Random 500 internal server error in production build on page refresh

[abhi12299]

Problem: In production build, I get a random 500 server error on production when I refresh a page. It stays there for around 5-10 reloads and then goes away. Client side routing works though. This occurs in my personal website https://iabhishek.dev randomly. On inspection, the following code gives errors randomly (line 290)

Code for my website can be found at https://github.com/abhi12299/website

The following is my code for getInitialProps in / page:

The getCookie function:

The authenticate function:

The getLatestPosts function:

[timneutkens]

Not sure exactly, I can't reproduce the issue described.

---

Please follow the issue template.

https://github.com/zeit/next.js/issues/new/choose

https://github.com/zeit/next.js/issues/new?template=8.Question_about_next.md

# Question about Next.js

GitHub Issues are reserved for Bug reports and Feature requests. The best place to get your question answered is to post it on https://spectrum.chat/next-js.

[abhi12299]

Hi. If I knew how to reproduce the issue, I would have provided more details. But this happens randomly. Please read the issue description again. It just happens on production. On spectrum, the issue would be left unnoticeable. It's a big problem for me, that's why I raised this issue here.

[kodcuserkan]

Similar here any answers or links please?

[abhi12299]

@kodcuserkan I got this error for a few days without any knowledge of what happened wrong. But it was fixed without any code change after a couple of days.

[hems]

Hi. If I knew how to reproduce the issue, I would have provided more details. But this happens randomly. Please read the issue description again. It just happens on production. On spectrum, the issue would be left unnoticeable. It's a big problem for me, that's why I raised this issue here.

I get the same error here, app works fine locally but when deployed ( I'm using Heroku as a simple test ) the app loads and yields a 500 on the index page without a stack trace or any other information attached to it.

Exactly on the same place, you get the error

[hems]

I'm also using getInitialProps on my Page, also the funny thing is I check the server logs and i cant see the 500 error in the logs.

The error seems to happen without stack trace and without being posted to the console on the client and on the server.

[hems]

Not sure if it's relevant but I'm using webpackChunkName when loading 1 of the files, but that's unrelated i guess cause that file loads perfectly.

Also, i can see the index page behaves differently locally and on the deployed server.

On the deployed version on heroku I get 500 error on the index page but on the client, i don't et that error.

[hems]

Ok so turns out in my case NODE_ENV is being set to "production" even tough Heroku NODE_ENV is set to "staging".

I got to figure out what is messing this up.

[abhi12299]

@hems Did you find something regarding the issue? It was a real pain for me back then.

[hems]

@hems Did you find something regarding the issue? It was a real pain for me back then.

I installed source maps and also found a weird error because as i said my NODE_ENV was always "production" even on my staging and production environment..

https://github.com/zeit/next-plugins/tree/master/packages/next-source-maps

[JClackett]

Having a similar issue when deploying to Now, client side navigation works fine, but refreshing on certain pages causes a 500, this doesnt happen on heroku, so i assume theres something going on with the Now node environment thats causing this issue, but its hard to find any logs

[JClackett]

After checking some more, i notice that the pages that throw a 500 are the ones that have nested/mulitple HOC's around the page. Im using apollo, so am wrapping pages in the withApollo hoc but then have a few layout HOC's as well sometimes.

export default withApollo(SomePage) // Works fine

export default withApollo(withSideBar(SomeOtherPage)) // 500

[abhi12299]

@JClackett yup, with multiple HOCs I too had this problem.. maybe it has something to do with my implementation but then again, a lot of people have this problem

[erikleon]

experiencing this same problem and trying to determine if its cloudflare or next.js related but the error is very vague and no way to reproduce locally or on a QA/Staging environment

[vasc1]

yeap, I'm on this boat too

[JClackett]

I resorted to not using hocs now, bit annoying but it seems to be the cause

[bikramtuladhar]

We recently faced this kind of random 500 internal error, it turned out our API server (laravel) triggered API rate-limiting (60 request per minute). Rate limiting response was not logged as an error which made difficult to track the real issue.

[taivmo]

Hi, maybe this might help ! I create a file name : next.config.js in root folder (next to package.json) then in next.config.js file, I add : module.exports = { trailingSlash: true, } Hope this can help someone!

[balazsorban44]

This issue has been automatically locked due to no recent activity. If you are running into a similar issue, please create a new issue with the steps to reproduce. Thank you.