17.emergent.host
CONTENTAI_BASE_URL=https://viral-ai-content-3.preview.emergentagent.com