You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@camel.apache.org by fm...@apache.org on 2026/09/07 13:27:57 UTC
(camel-spring-boot) branch CAMEL-24498-restore-observability-bind-default created (now 41d713c3003)
This is an automated email from the ASF dual-hosted git repository.
Croway pushed a change to branch CAMEL-24498-restore-observability-bind-default
in repository https://gitbox.apache.org/repos/asf/camel-spring-boot.git
at 41d713c3003 CAMEL-24498: restore observability services bind default
This branch includes the following new commits:
new 41d713c3003 CAMEL-24498: restore observability services bind default
The 1 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
(camel-spring-boot) 01/01: CAMEL-24498: restore observability services bind default
Posted by fm...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
Croway pushed a commit to branch CAMEL-24498-restore-observability-bind-default
in repository https://gitbox.apache.org/repos/asf/camel-spring-boot.git
commit 41d713c3003d7e9475847198bd9dc406fcb678da
Author: Croway <fe...@gmail.com>
AuthorDate: Mon Sep 7 15:27:42 2026 +0200
CAMEL-24498: restore observability services bind default
---
.../src/main/doc/usage.adoc | 29 ++++------------------
...ervabilityServicesEnvironmentPostProcessor.java | 9 +++----
...bilityServicesEnvironmentPostProcessorTest.java | 6 ++---
.../pages/starters/observability-services.adoc | 29 ++++------------------
4 files changed, 15 insertions(+), 58 deletions(-)
diff --git a/components-starter/camel-observability-services-starter/src/main/doc/usage.adoc b/components-starter/camel-observability-services-starter/src/main/doc/usage.adoc
index bba4aec5c42..2715bc7e9e6 100644
--- a/components-starter/camel-observability-services-starter/src/main/doc/usage.adoc
+++ b/components-starter/camel-observability-services-starter/src/main/doc/usage.adoc
@@ -1,13 +1,12 @@
-Add the starter to the classpath and the endpoints below are available on a separate management listener,
-bound to loopback on port `9876`:
+Add the starter to the classpath and the endpoints below are available on a separate management listener on port `9876`:
[width="100%",cols="3,5",options="header"]
|===
| Endpoint | Purpose
-| `http://127.0.0.1:9876/observe/health` | Aggregate health
-| `http://127.0.0.1:9876/observe/health/live` | Kubernetes liveness probe
-| `http://127.0.0.1:9876/observe/health/ready` | Kubernetes readiness probe
-| `http://127.0.0.1:9876/observe/metrics` | Prometheus scrape endpoint
+| `/observe/health` | Aggregate health
+| `/observe/health/live` | Kubernetes liveness probe
+| `/observe/health/ready` | Kubernetes readiness probe
+| `/observe/metrics` | Prometheus scrape endpoint
|===
=== Injected defaults
@@ -19,7 +18,6 @@ configuration:
|===
| Property | Value | Notes
| `management.server.port` | `9876` | Management endpoints run on their own listener, separate from the application port.
-| `management.server.address` | `127.0.0.1` | The listener binds to loopback. See <<Exposing the management listener>>.
| `management.endpoints.web.exposure.include` | `health,prometheus` | Only these two endpoints are exposed.
| `management.endpoints.web.base-path` | `/observe` | Replaces the `/actuator` base path.
| `management.endpoints.web.path-mapping.prometheus` | `metrics` | Serves the Prometheus endpoint at `/observe/metrics`.
@@ -37,23 +35,6 @@ configuration:
| `management.endpoint.health.group.ready.show-details` | `always` | See <<Health detail exposure>>.
|===
-=== Exposing the management listener
-
-The management listener binds to `127.0.0.1`, so out of the box the health and metrics endpoints are reachable
-from the same host only. This matches the Spring Boot baseline, which ships no separate management listener at
-all: opening one on every interface should be a decision you make, not one the starter makes for you.
-
-A Kubernetes deployment whose kubelet probes or Prometheus scrapers reach the pod over the network has to widen
-the bind address:
-
-[source,properties]
-----
-management.server.address = 0.0.0.0
-----
-
-Pair that with a `NetworkPolicy`, or with authentication in front of the management port, so the endpoints are
-only reachable from your probes and your scrapers.
-
=== Health detail exposure
`/observe/health` uses `when-authorized`, the setting that shows the individual health indicators to an
diff --git a/components-starter/camel-observability-services-starter/src/main/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessor.java b/components-starter/camel-observability-services-starter/src/main/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessor.java
index ff0d1cfa9c3..81bd7f837fd 100644
--- a/components-starter/camel-observability-services-starter/src/main/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessor.java
+++ b/components-starter/camel-observability-services-starter/src/main/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessor.java
@@ -30,10 +30,9 @@ import org.springframework.core.env.MapPropertySource;
* user-provided configuration (application.properties, environment variables, system properties, ...) overrides them
* following the standard Spring Boot precedence rules.
* <p>
- * The defaults stay within the Spring Boot baseline: the management listener binds to loopback, and the aggregate
- * health endpoint only shows its details to an authorized caller. The {@code live} and {@code ready} probe groups
- * keep {@code show-details=always} because they are consumed unauthenticated by the kubelet, and the indicators
- * they contain report an availability state and nothing else.
