Hello Vercel Support Team,
I’d like to provide an important update regarding the production issue previously reported.
Key detail:
The NOT_FOUND error ID changes every time the issue occurs, for example:
-
gru1::lgqck-1770151655454-a4f4162e13ec -
gru1::vtvrd-1770151741831-7296466c5fb2
This confirms the problem is not deterministic and strongly points to edge routing or caching inconsistency, not application logic.
Observed behavior:
-
The same dynamic page loads successfully
-
Minutes later or after a refresh, it returns:
404: NOT_FOUND Code: NOT_FOUND Region: gru1 -
No new deployment occurs between successful access and the 404
-
The route and data still exist
Because the error ID changes, it indicates the request is being handled by different edge instances, some of which incorrectly respond with NOT_FOUND.
Why this is critical:
-
This breaks a core production route
-
Causes user-facing errors and SEO damage
-
Makes the platform unreliable
Request:
Please investigate:
-
Edge cache invalidation or corruption
-
ISR / routing resolution for dynamic routes
-
Region-specific behavior in
gru1 -
Any known issues causing intermittent 404s on existing pages
This is an urgent production issue affecting real users.
Thank you for your assistance,