feat(injector): External resource detection for OpenTelemetry-instrumented applications

.github/workflows/ci.yaml

@@ -59,6 +59,12 @@ jobs:
         run: |
           make lint
 
+      - name: lint C sources with clang-format
+        uses: jidicula/[email protected]
+        with:
+          clang-format-version: '19'
+          check-path: 'images/instrumentation/injector/src'
+
       - name: install Helm unittest plugin
         shell: bash
         run: |

Makefile

@@ -172,6 +172,10 @@ golangci-lint: golangci-lint-install
 	@echo "-------------------------------- (linting Go code)"
 	$(GOLANGCI_LINT) run
 
+.PHONY: golang-lint-fix
+golang-lint-fix: golangci-lint-install
+	$(GOLANGCI_LINT) run --fix
+
 .PHONY: helm-chart-lint
 helm-chart-lint:
 	@echo "-------------------------------- (linting Helm charts)"
@@ -200,12 +204,32 @@ prometheus-crd-version-check:
 	@echo "-------------------------------- (verifying the Prometheus CRD version is in sync)"
 	./test-resources/bin/prometheus-crd-version-check.sh
 
+.PHONY: c-lint-installed
+c-lint-installed:
+	@set +x
+	@if ! clang-format --version > /dev/null; then \
+	echo "error: clang-format is not installed. Run 'brew install clang-format' or similar."; \
+	exit 1; \
+	fi
+
+.PHONY: c-lint
+c-lint: c-lint-installed
+ifeq ("${CI}","true")
+	@echo "CI: skip linting C source files via make lint, will run as separate Github action job step"
+else
+	@echo "-------------------------------- (linting C source files)"
+	clang-format --dry-run --Werror images/instrumentation/injector/src/*.c
+endif
+
+.PHONY: c-lint-fix
+c-lint-fix: c-lint-installed
+	clang-format -i images/instrumentation/injector/src/*.c
+
 .PHONY: lint
-lint: golangci-lint helm-chart-lint shellcheck-lint prometheus-crd-version-check
+lint: golangci-lint helm-chart-lint shellcheck-lint prometheus-crd-version-check c-lint
 
 .PHONY: lint-fix
-lint-fix: golangci-lint ## Run golangci-lint linter and perform fixes
-	$(GOLANGCI_LINT) run --fix
+lint-fix: golang-lint-fix c-lint-fix
 
 ##@ Build
 

helm-chart/dash0-operator/README.md

@@ -672,9 +672,10 @@ tracing works, intended for the technically curious reader.
 You can safely skip this section if you are not interested in the technical details.
 
 Workloads in [monitored namespaces](#enable-dash0-monitoring-for-a-namespace) are instrumented by the Dash0 operator
-to enable tracing for [supported runtimes](#supported-runtimes) out of the box.
-This allows Dash0 users to avoid the hassle of manually adding the OpenTelemetry SDK to their applications—Dash0 takes
-care of it automatically.
+to enable tracing for [supported runtimes](#supported-runtimes) out of the box, and to improve Kubernetes-related
+resource attribute auto-detection.
+This allows Dash0 users to avoid the hassle of manually adding the OpenTelemetry SDK to their applications.
+Dash0 takes care of it automatically.
 
 To achieve this, the Dash0 operator adds an
 [init container](https://kubernetes.io/docs/concepts/workloads/pods/init-containers/) with the [Dash0 instrumentation
@@ -698,6 +699,16 @@ For example, it sets (or appends to) `NODE_OPTIONS` to activate the
 [Dash0 OpenTelemetry distribution for Node.js](https://github.com/dash0hq/opentelemetry-js-distribution) to collect
 tracing data from all Node.js workloads.
 
