execjs runs at initial request after deploy

[hopewise]

I am using AWS ECS to deploy my app, the image I am using has no nodejs, therefore, I am pre compiling at the host container, and S3 sync works fine.

However, when requesting the app, execjs starts and nodejs processes start working for a long time, I am not sure if it actually rebuild the assets at production server??

How can I prevent this? Can you please check here: https://stackoverflow.com/questions/63895516/why-nodejs-process-starts-when-requesting-rails-app

[PikachuEXE]

I don't understand when requesting the app part What client did you use and what types of resource are requested? You might want to try requesting using different clients (e.g. curl, Postman) And also try requesting different types of resources (HTML, CSS, JS, image)

I think we need those results to understand & analyze your issue

[hopewise]

when requesting the app I mean when opening the homepage in browser, the execjs get started and causes high CPU usage for a long time.

If I use curl, nodejs command will start:

PID USER      PR  NI    VIRT    RES    SHR S %CPU %MEM     TIME+ COMMAND
2313 app       20   0  950292  51048  19448 R 47.4  2.5   0:00.70 nodejs

ps aux
app       2694 34.3  4.5 978576 92884 ?        Rl   09:45   0:01 nodejs /tmp/execjs20200916-368-uzpmawjs

For curl command, first time, I get 504 gateway time-out , then, second curl, I get:

HTTP/2 200 
date: Wed, 16 Sep 2020 10:04:25 GMT
content-type: text/html; charset=utf-8
status: 200 OK
cache-control: max-age=0, private, must-revalidate
vary: Accept-Encoding
referrer-policy: strict-origin-when-cross-origin
x-permitted-cross-domain-policies: none
x-xss-protection: 1; mode=block
x-request-id: 36365117-7db6-4f82-a79f-d575fd6ce0bd
x-download-options: noopen
etag: W/"f4bb89ba5383f9e1d90e43aef5bc8a67"
x-frame-options: SAMEORIGIN
x-runtime: 0.019307
x-content-type-options: nosniff
set-cookie: XSRF-TOKEN=tNkG8EGyxBvADeySy5WgOTljAx4pbhZf0uH1pRPIMDczf5gPl5NBQNsHL6kti6aEs4MIRqk3fRARqqkVWHAQ5w%3D%3D; path=/
set-cookie: _dca_session=bVJMUmVIcjZzK0FmM1pFNXFqbXhldnNhVXhCYkREWnFKRkFWQno5TFNqMHRiU2VPcHRMa2VFcTZPbHJ0T1dQZUJGM2VMSnk4bU83bzJRT2VaZDZtYktMTTRPRW40Q2c1eGUvWklEZ3k2M1gwZk9OK3dkSXBET3g0bWlqcmdYQ050T1ZWNWVGNVVTYmliUVJYbVdzS2x4ajI3S3dOVU94VSt3VllSVzduMCthL1UxYVNIaTN3L2EwVEdDVFk4L1lMLS1nejU4SzJab0cvcFE1OTRPYlIrMmZnPT0%3D--3d67f57099f5c064f29adbda9a1af1ddf203602d; path=/; expires=Wed, 16 Sep 2020 12:04:25 GMT; HttpOnly
x-powered-by: Phusion Passenger 6.0.6
server: nginx/1.14.0 + Phusion Passenger 6.0.6

[PikachuEXE]

Do you use asset pipeline/webpacker/other methods to manage JS assets? But anyway it seems the app might be compiling the JS when HTML is requested You should try to see what assets (get the URLs, try JS assets first) are requested and try to request those URLs to see what happen Could be

• Some assets missing (?)
• Wrong config about asset for prod
• Other unknown issues

[hopewise]

I use webpacker, it seems I need to disable server side rendering? But it's not clear how to do it.

Thanks for your suggestions, I will try it.

[PikachuEXE]

I use webpacker too but I don't use asset sync anymore CDN is used for load assets from the app directly However if webpacker is compiling assets on prod that's probably something to do with config/setup instead of asset sync

[hopewise]

Thanks, I am following up at webpacker.