Previously, when the mappings endpoint was not available, the beans
that provide mapping descriptions were still created. This resulted
in unnecessary CPU and memory usage collecting and storing
information that would never by used.
This commit updates the auto-configuration for the mappings endpoint
so that all the beans that it creates are conditional on the endpoint
being available, rather than only the endpoint bean itself.
Closes gh-23977
This commit updates DataSourceUnwrapper to take a separate interface
type argument if the target datasource has to be unwrapped, given that
the target type is usually not an interface.
Closes gh-24697
Previously, it was possible for Spring Integration, including its
built-in Micrometer support, to be auto-configured before the
Micrometer auto-configuration had defined the MeterRegistry bean. This
resulted in missing Spring Integration metrics.
Spring Integration is unusual in having its own built-in Micrometer
support that it configures itself. Rather than providing
auto-configuration for Integration's Micrometer support (Which isn't
needed), this commit introduces some auto-configuration that just
affects the ordering of the auto-configuration classes. This ordering
ensures that the MeterRegistry bean has been defined by Spring
Integration is auto-configured. This ensures that the MeterRegistry
bean is known to the BeanFactory when Spring Integration goes looking
for it.
See gh-24095
When running on Java 11 (where `@PostConstruct` is no longer part of
the JRE) and without a dependency on jakarta-annotation-api,
`@PostContruct` annotions are silently dropped. This leads to obscure and
hard-to-track down changes in the behaviour of our auto-configuration
as the `@PostConstruct`-annotated methods are not invoked.
To allow users to run on Java 11 without having jakarta-annotation-api
on the classpath, this commit removes use of `@PostConstruct` from main
code. A Checkstyle rule has also been added to prevent its usage in
main code from being reintroduced.
Closes gh-23723
Previously, the test would make an HTTP request and, as soon as the
response was received, it would check the presence and value of the
http.server.requests meter. This create a race condition between the
meter being registered once the response had been flushed and the
meter's presence being checked. If the check won the race, the test
would fail.
This commit updates the test to wait for up to 5 seconds for the
meter to be present and have a count of 1, matching the single request
that has been made.
Fixes gh-23919
Add a dedicate condition annotation to detect when Spring Security is
available but has not been configured by the user. The new annotation
helps simplify quite a few of our auto-configuration classes.
See gh-23421
Replace `WebSecurityConfigurer` and `WebSecurityConfigurerAdapter`
configurations with `WebSecurityCustomizer` or `SecurityFilterChain`
beans.
Closes gh-23421
Previously, the test would make an HTTP request and, as soon as the
response was received, it would check the presence and value of the
http.server.requests meter. This create a race condition between the
meter being registered once the response had been flushed and the
meter's presence being checked. If the check won the race, the test
would fail.
This commit updates the test to wait for up to 5 seconds for the
meter to be present and have a count of 1, matching the single request
that has been made.
Fixes gh-23863
Previously, the base path of a servlet-based management server could be
configured using management.server.servlet.context-path but there was no
equivalent property for WebFlux.
This commit introduces a new property, management.server.base-path,
that can be used with both servlet and reactive management servers. The
existing servlet-specific property has been deprecated in favour of the
new general property. When using the servlet stack, if both the general
property and the servlet-specific property are set, the new general
property takes precedence. When using the reactive stack, only the new
general property is considered.
Closes gh-22906
This commit makes sure to defer registration of hibernate statistics
outside of the singleton lock as it can lead to deadlocks when the
EntityManagerFactory is initialized in deferred mode.
Closes gh-23740
This commit adds support for Redis cache metrics. Users can opt-in for
statistics using the "spring.cache.redis.enable-statistics" property.
Closes gh-22701
Constructor calls like new AtomicInteger(0) cause a volatile write that
can be saved in cases where the constructor parameter is the default
value.
See gh-23575
Previously, all MeterFilter beans were applied to all MeterRegistry
beans. As a result, when a composite registry was auto-configured, both
the composite and all of its delegates would have the same MeterFilters
applied. This made it impossible for one of the delegate registries to
have a locally-configured filter that would allow a meter that would be
denied by one of the MeterFilter beans applied to the composite.
This commit update MeterRegistryConfigurer to skips the auto-configured
composite meter registry when applying MeterFilter beans to
MeterRegistry beans. As a result, the composite's filters will no
longer deny a meter before it reaches a delegate that would have
accepted it due to one of its locally-configured filters.
Closes gh-23381
With the introduction of health indicators that only require the
CqlSession, this commit deprecates the health indicators that require
Spring Data since the latter build on top of the former.
