Capture proxy related environment variables at Curl client construction to avoid crash

[xlamorlette-datadog]

Description

Read the proxy related environment variables when the Curl client is created (in CurlImpl constructor), and, later, pass the values to libcurl explicitly on each request (in CurlImpl::post()).

Motivation

In some (new) tests for the Datadog Nginx module, we get crashes during shutdown.
This is because the tracer's libcurl background thread reads proxy environment variables via getenv() on every new connection. But when Nginx shuts down, it changes its worker's process title, which rewrites its environment. This leads to a segfault due to a race condition.

Additional Notes

This mirrors the approach used in dd-trace-php (in ddtrace_comes_minit_proxy_env()).

I cleaned a bit the surrounding code, notably by extracting CurlImpl::configure_handle() from CurlImpl::post().

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-06-12 18:31:48

Comparing candidate commit 6d126a4 in PR branch xlamorlette/fix-curl-getenv-crash with baseline commit d0605a7 in branch main.

Found 3 performance improvements and 0 performance regressions! Performance is the same for 5 metrics, 0 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

scenario:BM_TraceID_ParseHex/128bit

• 🟩 execution_time [-8.852ns; -8.692ns] or [-6.588%; -6.469%]

scenario:BM_TraceID_ParseHex/64bit

• 🟩 execution_time [-5.826ns; -5.803ns] or [-7.758%; -7.728%]

scenario:BM_TraceTinyCCSource

• 🟩 execution_time [-2.058ms; -1.654ms] or [-2.627%; -2.111%]

[zacharycmontoya]

LGTM with one small refactor comment