input_metrics: discard cmetrics with no series #10700
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
WalkthroughThe changes introduce stricter handling of empty metrics payloads throughout the metrics processing and output pipeline. They add a utility to check for empty metrics, update macro definitions to use C boolean types, and ensure that output plugins skip flushes when metrics data is empty, thus preventing attempts to send empty payloads. Changes
Sequence Diagram(s)sequenceDiagram
participant Input as Input Plugin
participant Processor as Metrics Processor
participant Output as Output Plugin
Input->>Processor: Provide metrics context
Processor->>Processor: Process metrics (may filter out all data)
Processor->>Output: Pass processed metrics context
Output->>Output: Check if metrics context is empty (flb_metrics_is_empty)
alt Metrics context is empty
Output-->>Output: Log debug, skip flush, do not send payload
else Metrics context is not empty
Output->>Output: Serialize and send payload
end
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~18 minutes Assessment against linked issues
Assessment against linked issues: Out-of-scope changesNo out-of-scope changes were found. Poem
Note 🔌 MCP (Model Context Protocol) integration is now available in Early Access!Pro users can now connect to remote MCP servers under the Integrations page to get reviews and chat conversations that understand additional development context. ✨ Finishing Touches
🧪 Generate unit tests
🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
CodeRabbit Configuration File (
|
Signed-off-by: Eduardo Silva <[email protected]>
Signed-off-by: Eduardo Silva <[email protected]>
When ingesting a CMetrics context into the pipeline, it may contain metadata but no actual data points (e.g., counters, gauges, histograms). Such metrics should be dropped and not ingested. Signed-off-by: Eduardo Silva <[email protected]>
When a metrics context is about to be flushed, it may pass through one or more processors before reaching the output plugin. Processing can result in a valid metrics context with no time series data, which is useless for output plugins. This change detects that case, skips the flush, and returns FLB_OK to the engine. Signed-off-by: Eduardo Silva <[email protected]>
edsiper
requested review from
leonardo-albertovich,
fujimotos and
koleini
as code owners
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (4)
include/fluent-bit/flb_metrics.h (1)
95-99: Missing<stdbool.h>include for public header
flb_metrics_is_empty()returnsbool, but this header itself doesn’t pull in<stdbool.h>.
It currently works only because downstream headers happen to includeflb_macros.h, which now adds it.
Adding the include here removes the hidden dependency chain.#include <cmetrics/cmt_filter.h> +#include <stdbool.h>src/flb_input_metric.c (1)
72-79: Tiny optimisation – avoid double checkYou’ve already selected
encode_context;encode_context == NULLcannot be true because both branches set it to eitherout_contextorcmt.
The explicit NULL test can be dropped leaving only the emptiness check and saving one conditional branch.Not critical – feel free to ignore.
include/fluent-bit/flb_output.h (2)
673-679: Edge-case skip is good – but keep symmetry with logsNice addition. Consider also checking
total_events == 0for metrics chunks (mirroring the logs check) – some future encoder may setsize==0but non-zero events or vice-versa.
736-740: Unused variable in non-metrics builds
encode_contextis declared at function scope but is only referenced inside the METRICS branch.
WithFLB_HAVE_METRICSdisabled some compilers will warn.Either wrap the declaration in the same
#ifdefor mark itFLB_UNUSED.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (5)
include/fluent-bit/flb_macros.h(1 hunks)include/fluent-bit/flb_metrics.h(2 hunks)include/fluent-bit/flb_output.h(9 hunks)src/flb_input_metric.c(3 hunks)src/flb_metrics.c(1 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (28)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit (Arm64), amd64_arm64, -DCMAKE_SYSTEM_NAME=Windows -DCMA...
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit, x64, x64-windows-static, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 32bit, x86, x86-windows-static, 3.31.6)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_COVERAGE=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, clang, clang++)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-24.04, clang-14)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, gcc, g++)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-centos-7
- GitHub Check: PR - fuzzing test
🔇 Additional comments (2)
include/fluent-bit/flb_macros.h (1)
23-27: Boolean macro switch may leak into pre-C99 translation unitsReplacing the integer literals with
true/falseis cleaner, but it silently promotesFLB_TRUE/FLB_FALSEto the_Booltype.
Any compilation unit that (a) still builds in C90 mode or (b) relies on var-arg promotion differences might now get a surprise warning/error.No change requested, just verify all modules are being compiled with
-std=c99(or later) and check for new warnings in CI.include/fluent-bit/flb_output.h (1)
898-905:flb_event_chunk_create()with NULL data may assertWhen
serialization_buffer_offset == 0we freep_bufthen create a chunk withdata=NULLandsize=0.
