This section goes into more detail about how you should use Spring Boot. It covers topics such as build systems, auto-configuration, and how to run your applications. We also cover some Spring Boot best practices. Although there is nothing particularly special about Spring Boot (it is just another library that you can consume), there are a few recommendations that, when followed, make your development process a little easier.
If you are starting out with Spring Boot, you should probably read the Getting Started guide before diving into this section.
1. Build Systems
It is strongly recommended that you choose a build system that supports dependency management and that can consume artifacts published to the “Maven Central” repository. We would recommend that you choose Maven or Gradle. It is possible to get Spring Boot to work with other build systems (Ant, for example), but they are not particularly well supported.
1.1. Dependency Management
Each release of Spring Boot provides a curated list of dependencies that it supports. In practice, you do not need to provide a version for any of these dependencies in your build configuration, as Spring Boot manages that for you. When you upgrade Spring Boot itself, these dependencies are upgraded as well in a consistent way.
You can still specify a version and override Spring Boot’s recommendations if you need to do so. |
The curated list contains all the Spring modules that you can use with Spring Boot as well as a refined list of third party libraries.
The list is available as a standard Bills of Materials (spring-boot-dependencies
) that can be used with both Maven and Gradle.
Each release of Spring Boot is associated with a base version of the Spring Framework. We highly recommend that you do not specify its version. |
1.2. Maven
To learn about using Spring Boot with Maven, see the documentation for Spring Boot’s Maven plugin:
1.3. Gradle
To learn about using Spring Boot with Gradle, see the documentation for Spring Boot’s Gradle plugin:
1.4. Ant
It is possible to build a Spring Boot project using Apache Ant+Ivy.
The spring-boot-antlib
“AntLib” module is also available to help Ant create executable jars.
To declare dependencies, a typical ivy.xml
file looks something like the following example:
<ivy-module version="2.0">
<info organisation="org.springframework.boot" module="spring-boot-sample-ant" />
<configurations>
<conf name="compile" description="everything needed to compile this module" />
<conf name="runtime" extends="compile" description="everything needed to run this module" />
</configurations>
<dependencies>
<dependency org="org.springframework.boot" name="spring-boot-starter"
rev="${spring-boot.version}" conf="compile" />
</dependencies>
</ivy-module>
A typical build.xml
looks like the following example:
<project
xmlns:ivy="antlib:org.apache.ivy.ant"
xmlns:spring-boot="antlib:org.springframework.boot.ant"
name="myapp" default="build">
<property name="spring-boot.version" value="3.2.5" />
<target name="resolve" description="--> retrieve dependencies with ivy">
<ivy:retrieve pattern="lib/[conf]/[artifact]-[type]-[revision].[ext]" />
</target>
<target name="classpaths" depends="resolve">
<path id="compile.classpath">
<fileset dir="lib/compile" includes="*.jar" />
</path>
</target>
<target name="init" depends="classpaths">
<mkdir dir="build/classes" />
</target>
<target name="compile" depends="init" description="compile">
<javac srcdir="src/main/java" destdir="build/classes" classpathref="compile.classpath" />
</target>
<target name="build" depends="compile">
<spring-boot:exejar destfile="build/myapp.jar" classes="build/classes">
<spring-boot:lib>
<fileset dir="lib/runtime" />
</spring-boot:lib>
</spring-boot:exejar>
</target>
</project>
If you do not want to use the spring-boot-antlib module, see the howto.html “How-to” .
|
1.5. Starters
Starters are a set of convenient dependency descriptors that you can include in your application.
You get a one-stop shop for all the Spring and related technologies that you need without having to hunt through sample code and copy-paste loads of dependency descriptors.
For example, if you want to get started using Spring and JPA for database access, include the spring-boot-starter-data-jpa
dependency in your project.
The starters contain a lot of the dependencies that you need to get a project up and running quickly and with a consistent, supported set of managed transitive dependencies.
