Fluentd
gc_stat
The in_gc_stat Input plugin periodically emits the statistics of the Ruby garbage collector of the running Fluentd process. Each record is the value returned by GC.stat.
It is included in Fluentd's core.
This plugin is a diagnostic tool to investigate a problem of Fluentd itself, such as a memory leak or an unexpected slowdown. Do not configure it in a production environment. Enable it only while you are collecting the data for a specific investigation, and remove it once the investigation is done.
Example Configuration
<source>
@type gc_stat
tag gc_stat
emit_interval 10
</source>
<match gc_stat>
@type stdout
</match>
Refer to the Configuration File article for the basic structure and syntax of the configuration file.
Plugin Helpers
Parameters
See Common Parameters.
@type (required)
The value must be gc_stat.
tag (required)
| type | default | version |
|---|---|---|
| string | nil | 0.10.16 |
The tag assigned to the emitted events.
emit_interval
| type | default | version |
|---|---|---|
| time | 60 | 0.10.16 |
The interval between the emits.
use_symbol_keys
| type | default | version |
|---|---|---|
| bool | true | 1.11.0 |
If true, the keys of the record are the symbols returned by GC.stat. If false, they are converted to strings.
The default value assumes that the events are sent to out_stdout, which serializes the symbols as strings anyway. Set this parameter to false when you send the events to another plugin, because a plugin which looks up a field by its name expects string keys. For example, the csv formatter emits empty values for fields count unless use_symbol_keys false is set.
Output Example
With the example configuration above, the events look like this:
2026-08-24 10:38:04.387923158 +0900 gc_stat: {"count":20,"time":38,"marking_time":28,"sweeping_time":9,"heap_allocated_pages":184,"heap_empty_pages":0,"heap_allocatable_slots":0,"heap_available_slots":170033,"heap_live_slots":128266,"heap_free_slots":41767,"heap_final_slots":0,"heap_marked_slots":118143,"heap_eden_pages":184,"total_allocated_pages":184,"total_freed_pages":0,"total_allocated_objects":414773,"total_freed_objects":286507,"malloc_increase_bytes":1117672,"malloc_increase_bytes_limit":16777216,"minor_gc_count":15,"major_gc_count":5,"compact_count":0,"read_barrier_faults":0,"total_moved_objects":0,"remembered_wb_unprotected_objects":0,"remembered_wb_unprotected_objects_limit":676,"old_objects":116402,"old_objects_limit":135202,"oldmalloc_increase_bytes":3586192,"oldmalloc_increase_bytes_limit":16777216}
The set of the keys depends on the Ruby version which runs Fluentd. See GC.stat for the meaning of each key.
Multi-Process Environment
If you use this plugin under the multi-process environment, each worker emits the statistics of its own process. Since the record itself has no field to tell the workers apart, embed worker_id in the tag:
<source>
@type gc_stat
tag "gc_stat.#{worker_id}"
</source>
See config article for the embedded Ruby code.