If the helper insists on a non-NULL buffer even for zero length this will abort at runtime.Please verify the helper tolerates
(data==NULL && size==0)across all builds, or allocate a 1-byte dummy buffer instead.
Comment on lines
+380
to
+387
| bool flb_metrics_is_empty(struct cmt *cmt) | ||
| { | ||
| return cfl_list_is_empty(&cmt->counters) && | ||
| cfl_list_is_empty(&cmt->gauges) && | ||
| cfl_list_is_empty(&cmt->histograms) && | ||
| cfl_list_is_empty(&cmt->summaries) && | ||
| cfl_list_is_empty(&cmt->untypeds); | ||
| } |
There was a problem hiding this comment.
Guard against NULL input in flb_metrics_is_empty()
If a caller ever passes a NULL struct cmt *, the function will seg-fault.
A defensive early-return costs nothing and prevents a crash path that’s very hard to diagnose at runtime.
bool flb_metrics_is_empty(struct cmt *cmt)
{
- return cfl_list_is_empty(&cmt->counters) &&
+ if (cmt == NULL) {
+ return true;
+ }
+
+ return cfl_list_is_empty(&cmt->counters) &&
...
}📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
Suggested change
| bool flb_metrics_is_empty(struct cmt *cmt) | |
| { | |
| return cfl_list_is_empty(&cmt->counters) && | |
| cfl_list_is_empty(&cmt->gauges) && | |
| cfl_list_is_empty(&cmt->histograms) && | |
| cfl_list_is_empty(&cmt->summaries) && | |
| cfl_list_is_empty(&cmt->untypeds); | |
| } | |
| bool flb_metrics_is_empty(struct cmt *cmt) | |
| { | |
| if (cmt == NULL) { | |
| return true; | |
| } | |
| return cfl_list_is_empty(&cmt->counters) && | |
| cfl_list_is_empty(&cmt->gauges) && | |
| cfl_list_is_empty(&cmt->histograms) && | |
| cfl_list_is_empty(&cmt->summaries) && | |
| cfl_list_is_empty(&cmt->untypeds); | |
| } |
🤖 Prompt for AI Agents
In src/flb_metrics.c around lines 380 to 387, the function flb_metrics_is_empty
does not check if the input pointer cmt is NULL, which can cause a segmentation
fault. Add a defensive check at the start of the function to return true (or an
appropriate value) immediately if cmt is NULL to prevent crashes.
There was a problem hiding this comment.
use FLB_TRUE instead true
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Check the current definition of FLB_TRUE and FLB_FALSE
cat include/fluent-bit/flb_macros.h | grep -A2 -B2 "FLB_TRUE\|FLB_FALSE"
Length of output: 173
🏁 Script executed:
#!/bin/bash
# Check how other functions handle NULL checks with boolean returns
rg -A3 "if \(.*== NULL\)" src/flb_metrics.c | head -20
Length of output: 56
🏁 Script executed:
#!/bin/bash
# Look for other bool-returning functions that use FLB_TRUE
rg "bool.*\(" include/fluent-bit/ -A5 | grep -B3 "FLB_TRUE\|return true"
Length of output: 74
@edsiper, you're right! For consistency with the rest of the codebase, here's the updated suggestion using FLB_TRUE:
bool flb_metrics_is_empty(struct cmt *cmt)
{
+ if (cmt == NULL) {
+ return FLB_TRUE;
+ }
+
return cfl_list_is_empty(&cmt->counters) &&
cfl_list_is_empty(&cmt->gauges) &&
cfl_list_is_empty(&cmt->histograms) &&
cfl_list_is_empty(&cmt->summaries) &&
cfl_list_is_empty(&cmt->untypeds);
}This maintains consistency with the project's macro usage pattern, even though FLB_TRUE is now defined as the boolean literal true.
🧬 Code Graph Analysis Results
Relevant snippets from src/flb_metrics.c (lines 22-35)
static int id_exists(int id, struct flb_metrics *metrics)
{
struct mk_list *head;
struct flb_metric *metric;
mk_list_foreach(head, &metrics->list) {
metric = mk_list_entry(head, struct flb_metric, _head);
if (metric->id == id) {
return FLB_TRUE;
}
}
return FLB_FALSE;
}- This function checks if a metric ID already exists in the metrics list.
- It returns
FLB_TRUEif the ID exists, otherwiseFLB_FALSE. - The user comment about using
FLB_TRUEinstead oftrueis directly relevant here since this function returnsFLB_TRUEandFLB_FALSEmacros instead of standard Ctrue/false.