The following application starters are provided by Spring Boot under the org.springframework.boot
group:
Name | Description |
---|---|
Core starter, including auto-configuration support, logging and YAML |
|
Starter for JMS messaging using Apache ActiveMQ |
|
Starter for using Spring AMQP and Rabbit MQ |
|
Starter for aspect-oriented programming with Spring AOP and AspectJ |
|
Starter for JMS messaging using Apache Artemis |
|
Starter for using Spring Batch |
|
Starter for using Spring Framework’s caching support |
|
Starter for using Cassandra distributed database and Spring Data Cassandra |
|
Starter for using Cassandra distributed database and Spring Data Cassandra Reactive |
|
Starter for using Couchbase document-oriented database and Spring Data Couchbase |
|
Starter for using Couchbase document-oriented database and Spring Data Couchbase Reactive |
|
Starter for using Elasticsearch search and analytics engine and Spring Data Elasticsearch |
|
Starter for using Spring Data JDBC |
|
Starter for using Spring Data JPA with Hibernate |
|
Starter for using Spring Data LDAP |
|
Starter for using MongoDB document-oriented database and Spring Data MongoDB |
|
Starter for using MongoDB document-oriented database and Spring Data MongoDB Reactive |
|
Starter for using Neo4j graph database and Spring Data Neo4j |
|
Starter for using Spring Data R2DBC |
|
Starter for using Redis key-value data store with Spring Data Redis and the Lettuce client |
|
Starter for using Redis key-value data store with Spring Data Redis reactive and the Lettuce client |
|
Starter for exposing Spring Data repositories over REST using Spring Data REST and Spring MVC |
|
Starter for building MVC web applications using FreeMarker views |
|
Starter for building GraphQL applications with Spring GraphQL |
|
Starter for building MVC web applications using Groovy Templates views |
|
Starter for building hypermedia-based RESTful web application with Spring MVC and Spring HATEOAS |
|
Starter for using Spring Integration |
|
Starter for using JDBC with the HikariCP connection pool |
|
Starter for building RESTful web applications using JAX-RS and Jersey. An alternative to |
|
Starter for using jOOQ to access SQL databases with JDBC. An alternative to |
|
Starter for reading and writing json |
|
Starter for using Java Mail and Spring Framework’s email sending support |
|
Starter for building web applications using Mustache views |
|
Starter for using Spring Authorization Server features |
|
Starter for using Spring Security’s OAuth2/OpenID Connect client features |
|
Starter for using Spring Security’s OAuth2 resource server features |
|
Starter for using Spring for Apache Pulsar |
|
Starter for using Spring for Apache Pulsar Reactive |
|
Starter for using the Quartz scheduler |
|
Starter for building RSocket clients and servers |
|
Starter for using Spring Security |
|
Starter for testing Spring Boot applications with libraries including JUnit Jupiter, Hamcrest and Mockito |
|
Starter for building MVC web applications using Thymeleaf views |
|
Starter for using Java Bean Validation with Hibernate Validator |
|
Starter for building web, including RESTful, applications using Spring MVC. Uses Tomcat as the default embedded container |
|
Starter for using Spring Web Services |
|
Starter for building WebFlux applications using Spring Framework’s Reactive Web support |
|
Starter for building WebSocket applications using Spring Framework’s MVC WebSocket support |
In addition to the application starters, the following starters can be used to add production ready features:
Name | Description |
---|---|
Starter for using Spring Boot’s Actuator which provides production ready features to help you monitor and manage your application |
Finally, Spring Boot also includes the following starters that can be used if you want to exclude or swap specific technical facets:
Name | Description |
---|---|
Starter for using Jetty as the embedded servlet container. An alternative to |
|
Starter for using Log4j2 for logging. An alternative to |
|
Starter for logging using Logback. Default logging starter |
|
Starter for using Reactor Netty as the embedded reactive HTTP server. |
|
Starter for using Tomcat as the embedded servlet container. Default servlet container starter used by |
|
Starter for using Undertow as the embedded servlet container. An alternative to |
To learn how to swap technical facets, please see the how-to documentation for swapping web server and logging system.
For a list of additional community contributed starters, see the README file in the spring-boot-starters module on GitHub.
|
2. Structuring Your Code
Spring Boot does not require any specific code layout to work. However, there are some best practices that help.
If you wish to enforce a structure based on domains, take a look at Spring Modulith. |
2.1. Using the “default” Package
When a class does not include a package
declaration, it is considered to be in the “default package”.
The use of the “default package” is generally discouraged and should be avoided.
It can cause particular problems for Spring Boot applications that use the @ComponentScan
, @ConfigurationPropertiesScan
, @EntityScan
, or @SpringBootApplication
annotations, since every class from every jar is read.
