Adding explicit tls roots configuration to stackdriver

[msdrigg]

Fixes #95

Changes

When upgrading to tonic 0.12, tonic stopped implicitly setting up tls roots even when the features are enabled for it. This PR makes this setup explicit in stackdriver.

It doesn't change user-facing API and the change is relatively trivial. I manually tested this in my repo and the error went away. I don't think there is a good way to unit test this unless we want the unit tests to verify connectivity to gcp servers.

Merge requirement checklist

• CONTRIBUTING guidelines followed
• Unit tests added/updated (if applicable)
• Appropriate CHANGELOG.md files updated for non-trivial, user-facing changes
• Changes in public API reviewed (if applicable)

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: msdrigg / name: Scott Driggers (ccd6828)
• ✅ login: lalitb / name: Lalit Kumar Bhasin (127827a)

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 6 lines in your changes missing coverage. Please review.

Project coverage is 56.4%. Comparing base (56a6f45) to head (127827a).
Report is 26 commits behind head on main.

Files with missing lines	Patch %	Lines
opentelemetry-stackdriver/src/lib.rs	0.0%	6 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##            main     #96     +/-   ##
=======================================
+ Coverage   52.3%   56.4%   +4.1%     
=======================================
  Files         38      39      +1     
  Lines       4967    5446    +479     
=======================================
+ Hits        2598    3074    +476     
- Misses      2369    2372      +3     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[msdrigg]

So the only point of discussion I see is that with this fix: If both webpki-roots and native-roots features are enabled then here I default to webpki roots. I don't know if this is different from the previous tonic implicit behavior. Another option would be to error out if both are enabled.

[msdrigg]

I see a generated_code_is_fresh test is failing, what should I do to fix this?

[fiadliel]

@msdrigg The changed code needs to be committed. I opened #102 to do the PR part of this. Hopefully somebody will eventually review it 😄

[msdrigg]

Are there any tests that need to be written for this? I have been using this in my own project for the past month and it has resolved the errors for me.

[lalitb]

I think the changes are trivial, and we can add the tests afterwards if required.

[msdrigg]

Could we get a release for this?

[lalitb]

Could we get a release for this?

There is otel-rust release planned during this week. We can have release from contrib repo subsequently.