Relevant snippets from src/flb_metrics.c (lines 37-51)
static int id_get(struct flb_metrics *metrics)
{
int id;
int ret = FLB_FALSE;
/* Try to use 'count' as an id */
id = metrics->count;
while ((ret = id_exists(id, metrics)) == FLB_TRUE) {
id++;
}
return id;
}- This function generates a new unique metric ID.
- It uses
FLB_TRUEandFLB_FALSEmacros for boolean logic. - This is another place where the user comment about using
FLB_TRUEinstead oftrueis relevant.
Relevant snippet from include/fluent-bit/flb_mem.h (lines 84-96)
void *flb_calloc(size_t n, const size_t size) {
if (size == 0) {
return NULL;
}
#ifdef FLB_HAVE_TESTS_OSSFUZZ
// Add chance of failure. Used by fuzzing to test error-handling code.
if (flb_fuzz_get_probability(1)) {
return NULL;
}
#endif
return calloc(n, size);
}- This snippet shows the
flb_callocfunction used inflb_metrics_createfor memory allocation. - It is relevant for understanding memory allocation and error handling in the metrics code.
Relevant snippet from src/flb_sds.c (lines 389-399)
void flb_sds_destroy(flb_sds_t s)
{
struct flb_sds *head;
if (!s) {
return;
}
head = FLB_SDS_HEADER(s);
flb_free(head);
}- This snippet shows how
flb_sds_destroyfrees string data structures used for metric titles. - Relevant for understanding resource cleanup in
flb_metrics_destroy.
Relevant snippet from src/flb_sds.c (lines 58-76)
flb_sds_t flb_sds_create_len(const char *str, int len)
{
flb_sds_t s;
struct flb_sds *head;
s = sds_alloc(len);
if (!s) {
return NULL;
}
if (str) {
memcpy(s, str, len);
s[len] = '\0';
head = FLB_SDS_HEADER(s);
head->len = len;
}
return s;
}- This snippet shows how metric titles are created with a length limit.
- Relevant for understanding how titles are handled in
flb_metrics_addandflb_metrics_title.
Relevant snippet from lib/cmetrics/src/cmt_counter.c (lines 26-81)
struct cmt_counter *cmt_counter_create(struct cmt *cmt,
char *ns, char *subsystem,
char *name, char *help,
int label_count, char **label_keys)
{
int ret;
struct cmt_counter *counter;
if (!ns) {
cmt_log_error(cmt, "null ns not allowed");
return NULL;
}
if (!subsystem) {
cmt_log_error(cmt, "null subsystem not allowed");
return NULL;
}
if (!name || strlen(name) == 0) {
cmt_log_error(cmt, "undefined name");
return NULL;
}
if (!help || strlen(help) == 0) {
cmt_log_error(cmt, "undefined help");
return NULL;
}
counter = calloc(1, sizeof(struct cmt_counter));
if (!counter) {
cmt_errno();
return NULL;
}
cfl_list_add(&counter->_head, &cmt->counters);
ret = cmt_opts_init(&counter->opts, ns, subsystem, name, help);
if (ret == -1) {
cmt_log_error(cmt, "unable to initialize options for counter");
cmt_counter_destroy(counter);
return NULL;
}
/* Create the map */
counter->map = cmt_map_create(CMT_COUNTER, &counter->opts, label_count, label_keys,
(void *) counter);
if (!counter->map) {
cmt_log_error(cmt, "unable to allocate map for counter");
cmt_counter_destroy(counter);
return NULL;
}
/* set default counter aggregation type to cumulative */
counter->aggregation_type = CMT_AGGREGATION_TYPE_CUMULATIVE;
counter->cmt = cmt;
return counter;
}- This snippet shows how counters are created in the cmetrics library, which is used in
flb_metrics.cfor attaching metrics. - Relevant for understanding the creation of counters in functions like
attach_uptime.
Relevant snippet from lib/cmetrics/src/cmt_gauge.c (lines 27-81)
struct cmt_gauge *cmt_gauge_create(struct cmt *cmt,
char *ns, char *subsystem, char *name,
char *help, int label_count, char **label_keys)
{
int ret;
struct cmt_gauge *gauge;
if (!ns) {
cmt_log_error(cmt, "null ns not allowed");
return NULL;
}
if (!subsystem) {
cmt_log_error(cmt, "null subsystem not allowed");
return NULL;
}
if (!name || strlen(name) == 0) {
cmt_log_error(cmt, "undefined name");
return NULL;
}
if (!help || strlen(help) == 0) {
cmt_log_error(cmt, "undefined help");
return NULL;
}
gauge = calloc(1, sizeof(struct cmt_gauge));
if (!gauge) {
cmt_errno();
return NULL;
}
cfl_list_add(&gauge->_head, &cmt->gauges);
/* Initialize options */
ret = cmt_opts_init(&gauge->opts, ns, subsystem, name, help);
if (ret == -1) {
cmt_log_error(cmt, "unable to initialize options for gauge");
cmt_gauge_destroy(gauge);
return NULL;
}
/* Create the map */
gauge->map = cmt_map_create(CMT_GAUGE, &gauge->opts, label_count, label_keys,
(void *) gauge);
if (!gauge->map) {
cmt_log_error(cmt, "unable to allocate map for gauge");
cmt_gauge_destroy(gauge);
return NULL;
}
gauge->cmt = cmt;
return gauge;
}- This snippet shows how gauges are created in the cmetrics library, used in
flb_metrics.cfor attaching metrics like process start time and build info. - Relevant for understanding the metric attachment functions.