Closes gh-23226
Prior to this commit, Spring Boot would auto-configure both
Elasticsearch variants: `RestClient` ("Low Level" client) and
`RestHighLevelClient` ("High Level" client).
Since one can be derived from the other, this would create complex and
unclear situations depending on what developers provided with their
configuration.
`RestHighLevelClient` is mostly for actual use of the Elasticsearch API,
with support for specific methods and (de)serialization. On the other
hand, `RestClient` is merely wrapping the Apache HTTP client for
load-balancing support and low level HTTP features.
This commit completely removes the support for `RestClient` in Spring
Boot and now requires the presence of the
`org.elasticsearch.client:elasticsearch-rest-high-level-client`
dependency for REST client support with Elasticsearch.
Closes gh-22358
This commit builds on top of gh-22603 and exposes data collected by the
`BufferingApplicationStartup` on a dedicated `"/startup"` Actuator
endpoint.
Closes gh-23213
Rename `CompositeMeterRegistryAutoConfiguration` to
`MeterRegistryAutoConfiguration` since it can also create non-composite
registries.
Closes gh-22988
Update metrics auto-configurations so that they are auto-configured
after `CompositeMeterRegistryAutoConfiguration` in order to ensure
the `MeterRegistry` bean has been defined.
Prior to this commit, metrics auto-configurations that depended on a
`MeterRegistry` has `@AutoConfigureAfter(MetricsAutoConfiguration.class)`
which is not sufficient since `MetricsAutoConfiguration` does not export
a `MeterRegistry`.
See gh-21134
Previously, the endpoints' responses could occasionally exceed
WebClient's in-memory buffer limt, for example if the threads endpoint
was reporting a large number of threads or the threads had large
stacks.
This commit disables WebClient's in-memory buffer size limit so that
the tests passing is not dependent on the size of the endpoints'
responses.
Closes gh-22743
Prior to this commit, the livenessState and readinessState health
indicators would not be configured automatically and would be missing
from the "liveness" and "readiness" health groups, leading to 404s when
hitting the `/actuator/health/liveness` or `/actuator/health/readiness`.
This commit ensures that the health indicators beans have the proper
name and revisits the auto-configuration conditions to reflect that as
well.
Fixes gh-22562
This commit replaces the Neo4j-OGM based health checks with one based on
the Neo4j Java driver. A Reactive variant is also added in this commit.
See gh-22302
This commit harmonizes dependency declarations for Jackson in the
actuator. Both Jackson and JSR 310 are back to optional in the core
actuator module and mandatory when using the auto-configuration.
Closes gh-22624
The generated snippets were not declared as an output to the
`spring-boot-actuator-autoconfigure` test goal so when the test was
pulled from the cache, it did not contain the `generated-snippets`
directory. This directory is required as an input to the Asciidoctor
plugin.
See gh-22555
This commit introduces a new property to globally disable metrics
export. In integration tests, this property is automatically set to
disable everything but in-memory metrics.
This commit also introduces a `@AutoConfigureMetrics` annotation that
can be used for integration tests that require metrics export to operate
as they would in an application.
See gh-21658
Prior to this commit, the application availability infrastructure
would mix the `AvailabilityState`, the `HealthIndicator` and the
`HealthGroup` concepts and would not align with the rest.
This commit auto-configures the livenessState and readinessState
health indicators with the relevant configuration properties.
Unlike other indicators, they are not enabled by default but might
be in future versions.
This also moves the `management.health.probes.enabled` property
to `management.endpoint.health.probes.enabled` since "probes" here
is not a health indicator but rather a configuration flag for the
health endpoint.
Finally, the probes auto-configuration is refined to automatically
add liveness and readiness indicators for the probes group if
they're not already present.
Closes gh-22107
Prior to this commit, configuring a reactive Elasticsearch client would
auto-configure an Actuator Health check using a synchronous client, with
the default configuration properties (so tarting localhost:9200).
This would lead to false reports of unhealthy Elasticsearch clusters
when using reactive clients.
This commit reproduces the logic for MongoDB repositories: if a reactive
variant is available, it is selected for the health check
infrastructure.
See gh-21042
Update `Group` properties so that the `showDetails` value does not
inherit `Show.NEVER`. Prior to this commit, the `Group` properties
would not correctly inherit a `showDetails` value from the main
`management.endpoint.health.show-details` property.
See gh-22022
Previously, Spring Boot's modules published Gradle Module Metadata
(GMM) the declared a platform dependency on spring-boot-dependencies.
This provided versions for each module's own dependencies but also had
they unwanted side-effect of pulling in spring-boot-dependencies
constraints which would influence the version of other dependencies
declared in the same configuration. This was undesirable as users
should be able to opt in to this level of dependency management, either
by using the dependency management plugin or by using Gradle's built-in
support via a platform dependency on spring-boot-dependencies.
