TLDR:

Why is it happening? Cache
Can you solve it completely? No
Can you solve it partially? Yes
Should/Must you solve it? Not really

This happens most of the time because the browser caches the index.html (and a few other css and js) file of the webapp. Try turning off caching for the file by adding these to the nginx config:

location ~* \.html?$ {
  expires -1;
  add_header Pragma "no-cache";
  add_header Cache-Control "no-store, must-revalidate";
}

If you want to trigger a reload on the client side for this kind of errors, try this on your frontend app's router:

router.onError(error => {
  if (/loading chunk \d* failed./i.test(error.message)) {
    window.location.reload()
  }
})

Caution: possibility of infinite reload if reload does not resolve the issue

Credit: Read more technical steps here on this article that already explains a lot more than I planned to write. Learn more about different chunk errors here.

Facts

• Not all chunk errors are of same type, or for same reason.
• Reachable chunks does not mean they can be loaded. (reach !== load)
• Sometimes developers confuse syntax errors with chunk errors.
• Sometimes it can happen due to missing env variables.

My assumption on OPs description:

• It always should happen after e production release, but you might get reports at random times, because not all the users visit within a close timeframe after a release.
• It is mostly iOS, because iOS users don't usually clear their recent apps (or browser) and whenever they revisit, their session is mostly idle and tries to access the cache, other OS/Devices let's them clear the apps and they visit from a new session more often (compared to iOS).