docs: monitoring: document undocumented metrics and fix tables#2378
docs: monitoring: document undocumented metrics and fix tables#2378eschabell wants to merge 1 commit intofluent:masterfrom
Conversation
|
Note Currently processing new changes in this PR. This may take a few minutes, please wait... ✏️ Tip: You can disable in-progress messages and the fortune message in your review settings. Tip CodeRabbit can use Trivy to scan for security misconfigurations and secrets in Infrastructure as Code files.Add a .trivyignore file to your project to customize which findings Trivy reports. 📝 WalkthroughWalkthroughReorganized and reformatted v1/v2 metric tables in Changes
Sequence Diagram(s)(Skipped — documentation-only changes.) Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
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. Comment |
There was a problem hiding this comment.
Pull request overview
This pull request updates the Fluent Bit monitoring documentation by adding four previously undocumented metrics, correcting units and grammar in existing metric descriptions, and improving table organization through alphabetical sorting and normalized formatting.
Changes:
- Added four new ring buffer and output capacity metrics to the v2 metrics table
- Corrected units for upstream connection metrics from "bytes" to "connections"
- Fixed grammar issues in multiple metric descriptions and corrected a stray markdown bold
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
|
@coderabbitai review |
✅ Actions performedReview triggered.
|
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 1 out of 1 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
eschabell
force-pushed
the
erics_monitoring_fixes
branch
from
ca93ce4 to
a3a7b41
Compare
|
@cosmo0920 and @patrick-stephens ready for review! |
eschabell
force-pushed
the
erics_monitoring_fixes
branch
from
a3a7b41 to
2faebed
Compare
|
Hi @eschabell! I noticed that this PR does not include the
|
eschabell
force-pushed
the
erics_monitoring_fixes
branch
from
2faebed to
1b12e11
Compare
|
Thanks @c-neto, added those to the PR, appreciate that so tagging you for another review! |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
🧹 Nitpick comments (1)
administration/monitoring.md (1)
210-212: Ring buffer metrics are missing an availability qualifier.Every other conditionally-emitted metric in this table (files, long_line, memrb) has a note explaining when it appears. These three ring buffer metrics have no such qualifier, leaving users guessing when to expect them.
Based on learnings, ring buffer parameters use the
thread.ring_buffer.*configuration prefix (e.g.,thread.ring_buffer.capacity), so a note along these lines would align with the pattern used elsewhere in the table:📝 Suggested addition (same pattern as surrounding rows)
-| `fluentbit_input_ring_buffer_retries_total` | name: the name or alias for the input instance | The number of ring buffer write retries. | counter | retries | -| `fluentbit_input_ring_buffer_retry_failures_total` | name: the name or alias for the input instance | The number of ring buffer write retry failures. | counter | failures | -| `fluentbit_input_ring_buffer_writes_total` | name: the name or alias for the input instance | The number of ring buffer write operations. | counter | writes | +| `fluentbit_input_ring_buffer_retries_total` | name: the name or alias for the input instance | The number of ring buffer write retries. Only available for input plugins configured with ring buffer mode (using `thread.ring_buffer.*` parameters). | counter | retries | +| `fluentbit_input_ring_buffer_retry_failures_total` | name: the name or alias for the input instance | The number of ring buffer write retry failures. Only available for input plugins configured with ring buffer mode (using `thread.ring_buffer.*` parameters). | counter | failures | +| `fluentbit_input_ring_buffer_writes_total` | name: the name or alias for the input instance | The number of ring buffer write operations. Only available for input plugins configured with ring buffer mode (using `thread.ring_buffer.*` parameters). | counter | writes |Based on learnings, the correct configuration key prefix for ring buffer parameters is
thread.(notthreaded.).🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@administration/monitoring.md` around lines 210 - 212, Add an availability qualifier for the three ring buffer metrics (`fluentbit_input_ring_buffer_retries_total`, `fluentbit_input_ring_buffer_retry_failures_total`, `fluentbit_input_ring_buffer_writes_total`) matching the pattern used for other conditional metrics (e.g., files, long_line, memrb): state that these appear only when ring buffering is enabled via the `thread.ring_buffer.*` configuration keys (use the correct `thread.` prefix, not `threaded.`), and include a short note in the table cell explaining when users should expect the metrics to be emitted.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@administration/monitoring.md`:
- Line 231: Replace the incorrect metric name
`fluentbit_input_chunks.storage_chunks` with the correct Prometheus-compliant
metric `fluentbit_input_storage_chunks` in the documentation table; ensure the
period is removed and the underscore-only convention used elsewhere is followed
so the entry matches the actual Fluent Bit metric naming (e.g., aligns with
`fluentbit_input_storage_chunks` and related rows).