+ * The aggregate health endpoint only shows its details to an authorized caller. The {@code live} and {@code ready}
+ * probe groups keep {@code show-details=always} because they are consumed unauthenticated by the kubelet, and the
+ * indicators they contain report an availability state and nothing else.
*/
public class ObservabilityServicesEnvironmentPostProcessor implements EnvironmentPostProcessor, Ordered {
@@ -43,8 +42,6 @@ public class ObservabilityServicesEnvironmentPostProcessor implements Environmen
public void postProcessEnvironment(ConfigurableEnvironment environment, SpringApplication application) {
Map<String, Object> defaults = new LinkedHashMap<>();
defaults.put("management.server.port", "9876");
- // bind the management listener to loopback; exposing it beyond the host is a conscious step
- defaults.put("management.server.address", "127.0.0.1");
defaults.put("management.endpoints.web.exposure.include", "health,prometheus");
defaults.put("management.endpoints.web.base-path", "/observe");
defaults.put("management.endpoints.web.path-mapping.prometheus", "metrics");
diff --git a/components-starter/camel-observability-services-starter/src/test/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessorTest.java b/components-starter/camel-observability-services-starter/src/test/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessorTest.java
index d27752f7736..5d6e150d82c 100644
--- a/components-starter/camel-observability-services-starter/src/test/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessorTest.java
+++ b/components-starter/camel-observability-services-starter/src/test/java/org/apache/camel/observability/services/springboot/ObservabilityServicesEnvironmentPostProcessorTest.java
@@ -58,9 +58,8 @@ public class ObservabilityServicesEnvironmentPostProcessorTest {
}
@Test
- public void managementListenerBindsToLoopback() {
- // the port alone would put a second listener on every interface
- assertEquals("127.0.0.1", environment.getProperty("management.server.address"));
+ public void managementListenerUsesSpringBootDefaultBindAddress() {
+ assertNull(environment.getProperty("management.server.address"));
}
@Test
@@ -89,7 +88,6 @@ public class ObservabilityServicesEnvironmentPostProcessorTest {
Set<String> expected = new LinkedHashSet<>(Arrays.asList(
"management.server.port",
- "management.server.address",
"management.endpoints.web.exposure.include",
"management.endpoints.web.base-path",
"management.endpoints.web.path-mapping.prometheus",
diff --git a/docs/spring-boot/modules/ROOT/pages/starters/observability-services.adoc b/docs/spring-boot/modules/ROOT/pages/starters/observability-services.adoc
index 93a054b48c2..4f2db000045 100644
--- a/docs/spring-boot/modules/ROOT/pages/starters/observability-services.adoc
+++ b/docs/spring-boot/modules/ROOT/pages/starters/observability-services.adoc
@@ -24,16 +24,15 @@ The defaults are registered as the lowest precedence property source, so anythin
== Usage
-Add the starter to the classpath and the endpoints below are available on a separate management listener,
-bound to loopback on port `9876`:
+Add the starter to the classpath and the endpoints below are available on a separate management listener on port `9876`:
[width="100%",cols="3,5",options="header"]
|===
| Endpoint | Purpose
-| `http://127.0.0.1:9876/observe/health` | Aggregate health
-| `http://127.0.0.1:9876/observe/health/live` | Kubernetes liveness probe
-| `http://127.0.0.1:9876/observe/health/ready` | Kubernetes readiness probe
-| `http://127.0.0.1:9876/observe/metrics` | Prometheus scrape endpoint
+| `/observe/health` | Aggregate health
+| `/observe/health/live` | Kubernetes liveness probe
+| `/observe/health/ready` | Kubernetes readiness probe
+| `/observe/metrics` | Prometheus scrape endpoint
|===
=== Injected defaults
@@ -45,7 +44,6 @@ configuration:
|===
| Property | Value | Notes
| `management.server.port` | `9876` | Management endpoints run on their own listener, separate from the application port.
-| `management.server.address` | `127.0.0.1` | The listener binds to loopback. See <<Exposing the management listener>>.
| `management.endpoints.web.exposure.include` | `health,prometheus` | Only these two endpoints are exposed.
| `management.endpoints.web.base-path` | `/observe` | Replaces the `/actuator` base path.
| `management.endpoints.web.path-mapping.prometheus` | `metrics` | Serves the Prometheus endpoint at `/observe/metrics`.
@@ -63,23 +61,6 @@ configuration:
| `management.endpoint.health.group.ready.show-details` | `always` | See <<Health detail exposure>>.
|===
-=== Exposing the management listener
-
-The management listener binds to `127.0.0.1`, so out of the box the health and metrics endpoints are reachable
-from the same host only. This matches the Spring Boot baseline, which ships no separate management listener at
-all: opening one on every interface should be a decision you make, not one the starter makes for you.
-
-A Kubernetes deployment whose kubelet probes or Prometheus scrapers reach the pod over the network has to widen
-the bind address:
-
-[source,properties]
-----
-management.server.address = 0.0.0.0
-----
-
-Pair that with a `NetworkPolicy`, or with authentication in front of the management port, so the endpoints are
-only reachable from your probes and your scrapers.
-
=== Health detail exposure
`/observe/health` uses `when-authorized`, the setting that shows the individual health indicators to an