+Additionally, the Dash0 injector automatically improves Kubernetes-related resource attributes as follows: The operator
+sets the environment variables `DASH0_NAMESPACE_NAME`, `DASH0_POD_NAME`, `DASH0_POD_UID` and `DASH0_CONTAINER_NAME` on
+workloads.
+The Dash0 injector binary picks these values up and uses them to populate the resource attributes `k8s.namespace.name`,
+`k8s.pod.name`, `k8s.pod.uid` and `k8s.container.name` via the `OTEL_RESOURCE_ATTRIBUTES` environment variable.
+If `OTEL_RESOURCE_ATTRIBUTES` is already set on the process, the key-value pairs for these attributes are appended to
+the existing value of `OTEL_RESOURCE_ATTRIBUTES`.
+If `OTEL_RESOURCE_ATTRIBUTES` was not set on the process, the Dash0 injector will add `OTEL_RESOURCE_ATTRIBUTES` as
+a new environment variable.
+
 If you are curious, the source code for the injector is open source and can be found
 [here](https://github.com/dash0hq/dash0-operator/blob/main/images/instrumentation/injector/src/dash0_injector.c).
 

images/instrumentation/injector/src/dash0_injector.c

@@ -1,112 +1,195 @@
-#include <unistd.h>
 #include <stdint.h>
+#include <unistd.h>
 
 #define ALIGN (sizeof(size_t))
 #define UCHAR_MAX 255
-#define ONES ((size_t)-1/UCHAR_MAX)
-#define HIGHS (ONES * (UCHAR_MAX/2+1))
-#define HASZERO(x) ((x)-ONES & ~(x) & HIGHS)
+#define ONES ((size_t)-1 / UCHAR_MAX)
+#define HIGHS (ONES * (UCHAR_MAX / 2 + 1))
+#define HASZERO(x) ((x) - ONES & ~(x) & HIGHS)
 
 #define NODE_OPTIONS_ENV_VAR_NAME "NODE_OPTIONS"
-#define NODE_OPTIONS_DASH0_REQUIRE "--require /__dash0__/instrumentation/node.js/node_modules/@dash0hq/opentelemetry"
+#define NODE_OPTIONS_DASH0_REQUIRE                                             \
+  "--require "                                                                 \
+  "/__dash0__/instrumentation/node.js/node_modules/@dash0hq/opentelemetry"
+#define OTEL_RESOURCE_ATTRIBUTES_ENV_VAR_NAME "OTEL_RESOURCE_ATTRIBUTES"
+#define DASH0_NAMESPACE_NAME_ENV_VAR_NAME "DASH0_NAMESPACE_NAME"
+#define DASH0_POD_UID_ENV_VAR_NAME "DASH0_POD_UID"
+#define DASH0_POD_NAME_ENV_VAR_NAME "DASH0_POD_NAME"
+#define DASH0_POD_CONTAINER_NAME_VAR_NAME "DASH0_CONTAINER_NAME"
 
 extern char **__environ;
 
-size_t __strlen(const char *s)
-{
-	const char *a = s;
-	const size_t *w;
-	for (; (uintptr_t)s % ALIGN; s++) if (!*s) return s-a;
-	for (w = (const void *)s; !HASZERO(*w); w++);
-	for (s = (const void *)w; *s; s++);
-	return s-a;
+size_t __strlen(const char *s) {
+  const char *a = s;
+  const size_t *w;
+  for (; (uintptr_t)s % ALIGN; s++)
+    if (!*s)
+      return s - a;
+  for (w = (const void *)s; !HASZERO(*w); w++)
+    ;
+  for (s = (const void *)w; *s; s++)
+    ;
+  return s - a;
 }
 
-char *__strchrnul(const char *s, int c)
-{
-	size_t *w, k;
-
-	c = (unsigned char)c;
-	if (!c) return (char *)s + __strlen(s);
-
-	for (; (uintptr_t)s % ALIGN; s++)
-		if (!*s || *(unsigned char *)s == c) return (char *)s;
-	k = ONES * c;
-	for (w = (void *)s; !HASZERO(*w) && !HASZERO(*w^k); w++);
-	for (s = (void *)w; *s && *(unsigned char *)s != c; s++);
-	return (char *)s;
+char *__strchrnul(const char *s, int c) {
+  size_t *w, k;
+
+  c = (unsigned char)c;
+  if (!c)
+    return (char *)s + __strlen(s);
+
+  for (; (uintptr_t)s % ALIGN; s++)
+    if (!*s || *(unsigned char *)s == c)
+      return (char *)s;
+  k = ONES * c;
+  for (w = (void *)s; !HASZERO(*w) && !HASZERO(*w ^ k); w++)
+    ;
+  for (s = (void *)w; *s && *(unsigned char *)s != c; s++)
+    ;
+  return (char *)s;
 }
 