- Line 193: Update the metric documentation for fluentbit_build_info: change its
description to state that it is an informational metric that always exports the
value 1 and carries build metadata via labels (e.g., hostname, version, os), and
change its unit from "seconds" to a dimensionless unit such as "1" (or
"dimensionless") so it follows Prometheus `_info` conventions; locate the entry
referencing `fluentbit_build_info` in the monitoring docs and update the
description text and the unit column accordingly.
---
Nitpick comments:
In `@administration/monitoring.md`:
- Around line 210-212: Add an availability qualifier for the three ring buffer
metrics (`fluentbit_input_ring_buffer_retries_total`,
`fluentbit_input_ring_buffer_retry_failures_total`,
`fluentbit_input_ring_buffer_writes_total`) matching the pattern used for other
conditional metrics (e.g., files, long_line, memrb): state that these appear
only when ring buffering is enabled via the `thread.ring_buffer.*` configuration
keys (use the correct `thread.` prefix, not `threaded.`), and include a short
note in the table cell explaining when users should expect the metrics to be
emitted.
1b12e11 to
48caef0
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
administration/monitoring.md (1)
210-212: Consider adding an availability note to the ring buffer metrics for consistency.
fluentbit_input_memrb_dropped_*(lines 206–207) notes "Only available for input plugins withstorage.typeset tomemrb", and the Tail-specific metrics note their plugin constraint. The ring buffer metrics are only emitted for threaded input plugins withthread.ring_buffer.*configured; adding a similar note would keep the table consistent and help users know when to expect these metrics.✏️ Suggested wording additions
-| `fluentbit_input_ring_buffer_retries_total` | name: the name or alias for the input instance | The number of ring buffer write retries. | counter | retries | -| `fluentbit_input_ring_buffer_retry_failures_total` | name: the name or alias for the input instance | The number of ring buffer write retry failures. | counter | failures | -| `fluentbit_input_ring_buffer_writes_total` | name: the name or alias for the input instance | The number of ring buffer write operations. | counter | writes | +| `fluentbit_input_ring_buffer_retries_total` | name: the name or alias for the input instance | The number of ring buffer write retries. Only available for input plugins with the ring buffer enabled (`thread.ring_buffer.*` configuration). | counter | retries | +| `fluentbit_input_ring_buffer_retry_failures_total` | name: the name or alias for the input instance | The number of ring buffer write retry failures. Only available for input plugins with the ring buffer enabled (`thread.ring_buffer.*` configuration). | counter | failures | +| `fluentbit_input_ring_buffer_writes_total` | name: the name or alias for the input instance | The number of ring buffer write operations. Only available for input plugins with the ring buffer enabled (`thread.ring_buffer.*` configuration). | counter | writes |Based on learnings: the correct configuration key prefix for ring buffer parameters in Fluent Bit input plugins is
thread.(e.g.,thread.ring_buffer.capacity,thread.ring_buffer.window,thread.ring_buffer.retry_limit).🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@administration/monitoring.md` around lines 210 - 212, Add an availability note to the ring buffer metric rows (`fluentbit_input_ring_buffer_retries_total`, `fluentbit_input_ring_buffer_retry_failures_total`, `fluentbit_input_ring_buffer_writes_total`) stating these metrics are only emitted for threaded input plugins when the thread.ring_buffer.* parameters are configured (e.g., thread.ring_buffer.capacity, thread.ring_buffer.window, thread.ring_buffer.retry_limit), matching the style of the existing `fluentbit_input_memrb_dropped_*` and Tail-specific notes for consistency.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@administration/monitoring.md`:
- Line 245: Update the metric name documented in the monitoring table from
fluentbit_storage_mem_chunk to the correct plural fluentbit_storage_mem_chunks
to match Fluent Bit's actual metric (registered in flb_storage.c as mem_chunks
and equivalent to chunks.mem_chunks); change the table entry label and any
references to the singular form so it aligns with adjacent storage metrics like
fs_chunks and fs_chunks_busy.