Relevant snippet from lib/cmetrics/src/cmt_gauge.c (lines 94-109)
int cmt_gauge_set(struct cmt_gauge *gauge, uint64_t timestamp, double val,
int labels_count, char **label_vals)
{
struct cmt_metric *metric;
metric = cmt_map_metric_get(&gauge->opts, gauge->map, labels_count, label_vals,
CMT_TRUE);
if (!metric) {
cmt_log_error(gauge->cmt, "unable to retrieve metric for gauge %s_%s_%s",
gauge->opts.ns, gauge->opts.subsystem,
gauge->opts.name);
return -1;
}
cmt_metric_set(metric, timestamp, val);
return 0;
}- This snippet shows how gauge values are set, using
CMT_TRUEmacro for boolean. - Relevant for understanding the boolean macro usage in metric setting functions.
[End of relevant snippets.]
Signed-off-by: Eduardo Silva <[email protected]>
There was a problem hiding this comment.
Actionable comments posted: 0
🔭 Outside diff range comments (1)
include/fluent-bit/flb_output.h (1)
49-58: Missing declaration forflb_metrics_is_empty
flb_metrics_is_empty()is used later in this header, butfluent-bit/flb_metrics.his not included.
Relying on transitive includes risks an implicit-declaration warning or build break if the order changes.+#include <fluent-bit/flb_metrics.h> /* for flb_metrics_is_empty */
🧹 Nitpick comments (1)
include/fluent-bit/flb_output.h (1)
667-679: Early-return guard looks good but consider symmetry with logsThe new guard skips a flush when the metrics event-chunk has
size == 0, mirroring the logs-path that checkstotal_events == 0.
Two suggestions:
- Use the same helper you added (
flb_metrics_is_empty) for clarity and to future-proof against a non-zero buffer that still encodes zero series.- Document in the comment why
sizeis the chosen criterion so future readers don’t wonder abouttotal_events.Otherwise the logic correctly avoids the unnecessary network call.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
include/fluent-bit/flb_output.h(9 hunks)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (28)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 32bit, x86, x86-windows-static, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit, x64, x64-windows-static, 3.31.6)
- GitHub Check: pr-windows-build / call-build-windows-package (Windows 64bit (Arm64), amd64_arm64, -DCMAKE_SYSTEM_NAME=Windows -DCMA...
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SIMD=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_THREAD=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SANITIZE_MEMORY=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_COVERAGE=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_UNDEFINED=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_SMALL=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=Off, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, clang, clang++)
- GitHub Check: run-ubuntu-unit-tests (-DSANITIZE_ADDRESS=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, gcc, g++)
- GitHub Check: run-ubuntu-unit-tests (-DFLB_JEMALLOC=On, 3.31.6, clang, clang++)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, gcc, g++, ubuntu-22.04, clang-12)
- GitHub Check: pr-compile-system-libs (-DFLB_PREFER_SYSTEM_LIBS=On, 3.31.6, clang, clang++, ubuntu-24.04, clang-14)
- GitHub Check: pr-compile-centos-7
- GitHub Check: PR - fuzzing test
🔇 Additional comments (1)
include/fluent-bit/flb_output.h (1)
903-916: Zero-metrics path may rely onflb_event_chunk_create()accepting NULL data
flb_event_chunk_create()is invoked withdata = NULLandsize = 0.
If that helper does not explicitly allow aNULLpayload the call will seg-fault, and CI won’t catch it until a metrics filter removes every series.Please confirm the helper’s contract or add a defensive fallback, e.g.
void *safe_buf = size ? buf : ""; flb_event_chunk_create(..., safe_buf, size);
|
thanks @edsiper, this solves the issue. I can see this message logged and no attempts to deliver empty chunks. [2025/08/12 15:17:13.233752000] [debug] [output] skipping flush for metrics event chunk with zero metrics after processing.It also deletes the unwanted chunks |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #10313
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
New Features
Bug Fixes
Documentation