We recommend that you follow Java’s recommended package naming conventions and use a reversed domain name (for example, com.example.project ).
|
2.2. Locating the Main Application Class
We generally recommend that you locate your main application class in a root package above other classes.
The @SpringBootApplication
annotation is often placed on your main class, and it implicitly defines a base “search package” for certain items.
For example, if you are writing a JPA application, the package of the @SpringBootApplication
annotated class is used to search for @Entity
items.
Using a root package also allows component scan to apply only on your project.
If you do not want to use @SpringBootApplication , the @EnableAutoConfiguration and @ComponentScan annotations that it imports defines that behavior so you can also use those instead.
|
The following listing shows a typical layout:
com +- example +- myapplication +- MyApplication.java | +- customer | +- Customer.java | +- CustomerController.java | +- CustomerService.java | +- CustomerRepository.java | +- order +- Order.java +- OrderController.java +- OrderService.java +- OrderRepository.java
The MyApplication.java
file would declare the main
method, along with the basic @SpringBootApplication
, as follows:
@SpringBootApplication
public class MyApplication {
public static void main(String[] args) {
SpringApplication.run(MyApplication.class, args);
}
}
@SpringBootApplication
class MyApplication
fun main(args: Array<String>) {
runApplication<MyApplication>(*args)
}
3. Configuration Classes
Spring Boot favors Java-based configuration.
Although it is possible to use SpringApplication
with XML sources, we generally recommend that your primary source be a single @Configuration
class.
Usually the class that defines the main
method is a good candidate as the primary @Configuration
.
Many Spring configuration examples have been published on the Internet that use XML configuration.
If possible, always try to use the equivalent Java-based configuration.
Searching for Enable* annotations can be a good starting point.
|
3.1. Importing Additional Configuration Classes
You need not put all your @Configuration
into a single class.
The @Import
annotation can be used to import additional configuration classes.
Alternatively, you can use @ComponentScan
to automatically pick up all Spring components, including @Configuration
classes.
4. Auto-configuration
Spring Boot auto-configuration attempts to automatically configure your Spring application based on the jar dependencies that you have added.
For example, if HSQLDB
is on your classpath, and you have not manually configured any database connection beans, then Spring Boot auto-configures an in-memory database.
You need to opt-in to auto-configuration by adding the @EnableAutoConfiguration
or @SpringBootApplication
annotations to one of your @Configuration
classes.
You should only ever add one @SpringBootApplication or @EnableAutoConfiguration annotation.
We generally recommend that you add one or the other to your primary @Configuration class only.
|
4.1. Gradually Replacing Auto-configuration
Auto-configuration is non-invasive.
At any point, you can start to define your own configuration to replace specific parts of the auto-configuration.
For example, if you add your own DataSource
bean, the default embedded database support backs away.
If you need to find out what auto-configuration is currently being applied, and why, start your application with the --debug
switch.
Doing so enables debug logs for a selection of core loggers and logs a conditions report to the console.
4.2. Disabling Specific Auto-configuration Classes
If you find that specific auto-configuration classes that you do not want are being applied, you can use the exclude attribute of @SpringBootApplication
to disable them, as shown in the following example:
@SpringBootApplication(exclude = { DataSourceAutoConfiguration.class })
public class MyApplication {
}
@SpringBootApplication(exclude = [DataSourceAutoConfiguration::class])
class MyApplication
If the class is not on the classpath, you can use the excludeName
attribute of the annotation and specify the fully qualified name instead.
If you prefer to use @EnableAutoConfiguration
rather than @SpringBootApplication
, exclude
and excludeName
are also available.
Finally, you can also control the list of auto-configuration classes to exclude by using the spring.autoconfigure.exclude
property.
You can define exclusions both at the annotation level and by using the property. |
Even though auto-configuration classes are public , the only aspect of the class that is considered public API is the name of the class which can be used for disabling the auto-configuration.
The actual contents of those classes, such as nested configuration classes or bean methods are for internal use only and we do not recommend using those directly.
|
4.3. Auto-configuration Packages
Auto-configuration packages are the packages that various auto-configured features look in by default when scanning for things such as entities and Spring Data repositories.
The @EnableAutoConfiguration
annotation (either directly or through its presence on @SpringBootApplication
) determines the default auto-configuration package.
Additional packages can be configured using the @AutoConfigurationPackage
annotation.