This commit reworks how Spring Boot's build uses
spring-boot-dependencies and spring-boot-parent to provide its own
dependency management. Configurations that aren't seen by consumers are
configured to extend a dependencyManagement configuration that has an
enforced platform dependency on spring-boot-parent. This enforces
spring-boot-parent's version constraints on Spring Boot's build without
making them visible to consumers. To ensure that the versions that
Spring Boot has been built against are visible to consumers, the
Maven publication that produces pom files and GMM for the published
modules is configured to use the resolved versions from the module's
runtime classpath.
Fixes gh-21911
This commit provides a CassandraDriverHealthIndicator and
CassandraDriverReactiveHealthIndicator that do not require Spring Data.
As a result, a health indicator for Cassandra is provided even if the
application does not use Spring Data.
See gh-20887
This commit updates the logic for handling binding exceptions in the
management context when it is separate from the application context.
The changes allow the exception details to be visible to
DefaultErrorAttributes without causing the servlet container to
detect an error condition.
Fixes gh-21036
Update `EndpointDiscoverer` so that `@Endpoint` and `@EndpointExtension`
beans are created as late as possible.
Prior to this commit, endpoint beans and extension beans would be
created during the discovery phase which could cause early bean
initialization. The problem was especially nasty when using an embedded
servlet container since `ServletEndpointRegistrar` is loaded as the
container is initialized. This would trigger discovery and load all
endpoint beans, including the health endpoint, and all health indicator
beans.
Fixes gh-20714
Prior to this commit, details about an exception would get dropped when
the management context was separate from the application context and
an actuator endpoint threw a binding exception.
This commit adds some logic to capture the exception so the management
context error handlers can add the appropriate attributes to the error
response.
Fixes gh-21036
This commit improves the backward-compatibility of the ErrorAttributes
interfaces by providing a default implementation of a new method. It
also encapsulates several parameters that control the inclusion or
exclusion of error attributes into a new ErrorAttributeOptions type to
make it easier and less intrusive to add additional options in the
future. This encapsulation also makes the handling of the
includeException option more similar to other options.
Fixes gh-21324
Prior to this commit, there was a property server.error.include-details
that allowed configuration of the message and errors attributes in a
server error response.
This commit separates the control of the message and errors attributes
into two separate properties named server.error.include-message and
server.error.include-binding-errors. When the message attribute is
excluded from a servlet response, the value is changed from a
hard-coded text value to an empty value.
Fixes gh-20505
Prior to this commit, there was a cycle between `StatusAggregator` and
`SimpleStatusAggregator`, which caused a static initialization bug -
depending on which class (the implementation or its interface) was
loaded first.
This commit turns the static field of the `StatusAggregator` interface
into a static method to avoid this problem.
Fixes gh-21211
This commit removes the support for Reactor Netty metrics since it
seems that Spring Boot should not use this feature:
* HTTP metrics are already covered by WebFlux
* TCP metrics are only meant to TCP server/clients
* allocator metrics are already provided by Netty and there is
no specific API to enable them here.
Closes gh-19388
This commit enables the production of TCP and buffer allocator metrics
for Reactor Netty, client and server.
When applications use auto-configured server
(`NettyReactiveWebServerFactory`) and client (through
`WebClient.Builder`) instances, metrics will be enabled.
Note that HTTP metrics are not enabled here, since similar metrics are
already produced at the WebFlux level. Also, to avoid cardinality
explosion of metrics (through the URI tag), Reactor Netty offers
configurable infrastructure to deduplicate URI tags by turning expanded
URI instances into templated URIs. This is not targeted for Spring
usage.
Closes gh-19388
Update `ConditionalOnAvailableEndpoint` so that it now uses the same
matching code as the endpoint filter. This allows the condition to
match endpoint IDs that contain a dash.
In order to share logic, the `ExposeExcludePropertyEndpointFilter` class
has been deprecated and its logic moved to a new `expose` package
under `IncludExcludeEndpointFilter`. This filter is used by both the
`OnAvailableEndpointCondition` and the auto-configuration classes.
Fixes gh-21044
Prior to this commit, the ManagementErrorEndpoint used to handle error
responses for the management servlet excluded stacktrace and exception
message details from the response unconditionally.
With this commit, the endpoint honors the
`server.error.include-stacktrace` and `server.error.include-details`
properties to conditionally include error details for consistency
with non-management error handling.
