[962] Nuxt & Clustering via Sticky session

HI,

I am using clustering nuxt.js (v 0.10.7) via sticky-session with 6 cpu. Each worker contains the same code, and for the part related to Nuxt :

const Nuxt = require('nuxt')
 const config = require('./config/nuxt_cfg.js')
 config.dev = !isProd
 const nuxt = new Nuxt(config)

 process.env.DEBUG = 'nuxt:*'
 // Render every route with nuxt.js
 app.use(nuxt.render)

 if (config.dev) {
   nuxt.build()
   .catch((error) => {
     console.error(error)
     process.exit(1)
   })
 }

I get error messages on the server terminal just after workers' creation.

server started on 3000 port
server started on 3000 port
server started on 3000 port
server started on 3000 port
server started on 3000 port
{ Error: ENOTEMPTY: directory not empty, rmdir '/home/matthieu/Documents/Dev_HPP/helping-pong_nuxt_local/.nuxt'
    at Error (native)
  errno: -39,
  code: 'ENOTEMPTY',
  syscall: 'rmdir',
  path: '/home/matthieu/Documents/Dev_HPP/helping-pong_nuxt_local/.nuxt' }
sticky-session: worker died

I wonder how do all nuxt applications share the nuxt module folder ?

Despite these errors it seems the application is successfully compiled (cf attached file) : strange ! Any idea of the origin of these "errors" ?

Bug.txt