5. Spring Beans and Dependency Injection
You are free to use any of the standard Spring Framework techniques to define your beans and their injected dependencies.
We generally recommend using constructor injection to wire up dependencies and @ComponentScan
to find beans.
If you structure your code as suggested above (locating your application class in a top package), you can add @ComponentScan
without any arguments or use the @SpringBootApplication
annotation which implicitly includes it.
All of your application components (@Component
, @Service
, @Repository
, @Controller
, and others) are automatically registered as Spring Beans.
The following example shows a @Service
Bean that uses constructor injection to obtain a required RiskAssessor
bean:
@Service
public class MyAccountService implements AccountService {
private final RiskAssessor riskAssessor;
public MyAccountService(RiskAssessor riskAssessor) {
this.riskAssessor = riskAssessor;
}
// ...
}
@Service
class MyAccountService(private val riskAssessor: RiskAssessor) : AccountService
If a bean has more than one constructor, you will need to mark the one you want Spring to use with @Autowired
:
@Service
public class MyAccountService implements AccountService {
private final RiskAssessor riskAssessor;
private final PrintStream out;
@Autowired
public MyAccountService(RiskAssessor riskAssessor) {
this.riskAssessor = riskAssessor;
this.out = System.out;
}
public MyAccountService(RiskAssessor riskAssessor, PrintStream out) {
this.riskAssessor = riskAssessor;
this.out = out;
}
// ...
}
@Service
class MyAccountService : AccountService {
private val riskAssessor: RiskAssessor
private val out: PrintStream
@Autowired
constructor(riskAssessor: RiskAssessor) {
this.riskAssessor = riskAssessor
out = System.out
}
constructor(riskAssessor: RiskAssessor, out: PrintStream) {
this.riskAssessor = riskAssessor
this.out = out
}
// ...
}
Notice how using constructor injection lets the riskAssessor field be marked as final , indicating that it cannot be subsequently changed.
|
6. Using the @SpringBootApplication Annotation
Many Spring Boot developers like their apps to use auto-configuration, component scan and be able to define extra configuration on their "application class".
A single @SpringBootApplication
annotation can be used to enable those three features, that is:
-
@EnableAutoConfiguration
: enable Spring Boot’s auto-configuration mechanism -
@ComponentScan
: enable@Component
scan on the package where the application is located (see the best practices) -
@SpringBootConfiguration
: enable registration of extra beans in the context or the import of additional configuration classes. An alternative to Spring’s standard@Configuration
that aids configuration detection in your integration tests.
// Same as @SpringBootConfiguration @EnableAutoConfiguration @ComponentScan
@SpringBootApplication
public class MyApplication {
public static void main(String[] args) {
SpringApplication.run(MyApplication.class, args);
}
}
// same as @SpringBootConfiguration @EnableAutoConfiguration @ComponentScan
@SpringBootApplication
class MyApplication
fun main(args: Array<String>) {
runApplication<MyApplication>(*args)
}
@SpringBootApplication also provides aliases to customize the attributes of @EnableAutoConfiguration and @ComponentScan .
|
None of these features are mandatory and you may choose to replace this single annotation by any of the features that it enables. For instance, you may not want to use component scan or configuration properties scan in your application: Java
Kotlin
In this example, |
7. Running Your Application
One of the biggest advantages of packaging your application as a jar and using an embedded HTTP server is that you can run your application as you would any other. The sample applies to debugging Spring Boot applications. You do not need any special IDE plugins or extensions.
This section only covers jar-based packaging. If you choose to package your application as a war file, see your server and IDE documentation. |
7.1. Running From an IDE
You can run a Spring Boot application from your IDE as a Java application.
However, you first need to import your project.
Import steps vary depending on your IDE and build system.
Most IDEs can import Maven projects directly.
For example, Eclipse users can select Import…
→ Existing Maven Projects
from the File
menu.
If you cannot directly import your project into your IDE, you may be able to generate IDE metadata by using a build plugin. Maven includes plugins for Eclipse and IDEA. Gradle offers plugins for various IDEs.
If you accidentally run a web application twice, you see a “Port already in use” error.
