nextjs - 💡(How to fix) Fix Next.js build fails with "WebAssembly.instantiate(): Out of memory" on shared hosting (O2switch / cPanel)

Link to the code that reproduces this issue

https://github.com/vercel/next.js/discussions/32001

To Reproduce

Deploy a Next.js application to a shared hosting environment using cPanel (O2switch in this case). Connect to the server via SSH or cPanel terminal. Install dependencies with npm install. Run npm run build (which executes next build

Current vs. Expected behavior

The build fails with the following error:

error - Failed to load next.config.js

Build error occurred RangeError: WebAssembly.instantiate(): Out of memory: wasm memory at internal/deps/cjs-module-lexer/dist/lexer.js

This suggests a memory limitation during the build process.

Expected behavior: The application should build successfully, as it does locally and in other environments.

Provide environment information

Next.js version: latest (also tested with several recent versions)
Node.js version: 18+
Hosting: O2switch (shared hosting with cPanel / Phusion Passenger)
OS: Linux (server environment)
Build works locally: yes
Build works on other platforms (e.g. VPS / Vercel): yes

Which area(s) are affected? (Select all that apply)

Not sure

Which stage(s) are affected? (Select all that apply)

Other (Deployed)

Additional context

This issue appears to be related to memory limits imposed by the hosting environment rather than the application itself. Several users report similar problems when building Next.js apps on shared hosting platforms using cPanel.

Workarounds such as building locally and uploading the .next folder seem to partially solve the issue, but they are not ideal for CI/CD workflows.

Additionally, some users report that using SSH instead of the cPanel terminal reduces limitations, but the issue can still occur depending on available resources.

It would be helpful to know:

Whether Next.js build has specific memory requirements that could be optimized for constrained environments If there are recommended configurations or flags for low-memory environments Or if this scenario is considered unsupported by design

Thanks in advance for your help.