Fixes gh-20989
Prior to this commit, default error responses included the message
from a handled exception. When the exception was a BindException, the
error responses could also include an errors attribute containing the
details of the binding failure. These details could leak information
about the application.
This commit removes the exception message and binding errors detail
from error responses by default, and introduces a
`server.error.include-details` property that can be used to cause
these details to be included in the response.
Fixes gh-20505
Update the `HealthEndpointGroups` customization support to use a
post-processor rather than a mutable registry. Although this approach
is slightly less flexible, it removes a lot of complexity from the
`HealthEndpointGroups` code. Specifically, it allows us to drop the
`HealthEndpointGroupsRegistry` interface entirely.
The probe health groups are now added via the post-processor if they
aren't already defined. Unlike the previous implementation, users are
no longer able to customize status aggregation and http status code
mapping rules _unless_ they also re-define the health indicators that
are members of the group.
See gh-20962
Update `AvailabilityProbesAutoConfiguration` to allow the
`management.health.probes.enabled` property to override the platform
detection logic. Prior to this commit, it was possible to use the
property to enable the probes, but it was not possible to disable
them when deploying to Kubernates.
See gh-20962
Relocate probe auto-configuration from the `kubernetes` package to
`availability` since probes could also be used on other platforms.
The classes have also been renamed to named to `AvailabilityProbes...`
See gh-20962
Rename `LivenessProbeHealthIndicator` to `LivenessStateHealthIndicator`
and `ReadinessProbeHealthIndicator` to `ReadinessStateHealthIndicator`.
Also introduce a general purpose `AvailabilityStateHealthIndicator`
class.
See gh-20962
Create a general purpose `AvailabilityState` interface and refactor
the existing `LivenessState` and `ReadinessState` to use it. A single
`AvailabilityChangeEvent` is now used to carry all availability state
updates.
This commit also renames `ApplicationAvailabilityProvider` to
`ApplicationAvailabilityBean` and extracts an `ApplicationAvailability`
interface that other beans can inject. The helps to hide the event
listener method, which is really internal.
Finally the state enums have been renamed as follows:
- `LivenessState.LIVE` -> `LivenessState.CORRECT`
- `ReadinessState.READY` -> `ReadinessState.ACCEPTING_TRAFFIC`
- `ReadinessState.UNREADY` -> `ReadinessState.REFUSING_TRAFFIC`
See gh-20962
This commit removes the duplication added temporarily in Spring Boot to
initialize a default WavefrontSender builder.
This commit also disables a test is failing at the moment, see
https://github.com/micrometer-metrics/micrometer/issues/1964
See gh-20854
This commit upgrades the Wavefront metrics export auto-configuration to
provide a `WavefrontSender` if necessary and use that to export metrics
rather than the http client Micrometer used previously.
As a result, the "read-timeout" and "connect-timeout" properties are no
longer honoured.
Closes gh-20810
StatsD no longer publishes metrics about itself and StatsDMetrics is
deprecated as a result. This commit removes the auto-configuration of
it.
Closes gh-20836
This commit exposes an additional property for Graphite that allows to
restore the previous default behaviour with regards to tags, i.e.
prefixing the ones defined by the "tagsAsPrefix" property.
Close gh-20834
This property is a left-over and was never used in Micrometer so this
commit deprecates its use so that it can be removed in the next feature
release.
Closes gh-20835
This commit fixes the AtlasProperties hierarchy so that it no longer is
a StepRegistryProperties. The AtlasConfig on the Micrometer side of
things does not share the common config hierarchy either and some
properties have different default and lifecycle.
Closes gh-20843
Prior to this commit, we were relying on the
`"spring.main.cloud-platform"` property for overriding cloud platform
detection and enabling liveness and readiness probes. Changes made in
gh-20553 have now been reverted.
This commit adds the `"management.health.probes.enabled"` configuration
property. The auto-configuration now enables the HTTP Probes and
`HealthIndicator` if this property is enabled, or if the Kubernetes
cloud platform is detected.
This property is `false` by default for now, since enabling this for all
Spring Boot applications would be a breaking change. In this case, the
global `"/actuator/health"` endpoint could report `OUT_OF_SERVICE`
during startup time because the application now reports the readiness as
well.
See gh-19593
This commit moves the core Liveness and Readiness support to its own
`availability` package. We've made this a core concept independent of
Kubernetes.
Spring Boot now produces `LivenessStateChanged` and
`ReadinessStateChanged` events as part of the typical application
lifecycle.
Liveness and Readiness Probes (`HealthIndicator` components and health
groups) are still configured only when deployed on Kubernetes.
This commit also improves the documentation around Probes best practices
and container lifecycle considerations.