Spring Tools users can use the Relaunch button rather than the Run button to ensure that any existing instance is closed.
|
7.2. Running as a Packaged Application
If you use the Spring Boot Maven or Gradle plugins to create an executable jar, you can run your application using java -jar
, as shown in the following example:
$ java -jar target/myapplication-0.0.1-SNAPSHOT.jar
It is also possible to run a packaged application with remote debugging support enabled. Doing so lets you attach a debugger to your packaged application, as shown in the following example:
$ java -agentlib:jdwp=server=y,transport=dt_socket,address=8000,suspend=n \
-jar target/myapplication-0.0.1-SNAPSHOT.jar
7.3. Using the Maven Plugin
The Spring Boot Maven plugin includes a run
goal that can be used to quickly compile and run your application.
Applications run in an exploded form, as they do in your IDE.
The following example shows a typical Maven command to run a Spring Boot application:
$ mvn spring-boot:run
You might also want to use the MAVEN_OPTS
operating system environment variable, as shown in the following example:
$ export MAVEN_OPTS=-Xmx1024m
7.4. Using the Gradle Plugin
The Spring Boot Gradle plugin also includes a bootRun
task that can be used to run your application in an exploded form.
The bootRun
task is added whenever you apply the org.springframework.boot
and java
plugins and is shown in the following example:
$ gradle bootRun
You might also want to use the JAVA_OPTS
operating system environment variable, as shown in the following example:
$ export JAVA_OPTS=-Xmx1024m
7.5. Hot Swapping
Since Spring Boot applications are plain Java applications, JVM hot-swapping should work out of the box. JVM hot swapping is somewhat limited with the bytecode that it can replace. For a more complete solution, JRebel can be used.
The spring-boot-devtools
module also includes support for quick application restarts.
See the Hot swapping “How-to” for details.
8. Developer Tools
Spring Boot includes an additional set of tools that can make the application development experience a little more pleasant.
The spring-boot-devtools
module can be included in any project to provide additional development-time features.
To include devtools support, add the module dependency to your build, as shown in the following listings for Maven and Gradle:
<dependencies>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-devtools</artifactId>
<optional>true</optional>
</dependency>
</dependencies>
dependencies {
developmentOnly("org.springframework.boot:spring-boot-devtools")
}
Devtools might cause classloading issues, in particular in multi-module projects. Diagnosing Classloading Issues explains how to diagnose and solve them. |
Developer tools are automatically disabled when running a fully packaged application.
If your application is launched from java -jar or if it is started from a special classloader, then it is considered a “production application”.
You can control this behavior by using the spring.devtools.restart.enabled system property.
To enable devtools, irrespective of the classloader used to launch your application, set the -Dspring.devtools.restart.enabled=true system property.
This must not be done in a production environment where running devtools is a security risk.
To disable devtools, exclude the dependency or set the -Dspring.devtools.restart.enabled=false system property.
|
Flagging the dependency as optional in Maven or using the developmentOnly configuration in Gradle (as shown above) prevents devtools from being transitively applied to other modules that use your project.
|
Repackaged archives do not contain devtools by default.
If you want to use a certain remote devtools feature, you need to include it.
When using the Maven plugin, set the excludeDevtools property to false .
When using the Gradle plugin, configure the task’s classpath to include the developmentOnly configuration.
|
8.1. Diagnosing Classloading Issues
As described in the Restart vs Reload section, restart functionality is implemented by using two classloaders. For most applications, this approach works well. However, it can sometimes cause classloading issues, in particular in multi-module projects.
To diagnose whether the classloading issues are indeed caused by devtools and its two classloaders, try disabling restart. If this solves your problems, customize the restart classloader to include your entire project.
8.2. Property Defaults
Several of the libraries supported by Spring Boot use caches to improve performance. For example, template engines cache compiled templates to avoid repeatedly parsing template files. Also, Spring MVC can add HTTP caching headers to responses when serving static resources.
While caching is very beneficial in production, it can be counter-productive during development, preventing you from seeing the changes you just made in your application. For this reason, spring-boot-devtools disables the caching options by default.
Cache options are usually configured by settings in your application.properties
file.
For example, Thymeleaf offers the spring.thymeleaf.cache
property.
Rather than needing to set these properties manually, the spring-boot-devtools
module automatically applies sensible development-time configuration.
The following table lists all the properties that are applied:
Name | Default Value |
---|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
If you do not want property defaults to be applied you can set spring.devtools.add-properties to false in your application.properties .
|
Because you need more information about web requests while developing Spring MVC and Spring WebFlux applications, developer tools suggests you to enable DEBUG
logging for the web
logging group.
