500 Server error, but not for everybody
EDIT
Got saved on cloudflare, was obviously my fault. Thanks for reading anyway š
Hello all,
I have a problem with the production of this website : https://launchbase.fr/
Maybe you have a 500 visiting it, maybe not, and I have no idea why......
For me right now, https://launchbase.fr/ works, but https://launchbase-3cl.pages.dev/ was a 500 for everybody (until I bulk redirected)
If I try ANY build, I have a 500 : https://d93ae5f9.launchbase-3cl.pages.dev/ (which sounds weird because well... it was working at some point at least ^^")
The weird part is now :
No issue locally (ofc)
But not issue either on the preproduction, which is a separate cf Pages but from the same github (different branch, bot merged right now so the exact same) : https://launchabse.preprod.tech/
Can't be a data problem, my back is on adonisjs elsewhere, it's just a SSR front with nuxt. Server is up and happy (and even when I turn it off or have a problem with it, it's clearly not the error I have ^^")
I'm starting to think it's a cf problem and will be solved automatically soon T_T
If I steam the logs in Cloudflare -> deployment -> details -> functions, I have this (see image)
I posted the same message on cloudflare pages help on discord because I have no idea if I'm facing a nuxt issue or a cf issue :/
Thanks for your help š
2 Replies
Hi @Omnislash. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
----------------------
Based on the information in the knowledge sources, intermittent or environment-specific 500 errors in Nuxt production deploymentsāespecially on platforms like Cloudflare Pagesācan be caused by several factors:
1. CDN/Cache Issues: If some users see a 500 error and others do not, it may be due to CDN caching or serving outdated assets. For example, if a new deployment invalidates old JS chunk URLs, users (or bots) with cached HTML referencing old chunks may get 500 errors when those chunks are missing. This is a known issue, and Nuxt has built-in chunk loading error handling, but regressions or CDN misconfiguration can still cause problems. If refreshing does not solve the issue, it may indicate the CDN is serving the wrong version or with incorrect headers, causing a mismatch between HTML and JS assetsĀ 500 | failed to fetch dynamically imported module error.
2. Cloudflare Settings: Certain Cloudflare optimizations can interfere with Nuxt SSR apps. You should disable "Rocket Loaderā¢", "Mirage", and "Email Address Obfuscation" in your Cloudflare dashboard to prevent unwanted script injection or hydration errorsĀ CDN Proxy.
3. Deployment Preset and Configuration: When deploying to Cloudflare Pages, you should use the correct Nitro preset (
cloudflare_pages) and ensure your build process matches the platform's requirements. If you use Nuxt Content, you must also connect a D1 database as required by the moduleĀ Cloudflare Pages.
4. Dependency or Build Issues: Sometimes, issues only appear in production due to differences in environment or missing dependencies. Make sure all dependencies are installed and up to date, and that your build process is clean (remove node_modules and package-lock.json before reinstalling).
5. Chunk Loading and Crawler Errors: There are known issues where crawlers or bots trigger 500 errors due to missing JS chunks after a deployment. Nuxt has implemented fixes to suppress these errors for bots, but if you see them for real users, it may be a sign of a deeper deployment or caching issueĀ suppress client-side errors by crawlers.
What you can do:
- Double-check your Cloudflare settings as above.
- Ensure your deployment uses the correct Nitro preset and all required bindings (like D1 for Nuxt Content).
- Clear CDN and browser caches to ensure users get the latest assets.
- If possible, disable aggressive caching or set up cache-busting headers for your assets.
- Review your deployment logs for any missing dependencies or build errors.
If the problem persists only on Cloudflare Pages and not locally or on other environments, it is likely related to CDN caching, Cloudflare-specific settings, or a mismatch between deployed assets and what is being served. The knowledge sources do not provide a single definitive fix, but these are the most common causes and solutions for the symptoms you describe.
----------------------
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
----------------------<@305772639438503936> Kapa.ai is still learning and improving, please let me know how I did by reacting below.