ci: limit the slowest modules with awk, not head, so sort does not fail the build

[davsclaus]

Claude Code on behalf of davsclaus

Related: CAMEL-24777 (assigned to gnodet; this PR does not take the ticket over, see the root-cause comment).

PR builds have been failing after Maven finished successfully:

Maven build completed with exit code: 0
sort: write error
##[error]Process completed with exit code 2.

e.g. 27150, 27147.

render_top_slowest_modules in reactor_timing.sh (added in #26147) pipes sort into head -n "$limit". incremental-build.sh ignores SIGPIPE (trap '' PIPE, #26528), so once head has its lines and closes the pipe, sort gets EPIPE, prints write error and exits 2; with pipefail and set -e that fails the step. Without the trap sort is killed by SIGPIPE instead (exit 141), which fails the same way, so reverting #26528 would not help.

The limit is now taken with awk -v n="$limit" 'NR <= n', which reads all of its input.

Tests

reactor_timing_test.sh gets a case that runs a 50000-module reactor through the function with SIGPIPE ignored. With LC_ALL=C: 26 passed with the fix; with head restored the new case fails (2 failures). (Without LC_ALL=C three existing duration tests fail on a comma-decimal locale, on main as well.)

🤖 Generated with Claude Code

[github-actions]

🌟 Thank you for your contribution to the Apache Camel project! 🌟
🤖 CI automation will test this PR automatically.

🐫 Apache Camel Committers, please review the following items:

• First-time contributors require MANUAL approval for the GitHub Actions to run
• You can use the command /component-test (camel-)component-name1 (camel-)component-name2.. to request a test from the test bot although they are normally detected and executed by CI.
• You can label PRs using skip-tests and test-dependents to fine-tune the checks executed by this PR.
• Build and test logs are available in the summary page. Only Apache Camel committers have access to the summary.

⚠️ Be careful when sharing logs. Review their contents before sharing them publicly.

[gnodet-bot]

Solid fix. Replacing head -n with awk 'NR <= n' correctly eliminates the SIGPIPE/EPIPE race: awk consumes all of sort's output so the pipe never breaks, regardless of the trap '' PIPE in the parent script. Edge cases (limit=0, limit > input, empty input) are fine. The 50k-module test covers the exact failure mode.

This review was generated by an AI agent, Hermès on behalf of @gnodet.