Max Golovanov
Max Golovanov
I want to mention that I added a sample how the fields can be populated via Semantic Context API. Technically if PAL logic that obtains these values is commented out...
@peacenix @lalitb There was a limit on the total number of concurrent simultaneous HTTPS connections that SDK makes to a single HTTP server in production. This is configured [here](https://github.com/microsoft/cpp_client_telemetry/blob/417dafc025fa2083a411522e57e39e258c03026b/lib/config/RuntimeConfig_Default.hpp#L22) ....
I think the fix makes sense in OpenTelemetry. But it's not exactly clear why we don't get the higher-level TPM limit to apply in 1DS. In 1DS C++ SDK the...
@lalitb @nishchith-cp - I have a change that'd expose this via C API, verifying in a test that `ext.os.ver` is populated. You can also use C++ semantic context pattern that'd...
@lalitb - we need it, but we identified an extra scenario that's not covered by the patch. I'll do another iteration on it.
It's sort of design intent. There used to be `EVT_SENT` constant in the older Aria SDK. The new SDK keeps track of the moment the batch is sent for async...
Can I ask so that when we implement this, we can specify what versions of Android deps to use via env vars. My team needs this layer to be compatible...
Current outstanding issue: - OpenTelemetry is now Released. But all of its released assemblies are still always labeled `v1.0.0.0`. - This will cause a **runtime crash** for anything that relies...
@reyang @cijothomas > Essentially, I believe a flexible range such as ">=v1.6.0-beta" should never be used. Now that the OpenTelemetry v1.8.0 is released and its assembly version remains constant `1.0.0.0`...
> > OpenTelemetry is now Released. But all of its released assemblies are still always labeled v1.0.0.0 > > Sorry, I didn't fully analyze this issue, but just want to...