Display When There is a Timeout in the Network Tab in Replay

[jas-kas]

When auditing mobile replay for an iOS+Swift UI mobile application, there was a timeout that occurred when running the mobile app. In Sentry's Replay Details > Network UI, we identify it as 'unknown' status code, but there's no clear indication that a timeout has happened.

See this replay.

[ryan953]

looking at the replay i see two network requests
each request took about 1min. 60,000ms = 60s
and in the errors tab there are 4 error: The request timed out
the timestamps first and last error in the list match with the 2 requests i saw in the network tab: the error happened about 1 minute after the request was started

there is no statusCode value in the json. We’re showing -- and unknown in the ui depending on where you look.

i think we can also show -- for cached responses.

---

Options:

It's not accurate to assume that all requests over a certain duration have timed out. Long polling operations are designed to work like that, and streaming results too will take a while to return all their data.

One thing we could think about is color-coding durations, highlighting fast/med/slow requests within the replay. This is similar to just sorting the table by duration though and so might not be a great idea.

Since there are errors captured, it might be feasible for the SDK to link up the error and the network request. It looks like this isn't happening, even within Issue Details. On Issue Details all i can see connecting the two bits of data is a Network Request Breadcrumb.

A third option could be to better highlight the Errors tab inside Replay Details when there are errors inside it. That's the best place to highlight actual problems, whereas the Network tab shows more raw-values that help tell the story.