This will give you information about the incoming request, which handler is processing it, the response outcome, and other details.
If you wish to log all request details (including potentially sensitive information), you can turn on the spring.mvc.log-request-details
or spring.codec.log-request-details
configuration properties.
8.3. Automatic Restart
Applications that use spring-boot-devtools
automatically restart whenever files on the classpath change.
This can be a useful feature when working in an IDE, as it gives a very fast feedback loop for code changes.
By default, any entry on the classpath that points to a directory is monitored for changes.
Note that certain resources, such as static assets and view templates, do not need to restart the application.
If you are restarting with Maven or Gradle using the build plugin you must leave the forking set to enabled .
If you disable forking, the isolated application classloader used by devtools will not be created and restarts will not operate properly.
|
Automatic restart works very well when used with LiveReload. See the LiveReload section for details. If you use JRebel, automatic restarts are disabled in favor of dynamic class reloading. Other devtools features (such as LiveReload and property overrides) can still be used. |
DevTools relies on the application context’s shutdown hook to close it during a restart.
It does not work correctly if you have disabled the shutdown hook (SpringApplication.setRegisterShutdownHook(false) ).
|
DevTools needs to customize the ResourceLoader used by the ApplicationContext .
If your application provides one already, it is going to be wrapped.
Direct override of the getResource method on the ApplicationContext is not supported.
|
Automatic restart is not supported when using AspectJ weaving. |
8.3.1. Logging Changes in Condition Evaluation
By default, each time your application restarts, a report showing the condition evaluation delta is logged. The report shows the changes to your application’s auto-configuration as you make changes such as adding or removing beans and setting configuration properties.
To disable the logging of the report, set the following property:
spring.devtools.restart.log-condition-evaluation-delta=false
spring:
devtools:
restart:
log-condition-evaluation-delta: false
8.3.2. Excluding Resources
Certain resources do not necessarily need to trigger a restart when they are changed.
For example, Thymeleaf templates can be edited in-place.
By default, changing resources in /META-INF/maven
, /META-INF/resources
, /resources
, /static
, /public
, or /templates
does not trigger a restart but does trigger a live reload.
If you want to customize these exclusions, you can use the spring.devtools.restart.exclude
property.
For example, to exclude only /static
and /public
you would set the following property:
spring.devtools.restart.exclude=static/**,public/**
spring:
devtools:
restart:
exclude: "static/**,public/**"
If you want to keep those defaults and add additional exclusions, use the spring.devtools.restart.additional-exclude property instead.
|
8.3.3. Watching Additional Paths
You may want your application to be restarted or reloaded when you make changes to files that are not on the classpath.
To do so, use the spring.devtools.restart.additional-paths
property to configure additional paths to watch for changes.
You can use the spring.devtools.restart.exclude
property described earlier to control whether changes beneath the additional paths trigger a full restart or a live reload.
8.3.4. Disabling Restart
If you do not want to use the restart feature, you can disable it by using the spring.devtools.restart.enabled
property.
In most cases, you can set this property in your application.properties
(doing so still initializes the restart classloader, but it does not watch for file changes).
If you need to completely disable restart support (for example, because it does not work with a specific library), you need to set the spring.devtools.restart.enabled
System
property to false
before calling SpringApplication.run(…)
, as shown in the following example:
@SpringBootApplication
public class MyApplication {
public static void main(String[] args) {
System.setProperty("spring.devtools.restart.enabled", "false");
SpringApplication.run(MyApplication.class, args);
}
}
@SpringBootApplication
object MyApplication {
@JvmStatic
fun main(args: Array<String>) {
System.setProperty("spring.devtools.restart.enabled", "false")
SpringApplication.run(MyApplication::class.java, *args)
}
}
8.3.5. Using a Trigger File
If you work with an IDE that continuously compiles changed files, you might prefer to trigger restarts only at specific times. To do so, you can use a “trigger file”, which is a special file that must be modified when you want to actually trigger a restart check.
Any update to the file will trigger a check, but restart only actually occurs if Devtools has detected it has something to do. |
To use a trigger file, set the spring.devtools.restart.trigger-file
property to the name (excluding any path) of your trigger file.
The trigger file must appear somewhere on your classpath.