-char *__strcpy(char *restrict dest, const char *restrict src)
-{
-	const unsigned char *s = src;
-	unsigned char *d = dest;
-	while ((*d++ = *s++));
-	return dest;
+char *__strcpy(char *restrict dest, const char *restrict src) {
+  const unsigned char *s = src;
+  unsigned char *d = dest;
+  while ((*d++ = *s++))
+    ;
+  return dest;
 }
 
-char *__strcat(char *restrict dest, const char *restrict src)
-{
-	__strcpy(dest + __strlen(dest), src);
-	return dest;
+char *__strcat(char *restrict dest, const char *restrict src) {
+  __strcpy(dest + __strlen(dest), src);
+  return dest;
 }
 
-int __strcmp(const char *l, const char *r)
-{
-	for (; *l==*r && *l; l++, r++);
-	return *(unsigned char *)l - *(unsigned char *)r;
+int __strcmp(const char *l, const char *r) {
+  for (; *l == *r && *l; l++, r++)
+    ;
+  return *(unsigned char *)l - *(unsigned char *)r;
 }
 
-int __strncmp(const char *_l, const char *_r, size_t n)
-{
-	const unsigned char *l=(void *)_l, *r=(void *)_r;
-	if (!n--) return 0;
-	for (; *l && *r && n && *l == *r ; l++, r++, n--);
-	return *l - *r;
+int __strncmp(const char *_l, const char *_r, size_t n) {
+  const unsigned char *l = (void *)_l, *r = (void *)_r;
+  if (!n--)
+    return 0;
+  for (; *l && *r && n && *l == *r; l++, r++, n--)
+    ;
+  return *l - *r;
 }
 
-char *__getenv(const char *name)
-{
-	size_t l = __strchrnul(name, '=') - name;
-	if (l && !name[l] && __environ)
-		for (char **e = __environ; *e; e++)
-			if (!__strncmp(name, *e, l) && l[*e] == '=')
-				return *e + l+1;
-	return 0;
+char *__getenv(const char *name) {
+  size_t l = __strchrnul(name, '=') - name;
+  if (l && !name[l] && __environ)
+    for (char **e = __environ; *e; e++)
+      if (!__strncmp(name, *e, l) && l[*e] == '=')
+        return *e + l + 1;
+  return 0;
 }
 
 /*
  * Buffers of statically-allocated memory that we can use to safely return to
  * the program manipulated values of env vars without dynamic allocations.
  */
-char val1[1012];
-char val2[1012];
-
-char *getenv(const char *name)
-{
-    char *origValue = __getenv(name);
-    int l = __strlen(name);
-
-    char *nodeOptionsVarName = NODE_OPTIONS_ENV_VAR_NAME;
-    if (__strcmp(name, nodeOptionsVarName) == 0)
-    {
-        if (__strlen(val1) == 0)
-        {
-            // Prepend our --require as the first item to the NODE_OPTIONS string.
-            char *nodeOptionsDash0Require = NODE_OPTIONS_DASH0_REQUIRE;
-            __strcat(val1, nodeOptionsDash0Require);
-
-            if (origValue != NULL)
-            {
-                // If NODE_OPTIONS were present, append the existing NODE_OPTIONS after our --require.
-                __strcat(val1, " ");
-                __strcat(val1, origValue);
-            }
+char cachedModifiedOtelResourceAttributesValue[1012];
+char cachedModifiedNodeOptionsValue[1012];
+
+char *getenv(const char *name) {
+  char *origValue = __getenv(name);
+  int l = __strlen(name);
+
+  char *otelResourceAttributesVarName = OTEL_RESOURCE_ATTRIBUTES_ENV_VAR_NAME;
+  char *nodeOptionsVarName = NODE_OPTIONS_ENV_VAR_NAME;
+  if (__strcmp(name, otelResourceAttributesVarName) == 0) {
+    if (__strlen(cachedModifiedOtelResourceAttributesValue) == 0) {
+      // This environment variable (OTEL_RESOURCE_ATTRIBUTES) has not been
+      // requested before, calculate the modified value and cache it.
+      char *namespaceName = __getenv(DASH0_NAMESPACE_NAME_ENV_VAR_NAME);
+      char *podUid = __getenv(DASH0_POD_UID_ENV_VAR_NAME);
+      char *podName = __getenv(DASH0_POD_NAME_ENV_VAR_NAME);
+      char *containerName = __getenv(DASH0_POD_CONTAINER_NAME_VAR_NAME);
+
+      int attributeCount = 0;
+
+      /*
+       * We do not perform octect escaping in the resource attributes as
+       * specified in
+       * https://opentelemetry.io/docs/specs/otel/resource/sdk/#specifying-resource-information-via-an-environment-variable
+       * because the values that are passed down to the injector comes from
+       * fields that Kubernetes already enforces to either conform to RFC 1035
+       * or RFC RFC 1123
+       * (https://kubernetes.io/docs/concepts/overview/working-with-objects/names/#dns-label-names),
+       * and in either case, none of the characters allowed require escaping
+       * based on https://www.w3.org/TR/baggage/#header-content
+       */
+
+      if (namespaceName != NULL && __strlen(namespaceName) > 0) {
+        __strcat(cachedModifiedOtelResourceAttributesValue,
+                 "k8s.namespace.name=");
+        __strcat(cachedModifiedOtelResourceAttributesValue, namespaceName);
+        attributeCount += 1;
+      }
+
+      if (podName != NULL && __strlen(podName) > 0) {
+        if (attributeCount > 0) {
+          __strcat(cachedModifiedOtelResourceAttributesValue, ",");
         }
 
