flb_metrics: change hot_reloaded_times from gauge to counter

[eschabell]

Registering it as a gauge causes incorrect behavior with PromQL
functions like rate() and increase(), which expect counters.

• src/flb_metrics.c: replace cmt_gauge_create() with cmt_counter_create() for hot_reloaded_times in attach_hot_reload_info()
• src/flb_metrics.c: replace struct cmt_gauge with struct cmt_counter
• src/flb_metrics.c: replace cmt_gauge_set() with cmt_counter_set()

Fixes #11479.

---

Testing

• [ N/A ] Example configuration file for the change
• [ N/A ] Debug log output from testing the change
• [ N/A ] Attached Valgrind output that shows no leaks or memory corruption was found

If this is a change to packaging of containers or native binaries then please confirm it works for all targets.

• [ N/A ] Run local packaging test showing all targets (including any new ones) build.
• [ N/A ] Set ok-package-test label to test for all targets (requires maintainer to do).

Documentation

• [ YES ] Documentation required for this feature

Docs PR for this change: fluent/fluent-bit-docs#2383

Backporting

• [ N/A ] Backport to latest stable release.

---

Fluent Bit is licensed under Apache 2.0, by submitting this pull request I understand that this code will be released under the terms of that license.

Summary by CodeRabbit

• Bug Fixes
  • Updated hot reload metric tracking to use more appropriate metric type for better accuracy in monitoring.

[coderabbitai]

📝 Walkthrough

Walkthrough

Changed the hot reload metric from a gauge to a counter in flb_metrics.c. The implementation now uses cmt_counter_create() instead of cmt_gauge_create() and updates the metric with cmt_counter_set() instead of cmt_gauge_set(), maintaining the same observable value source and hostname dimension handling.

Changes

Cohort / File(s)	Summary
Metric Type Implementation src/flb_metrics.c	Changed fluentbit_hot_reloaded_times metric from gauge to counter type. Replaced cmt_gauge_create() with cmt_counter_create(), added counter creation result check, and switched metric update from cmt_gauge_set() to cmt_counter_set().

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Suggested reviewers

• edsiper

Poem

🐰 A gauge became a counter bright,
Hot reloads now counted right!
No more confusion, up it goes,
Like my fluffy tail's proud pose! 🔄

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: converting the hot_reloaded_times metric from a gauge to a counter implementation.
Linked Issues check	✅ Passed	The code changes directly address issue #11479 by replacing cmt_gauge with cmt_counter for the hot_reloaded_times metric [#11479].
Out of Scope Changes check	✅ Passed	All changes in src/flb_metrics.c are directly scoped to converting the hot_reloaded_times metric type as required by the linked issue.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

src/flb_metrics.c (1)

367-370: Pre-existing: return values of all attach_* helpers are discarded.

All four attach_* calls return -1 on metric-creation failure, but those errors are swallowed here. This is not introduced by this PR and is out of scope, but worth tracking separately since a failed cmt_counter_create for any metric goes completely unnoticed.

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@src/flb_metrics.c` around lines 367 - 370, The four helper calls
attach_uptime, attach_process_start_time_seconds, attach_build_info, and
attach_hot_reload_info can return -1 on failure but their results are ignored
here; update the caller to check each return value and propagate or handle
errors (e.g., if any helper returns -1, immediately return -1 from the
surrounding function or log and clean up as appropriate) so a failed
cmt_counter_create is not silently swallowed—ensure you reference those helper
functions by name when adding the checks and propagate the error consistently.

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@src/flb_metrics.c`:
- Line 346: The call to cmt_counter_set in attach_hot_reload_info (and likewise
in attach_uptime) currently drops the return value, hiding failures when
allow_reset==0; update the code to check the int return from cmt_counter_set and
handle errors: if it returns -1 log a debug/error with context (function name,
hostname/metric name and the attempted val) or propagate the error upward as
appropriate so failures are observable; locate calls to cmt_counter_set in
attach_hot_reload_info and attach_uptime and add the simple return-value check
plus logging/propagation.

---

Nitpick comments:
In `@src/flb_metrics.c`:
- Around line 367-370: The four helper calls attach_uptime,
attach_process_start_time_seconds, attach_build_info, and attach_hot_reload_info
can return -1 on failure but their results are ignored here; update the caller
to check each return value and propagate or handle errors (e.g., if any helper
returns -1, immediately return -1 from the surrounding function or log and clean
up as appropriate) so a failed cmt_counter_create is not silently
swallowed—ensure you reference those helper functions by name when adding the
checks and propagate the error consistently.

---

ℹ️ Review info

Configuration used: defaults

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between bc966ed and 61140f0.

📒 Files selected for processing (1)

• src/flb_metrics.c

[eschabell]

@cosmo0920 or @lecaros the linting errors are on windows build files, can someone look at this change and review?