Do not abort XHR's on unload

[brian-mann]

Current behavior:

Whenever Cypress encounters the page unloading, it will automatically call xhr#abort on all open XHR's. This is unnecessary as the browser will automatically abort it under the hood.

Aborting it manually is not only incorrect, but this will trigger the native abort() handlers which can cause errors in the client application.

Desired behavior:

Allow the native browser behavior to propagate and simply update the UI state to match what would be seen by the Network tab in dev tools.

[arpowers]

related to #2374 ?

[jennifer-shehane]

@arpowers I can't be sure this is related to that issue since we were never able to reproduce #2374, but you're welcome to pull down the PR and see if the related issue is fixed. PR: #2969

[jennifer-shehane]

The code for this is done, but this has yet to be released. We'll update this issue and reference the changelog when it's released.

[brian-mann]

Released in 3.1.4.

[janhjordie]

Hey @brian-mann
I still have same issue in Electron. Also with the new 3.1.4.
And it both visual and headless Electron. It works just perfect in Chrome and Canary.
I'm running MacOS.

I reported this issue ealier #1652 (comment)

In my case is a bunch of SignalR request that runs in the application, and not any I actually use.
One of the SignalR request gets aborted and then stops the whole test

Please advise !

[sverrelorenzen]

Hi,

I have the same issue with Electron as @janhjordie.

Regards

[jennifer-shehane]

Hey @janhjordie and @sverrelorenzen, could you open a new issue detailing a reproducible example of your issue (app code + test code). Thank you!

[13322211021]

Got the same issue with Electron as @janhjordie.
Did you get a workaround?@janhjordie, @sverrelorenzen
I use cypress 3.4.1.

[jennifer-shehane]

This issue will be closed to further comment as the exact issue here was resolved and tested.

If you're experiencing a bug similar to this in Cypress, please open a new issue with a fully reproducible example that we can run. There may be a specific edge case with the issue that we need more detail to fix.