For example, if you have a project with the following structure:
src +- main +- resources +- .reloadtrigger
Then your trigger-file
property would be:
spring.devtools.restart.trigger-file=.reloadtrigger
spring:
devtools:
restart:
trigger-file: ".reloadtrigger"
Restarts will now only happen when the src/main/resources/.reloadtrigger
is updated.
You might want to set spring.devtools.restart.trigger-file as a global setting, so that all your projects behave in the same way.
|
Some IDEs have features that save you from needing to update your trigger file manually.
Spring Tools for Eclipse and IntelliJ IDEA (Ultimate Edition) both have such support.
With Spring Tools, you can use the “reload” button from the console view (as long as your trigger-file
is named .reloadtrigger
).
For IntelliJ IDEA, you can follow the instructions in their documentation.
8.3.6. Customizing the Restart Classloader
As described earlier in the Restart vs Reload section, restart functionality is implemented by using two classloaders. If this causes issues, you might need to customize what gets loaded by which classloader.
By default, any open project in your IDE is loaded with the “restart” classloader, and any regular .jar
file is loaded with the “base” classloader.
The same is true if you use mvn spring-boot:run
or gradle bootRun
: the project containing your @SpringBootApplication
is loaded with the “restart” classloader, and everything else with the “base” classloader.
You can instruct Spring Boot to load parts of your project with a different classloader by creating a META-INF/spring-devtools.properties
file.
The spring-devtools.properties
file can contain properties prefixed with restart.exclude
and restart.include
.
The include
elements are items that should be pulled up into the “restart” classloader, and the exclude
elements are items that should be pushed down into the “base” classloader.
The value of the property is a regex pattern that is applied to the classpath, as shown in the following example:
restart.exclude.companycommonlibs=/mycorp-common-[\\w\\d-\\.]+\\.jar
restart.include.projectcommon=/mycorp-myproj-[\\w\\d-\\.]+\\.jar
restart:
exclude:
companycommonlibs: "/mycorp-common-[\\w\\d-\\.]+\\.jar"
include:
projectcommon: "/mycorp-myproj-[\\w\\d-\\.]+\\.jar"
All property keys must be unique.
As long as a property starts with restart.include. or restart.exclude. it is considered.
|
All META-INF/spring-devtools.properties from the classpath are loaded.
You can package files inside your project, or in the libraries that the project consumes.
|
8.3.7. Known Limitations
Restart functionality does not work well with objects that are deserialized by using a standard ObjectInputStream
.
If you need to deserialize data, you may need to use Spring’s ConfigurableObjectInputStream
in combination with Thread.currentThread().getContextClassLoader()
.
Unfortunately, several third-party libraries deserialize without considering the context classloader. If you find such a problem, you need to request a fix with the original authors.
8.4. LiveReload
The spring-boot-devtools
module includes an embedded LiveReload server that can be used to trigger a browser refresh when a resource is changed.
LiveReload browser extensions are freely available for Chrome, Firefox and Safari.
You can find these extensions by searching 'LiveReload' in the marketplace or store of your chosen browser.
If you do not want to start the LiveReload server when your application runs, you can set the spring.devtools.livereload.enabled
property to false
.
You can only run one LiveReload server at a time. Before starting your application, ensure that no other LiveReload servers are running. If you start multiple applications from your IDE, only the first has LiveReload support. |
To trigger LiveReload when a file changes, Automatic Restart must be enabled. |
8.5. Global Settings
You can configure global devtools settings by adding any of the following files to the $HOME/.config/spring-boot
directory:
-
spring-boot-devtools.properties
-
spring-boot-devtools.yaml
-
spring-boot-devtools.yml
Any properties added to these files apply to all Spring Boot applications on your machine that use devtools.
For example, to configure restart to always use a trigger file, you would add the following property to your spring-boot-devtools
file:
spring.devtools.restart.trigger-file=.reloadtrigger
spring:
devtools:
restart:
trigger-file: ".reloadtrigger"
By default, $HOME
is the user’s home directory.
To customize this location, set the SPRING_DEVTOOLS_HOME
environment variable or the spring.devtools.home
system property.
If devtools configuration files are not found in $HOME/.config/spring-boot , the root of the $HOME directory is searched for the presence of a .spring-boot-devtools.properties file.