-        return val1;
+        __strcat(cachedModifiedOtelResourceAttributesValue, "k8s.pod.name=");
+        __strcat(cachedModifiedOtelResourceAttributesValue, podName);
+        attributeCount += 1;
+      }
+
+      if (podUid != NULL && __strlen(podUid) > 0) {
+        if (attributeCount > 0) {
+          __strcat(cachedModifiedOtelResourceAttributesValue, ",");
+        }
+
+        __strcat(cachedModifiedOtelResourceAttributesValue, "k8s.pod.uid=");
+        __strcat(cachedModifiedOtelResourceAttributesValue, podUid);
+        attributeCount += 1;
+      }
+
+      if (containerName != NULL && __strlen(containerName) > 0) {
+        if (attributeCount > 0) {
+          __strcat(cachedModifiedOtelResourceAttributesValue, ",");
+        }
+
+        __strcat(cachedModifiedOtelResourceAttributesValue,
+                 "k8s.container.name=");
+        __strcat(cachedModifiedOtelResourceAttributesValue, containerName);
+        attributeCount += 1;
+      }
+
+      if (origValue != NULL && __strlen(origValue) > 0) {
+        if (attributeCount > 0) {
+          __strcat(cachedModifiedOtelResourceAttributesValue, ",");
+        }
+
+        __strcat(cachedModifiedOtelResourceAttributesValue, origValue);
+      }
     }
 
-    return origValue;
-}
+    return cachedModifiedOtelResourceAttributesValue;
+  } else if (__strcmp(name, nodeOptionsVarName) == 0) {
+    if (__strlen(cachedModifiedNodeOptionsValue) == 0) {
+      // This environment variable (NODE_OPTIONS) has not been requested before,
+      // calculate the modified value and cache it.
+
+      // Prepend our --require as the first item to the NODE_OPTIONS string.
+      char *nodeOptionsDash0Require = NODE_OPTIONS_DASH0_REQUIRE;
+      __strcat(cachedModifiedNodeOptionsValue, nodeOptionsDash0Require);
+
+      if (origValue != NULL && __strlen(origValue) > 0) {
+        // If NODE_OPTIONS were present, append the existing NODE_OPTIONS after
+        // our --require.
+        __strcat(cachedModifiedNodeOptionsValue, " ");
+        __strcat(cachedModifiedNodeOptionsValue, origValue);
+      }
+    }
 
+    return cachedModifiedNodeOptionsValue;
+  }
+
+  return origValue;
+}

images/instrumentation/injector/test/app/index.js

@@ -34,6 +34,9 @@ function main () {
       process.stdout.write("; ")
       echoEnvVar("NODE_OPTIONS");
       break;
+    case "otel_resource_attributes":
+      echoEnvVar("OTEL_RESOURCE_ATTRIBUTES");
+      break;
     default:
       console.error(`unknown test case: ${testCase}`);
       process.exit(1)