A number of environment variables can be used to configure the behavior of the library. There are also functions to configure this behavior in code. The environment variables are convenient when troubleshooting problems.
Endpoint Overrides
GOOGLE_CLOUD_CPP_AUTOML_PREDICTION_SERVICE_ENDPOINT=...overrides theEndpointOption(which defaults to "automl.googleapis.com") used byMakePredictionServiceConnection().
GOOGLE_CLOUD_CPP_AUTO_ML_ENDPOINT=...overrides theEndpointOption(which defaults to "automl.googleapis.com") used byMakeAutoMlConnection().
GOOGLE_CLOUD_CPP_ENABLE_TRACING=rpc: turns on tracing for most gRPC calls. The library injects an additional Stub decorator that prints each gRPC request and response. Unless you have configured your own logging backend, you should also setGOOGLE_CLOUD_CPP_ENABLE_CLOGto produce any output on the program's console.
See Also
google::cloud::TracingComponentsOptionGOOGLE_CLOUD_CPP_TRACING_OPTIONS=...: modifies the behavior of gRPC tracing, including whether messages will be output on multiple lines, or whether string/bytes fields will be truncated.
See Also
google::cloud::GrpcTracingOptionsOptionGOOGLE_CLOUD_CPP_ENABLE_CLOG=yes: turns on logging in the library, basically the library always "logs" but the logging infrastructure has no backend to actually print anything until the application sets a backend or they set this environment variable.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-09-04 UTC."],[[["\u003cp\u003eThe webpage lists various versions of the AutoML C++ library, ranging from version 2.11.0 up to the latest release candidate 2.37.0-rc, with each version linking to its respective documentation.\u003c/p\u003e\n"],["\u003cp\u003eEnvironment variables like \u003ccode\u003eGOOGLE_CLOUD_CPP_AUTOML_PREDICTION_SERVICE_ENDPOINT\u003c/code\u003e and \u003ccode\u003eGOOGLE_CLOUD_CPP_AUTO_ML_ENDPOINT\u003c/code\u003e can be used to override the default endpoints for AutoML Prediction and AutoML services, respectively.\u003c/p\u003e\n"],["\u003cp\u003eTracing for gRPC calls can be enabled using \u003ccode\u003eGOOGLE_CLOUD_CPP_ENABLE_TRACING=rpc\u003c/code\u003e, and additional logging output can be achieved by setting \u003ccode\u003eGOOGLE_CLOUD_CPP_ENABLE_CLOG\u003c/code\u003e, allowing for better troubleshooting and monitoring.\u003c/p\u003e\n"],["\u003cp\u003eThe behavior of gRPC tracing can be modified through \u003ccode\u003eGOOGLE_CLOUD_CPP_TRACING_OPTIONS\u003c/code\u003e, providing control over output formatting and field truncation, which can be adjusted to fit individual needs.\u003c/p\u003e\n"],["\u003cp\u003eThe \u003ccode\u003eGOOGLE_CLOUD_PROJECT\u003c/code\u003e environment variable is used for GCP project configuration in examples and tests, though it does not directly impact the library's functionality.\u003c/p\u003e\n"]]],[],null,["Version 2.14.0keyboard_arrow_down\n\n- [2.42.0-rc (latest)](/cpp/docs/reference/automl/latest/automl-env)\n- [2.41.0](/cpp/docs/reference/automl/2.41.0/automl-env)\n- [2.40.0](/cpp/docs/reference/automl/2.40.0/automl-env)\n- [2.39.0](/cpp/docs/reference/automl/2.39.0/automl-env)\n- [2.38.0](/cpp/docs/reference/automl/2.38.0/automl-env)\n- [2.37.0](/cpp/docs/reference/automl/2.37.0/automl-env)\n- [2.36.0](/cpp/docs/reference/automl/2.36.0/automl-env)\n- [2.35.0](/cpp/docs/reference/automl/2.35.0/automl-env)\n- [2.34.0](/cpp/docs/reference/automl/2.34.0/automl-env)\n- [2.33.0](/cpp/docs/reference/automl/2.33.0/automl-env)\n- [2.32.0](/cpp/docs/reference/automl/2.32.0/automl-env)\n- [2.31.0](/cpp/docs/reference/automl/2.31.0/automl-env)\n- [2.30.0](/cpp/docs/reference/automl/2.30.0/automl-env)\n- [2.29.0](/cpp/docs/reference/automl/2.29.0/automl-env)\n- [2.28.0](/cpp/docs/reference/automl/2.28.0/automl-env)\n- [2.27.0](/cpp/docs/reference/automl/2.27.0/automl-env)\n- [2.26.0](/cpp/docs/reference/automl/2.26.0/automl-env)\n- [2.25.1](/cpp/docs/reference/automl/2.25.1/automl-env)\n- [2.24.0](/cpp/docs/reference/automl/2.24.0/automl-env)\n- [2.23.0](/cpp/docs/reference/automl/2.23.0/automl-env)\n- [2.22.1](/cpp/docs/reference/automl/2.22.1/automl-env)\n- [2.21.0](/cpp/docs/reference/automl/2.21.0/automl-env)\n- [2.20.0](/cpp/docs/reference/automl/2.20.0/automl-env)\n- [2.19.0](/cpp/docs/reference/automl/2.19.0/automl-env)\n- [2.18.0](/cpp/docs/reference/automl/2.18.0/automl-env)\n- [2.17.0](/cpp/docs/reference/automl/2.17.0/automl-env)\n- [2.16.0](/cpp/docs/reference/automl/2.16.0/automl-env)\n- [2.15.1](/cpp/docs/reference/automl/2.15.1/automl-env)\n- [2.14.0](/cpp/docs/reference/automl/2.14.0/automl-env)\n- [2.13.0](/cpp/docs/reference/automl/2.13.0/automl-env)\n- [2.12.0](/cpp/docs/reference/automl/2.12.0/automl-env)\n- [2.11.0](/cpp/docs/reference/automl/2.11.0/automl-env) \n\nEnvironment Variables\n=====================\n\nA number of environment variables can be used to configure the behavior of the library. There are also functions to configure this behavior in code. The environment variables are convenient when troubleshooting problems.\n\nEndpoint Overrides\n------------------\n\n- `GOOGLE_CLOUD_CPP_AUTOML_PREDICTION_SERVICE_ENDPOINT=...` overrides the `EndpointOption` (which defaults to \"automl.googleapis.com\") used by `MakePredictionServiceConnection()`.\n- `GOOGLE_CLOUD_CPP_AUTO_ML_ENDPOINT=...` overrides the `EndpointOption` (which defaults to \"automl.googleapis.com\") used by `MakeAutoMlConnection()`.\n\n###### See Also\n\n[`google::cloud::EndpointOption`](https://cloud.google.com/cpp/docs/reference/common/latest/structgoogle_1_1cloud_1_1EndpointOption.html)\n\nLogging\n-------\n\n`GOOGLE_CLOUD_CPP_ENABLE_TRACING=rpc`: turns on tracing for most gRPC calls. The library injects an additional Stub decorator that prints each gRPC request and response. Unless you have configured your own logging backend, you should also set `GOOGLE_CLOUD_CPP_ENABLE_CLOG` to produce any output on the program's console.\n\n###### See Also\n\n[`google::cloud::TracingComponentsOption`](https://cloud.google.com/cpp/docs/reference/common/latest/structgoogle_1_1cloud_1_1TracingComponentsOption.html)`GOOGLE_CLOUD_CPP_TRACING_OPTIONS=...`: modifies the behavior of gRPC tracing, including whether messages will be output on multiple lines, or whether string/bytes fields will be truncated.\n\n###### See Also\n\n[`google::cloud::GrpcTracingOptionsOption`](https://cloud.google.com/cpp/docs/reference/common/latest/structgoogle_1_1cloud_1_1GrpcTracingOptionsOption.html)`GOOGLE_CLOUD_CPP_ENABLE_CLOG=yes`: turns on logging in the library, basically the library always \"logs\" but the logging infrastructure has no backend to actually print anything until the application sets a backend or they set this environment variable.\n\n###### See Also\n\n[`google::cloud::LogBackend`](https://cloud.google.com/cpp/docs/reference/common/latest/classgoogle_1_1cloud_1_1LogBackend.html)\n\n###### See Also\n\n[`google::cloud::LogSink`](https://cloud.google.com/cpp/docs/reference/common/latest/classgoogle_1_1cloud_1_1LogSink.html)\n\nSetting the Default Project\n---------------------------\n\n`GOOGLE_CLOUD_PROJECT=...`: is used in examples and integration tests to configure the GCP project. This has no effect in the library."]]