---
Duplicate comments:
In `@administration/monitoring.md`:
- Line 193: The table row for fluentbit_build_info is incorrect: update the
description to state that fluentbit_build_info is a Prometheus _info metric that
always exports the constant value 1 and carries metadata (hostname, version, os)
as labels, and change the unit from "seconds" to "unitless" (or "1"); modify the
row for the metric named fluentbit_build_info in the monitoring document to
reflect this corrected description and unit.
---
Nitpick comments:
In `@administration/monitoring.md`:
- Around line 210-212: Add an availability note to the ring buffer metric rows
(`fluentbit_input_ring_buffer_retries_total`,
`fluentbit_input_ring_buffer_retry_failures_total`,
`fluentbit_input_ring_buffer_writes_total`) stating these metrics are only
emitted for threaded input plugins when the thread.ring_buffer.* parameters are
configured (e.g., thread.ring_buffer.capacity, thread.ring_buffer.window,
thread.ring_buffer.retry_limit), matching the style of the existing
`fluentbit_input_memrb_dropped_*` and Tail-specific notes for consistency.
- add fluentbit_input_ring_buffer_writes_total metric
- add fluentbit_input_ring_buffer_retries_total metric
- add fluentbit_input_ring_buffer_retry_failures_total metric
- add fluentbit_output_chunk_available_capacity_percent metric
- sort v2 metrics table alphabetically
- sort v2 storage layer table alphabetically
- fix unit for upstream connection metrics from bytes to connections
- fix grammar in storage_chunks_busy description
- fix grammar in fs_chunks_busy and fs_chunks_busy_bytes descriptions
- fix grammar in upstream connection metric descriptions
- fix stray markdown bold on Complete coverage bullet point
- remove trailing period from Mem_Buf_Limit in storage_overlimit
description
- normalize table column formatting
- fix unit for fluentbit_hot_reloaded_times metric
- fix unit for fluentbit_input_ring_buffer_retry_failures_total metric
- add 5 Tail-specific metrics to the v2 metrics table:
- fluentbit_input_files_closed_total
- fluentbit_input_files_opened_total
- fluentbit_input_files_rotated_total
- fluentbit_input_long_line_truncated_total
- fluentbit_input_multiline_truncated_total
- add missing fluentbit_input_long_line_skipped_total metric
(added in fluent-bit#11457 for tracking skipped long lines
when skip_long_lines is enabled)
- fix fluentbit_input_long_line_truncated_total description to
reference truncate_long_lines instead of skip_long_lines
- clean up fluentbit_build_info description to accurately reflect
the v2 cmetrics behavior (value is the init_time epoch timestamp)
- remove invalid fluentbit_input_chunks.storage_chunks row
- fix fluentbit_storage_mem_chunks metric name (add missing s)
Fixes fluent#2377 and fixes fluent#2379 (thanks to Carlos Neto <carlos.neto.dev@gmail.com>)
Signed-off-by: Eric D. Schabell <eric@schabell.org>
eschabell
force-pushed
the
erics_monitoring_fixes
branch
from
48caef0 to
28c4e8f
Compare
2 participants
Fixes #2377
Summary by CodeRabbit