Expose logs for symbolication and source context resolution

[jamescrosswell]

Problem Statement

When you can't see symbols or source context for events after uploading debug companion files and source code using the Sentry CLI, there's no easy way to determine what the problem might be.

sentry-cli can be run with debug verbosity, which helps resolve issues uploading files. The debug files area of the sentry project settings gives an overview of which files have been uploaded (and whether symbols/source are present). However, if symbols and source are successfully uploaded but Sentry is still not able to resolve the symbols/source for events, there is no information available to SDK users letting them know why (or how to fix this).

In this discussion thread, for example, I've been going back and forth with a user to try to get any kind of symbolication working for a sample app that he provided. After 3 days, we finally managed to get it working just by trial and error and, even then, using settings that are suboptimal and inappropriate for a production application.

Solution Brainstorm

When showing a stack trace for which Sentry is unable to resolve symbols or source context, it would be great if there was some way to view some "logs" for the process of resolving symbols/source that could indicate why resolution is failing.

Currently the ... menu lets you force the stack trace to be unsymbolicated. That menu could include an option to show "Symbolication Diagnostics" and/or "Source Context Diagnostics".

This might also be a good launching pad into some kind of tutorial to show people how to enrich their stack traces with symbolication and source context, if they're not aware of those features of Sentry.

Product Area

Unknown

[getsantry]

Auto-routing to @getsentry/product-owners-issues for triage ⏲️

[mrduncan]

@getsentry/product-owners-processing what information (if any) do we currently have available to show to users?

[getsantry]

Routing to @getsentry/product-owners-processing for triage ⏲️

[bruno-garcia]

cc @olksdr for processing and @matejminar for ideas of the UI
I believe we might have context on the issue to why it failed, we just don't show it?