See gh-19593
Prior to this commit and as of Spring Boot 2.2.0, we would advise
developers to use the Actuator health groups to define custom "liveness"
and "readiness" groups and configure them with subsets of existing
health indicators.
This commit addresses several limitations with that approach.
First, `LivenessState` and `ReadinessState` are promoted to first class
concepts in Spring Boot applications. These states should not only based
on periodic health checks. Applications should be able to track changes
(and adapt their behavior) or update states (when an error happens).
The `ApplicationStateProvider` can be injected and used by applications
components to get the current application state. Components can also
track specific `ApplicationEvent` to be notified of changes, like
`ReadinessStateChangedEvent` and `LivenessStateChangedEvent`.
Components can also publish such events with an
`ApplicationEventPublisher`. Spring Boot will track startup event and
application context state to update the liveness and readiness state of
the application. This infrastructure is available in the
main spring-boot module.
If Spring Boot Actuator is on the classpath, additional
`HealthIndicator` will be contributed to the application:
`"LivenessProveHealthIndicator"` and `"ReadinessProbeHealthIndicator"`.
Also, "liveness" and "readiness" Health groups will be defined if
they're not configured already.
Closes gh-19593
Prior to this commit, `HealthContributor` would be exposed under the
main `HealthEndpoint` and subgroups, `HealthEndpointGroups`. Groups are
driven by configuration properties and there was no way to contribute
programmatically new groups.
This commit introduces the `HealthEndpointGroupsRegistry` (a mutable
version of `HealthEndpointGroups`) and a
`HealthEndpointGroupsRegistryCustomizer`. This allows configurations to
add/remove groups during Actuator auto-configuration.
Closes gh-20554
This commit relaxes the class condition to ConnectionFactory, checking
if the connection factory is a connection pool and bind its metrics to
the registry accordingly.
Closes gh-20349
This commit adds metrics support for `ConnectionPool` beans.
See gh-19988
Co-authored-by: Mark Paluch <mpaluch@pivotal.io>
Co-authored-by: Tadaya Tsuyukubo <tadaya@ttddyy.net>
This commit adds an health indicator for R2DBC. If a validation query is
provided, it is used to validate the state of the database. If not, a
check of the connection is issued.
See gh-19988
Co-authored-by: Mark Paluch <mpaluch@pivotal.io>
Prior to this commit, Actuator endpoints would use the application
ObjectMapper instance for serializing payloads as JSON. This was
problematic in several cases:
* application-specific configuration would change the actuator endpoint
output.
* choosing a different JSON mapper implementation in the application
would break completely some endpoints.
Spring Boot Actuator already has a hard dependency on Jackson, and this
commit uses that fact to configure a shared `ObjectMapper` instance that
will be used by the Actuator infrastructure consistently, without
polluting the application context.
This `ObjectMapper` is used in Actuator for:
* JMX endpoints
* Spring MVC endpoints with an HTTP message converter
* Spring WebFlux endpoints with an `Encoder`
* Jersey endpoints with a `ContextResolver<ObjectMapper>`
For all web endpoints, this configuration is limited to the
actuator-specific media types such as
`"application/vnd.spring-boot.actuator.v3+json"`.
Fixes gh-12951
This partially re-applies the deprecation removal from commit
df1837a16b,
without removing CompositeHealthIndicator, HealthAggregator, and related
configuration that is required by Spring Cloud.
For consistency with SpringApplication, this commit disables bean
overriding by default in ApplicationContextRunner. Bean overriding can
be enabled again using withAllowBeanDefinitionOverriding.
Closes gh-18019
Update all dependencies declarations to use the form `scope(reference)`
rather than `scope reference`.
Prior to this commit we declared dependencies without parentheses unless
we were forced to add them due to an `exclude`.
Replace Gradle single quote strings with the double quote form
whenever possible. The change helps to being consistency to the
dependencies section where mostly single quotes were used, but
occasionally double quotes were required due to `${}` references.
Pedro Ivo Machado
Andy Wilkinson
Stephane Nicoll
Phillip Webb
Pascal Schumacher
bono007
Madhura Bhave
izeye
Phillip Webb
dreis2211
Artem Bilan
Stephane Nicoll
Andy Wilkinson
Brian Clozel
Pradipta Sarma
Martin Benda
smlc
AndrewDi
Emily Tsanova
Gerrit Meier
Nelson Osacky
Julio Gomez Diaz
Aleksander Lech
Eddú Meléndez
Lee Dobryden
Leo Li
Alexandre Dutra
Scott Frederick
Brian Clozel
Jon Schneider
Vedran Pavic
Andreas Born
Dave Syer
babjo
Christoph Strobl
wonwoo