This allows you to share the devtools global configuration with applications that are on an older version of Spring Boot that does not support the $HOME/.config/spring-boot location.
|
Profiles are not supported in devtools properties/yaml files. Any profiles activated in |
8.5.1. Configuring File System Watcher
FileSystemWatcher works by polling the class changes with a certain time interval, and then waiting for a predefined quiet period to make sure there are no more changes.
Since Spring Boot relies entirely on the IDE to compile and copy files into the location from where Spring Boot can read them, you might find that there are times when certain changes are not reflected when devtools restarts the application.
If you observe such problems constantly, try increasing the spring.devtools.restart.poll-interval
and spring.devtools.restart.quiet-period
parameters to the values that fit your development environment:
spring.devtools.restart.poll-interval=2s
spring.devtools.restart.quiet-period=1s
spring:
devtools:
restart:
poll-interval: "2s"
quiet-period: "1s"
The monitored classpath directories are now polled every 2 seconds for changes, and a 1 second quiet period is maintained to make sure there are no additional class changes.
8.6. Remote Applications
The Spring Boot developer tools are not limited to local development. You can also use several features when running applications remotely. Remote support is opt-in as enabling it can be a security risk. It should only be enabled when running on a trusted network or when secured with SSL. If neither of these options is available to you, you should not use DevTools' remote support. You should never enable support on a production deployment.
To enable it, you need to make sure that devtools
is included in the repackaged archive, as shown in the following listing:
<build>
<plugins>
<plugin>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-maven-plugin</artifactId>
<configuration>
<excludeDevtools>false</excludeDevtools>
</configuration>
</plugin>
</plugins>
</build>
Then you need to set the spring.devtools.remote.secret
property.
Like any important password or secret, the value should be unique and strong such that it cannot be guessed or brute-forced.
Remote devtools support is provided in two parts: a server-side endpoint that accepts connections and a client application that you run in your IDE.
The server component is automatically enabled when the spring.devtools.remote.secret
property is set.
The client component must be launched manually.
Remote devtools is not supported for Spring WebFlux applications. |
8.6.1. Running the Remote Client Application
The remote client application is designed to be run from within your IDE.
You need to run org.springframework.boot.devtools.RemoteSpringApplication
with the same classpath as the remote project that you connect to.
The application’s single required argument is the remote URL to which it connects.
For example, if you are using Eclipse or Spring Tools and you have a project named my-app
that you have deployed to Cloud Foundry, you would do the following:
-
Select
Run Configurations…
from theRun
menu. -
Create a new
Java Application
“launch configuration”. -
Browse for the
my-app
project. -
Use
org.springframework.boot.devtools.RemoteSpringApplication
as the main class. -
Add
https://myapp.cfapps.io
to theProgram arguments
(or whatever your remote URL is).
A running remote client might resemble the following listing:
. ____ _ __ _ _ /\\ / ___'_ __ _ _(_)_ __ __ _ ___ _ \ \ \ \ ( ( )\___ | '_ | '_| | '_ \/ _` | | _ \___ _ __ ___| |_ ___ \ \ \ \ \\/ ___)| |_)| | | | | || (_| []::::::[] / -_) ' \/ _ \ _/ -_) ) ) ) ) ' |____| .__|_| |_|_| |_\__, | |_|_\___|_|_|_\___/\__\___|/ / / / =========|_|==============|___/===================================/_/_/_/ :: Spring Boot Remote :: (v3.2.5) 2024-04-18T17:04:07.711Z INFO 13353 --- [ main] o.s.b.devtools.RemoteSpringApplication : Starting RemoteSpringApplication v3.2.5 using Java 17.0.11 with PID 13353 (/Users/myuser/.m2/repository/org/springframework/boot/spring-boot-devtools/3.2.5/spring-boot-devtools-3.2.5.jar started by myuser in /opt/apps/) 2024-04-18T17:04:07.721Z INFO 13353 --- [ main] o.s.b.devtools.RemoteSpringApplication : No active profile set, falling back to 1 default profile: "default" 2024-04-18T17:04:08.597Z INFO 13353 --- [ main] o.s.b.d.a.OptionalLiveReloadServer : LiveReload server is running on port 35729 2024-04-18T17:04:08.648Z INFO 13353 --- [ main] o.s.b.devtools.RemoteSpringApplication : Started RemoteSpringApplication in 2.936 seconds (process running for 4.812)
Because the remote client is using the same classpath as the real application it can directly read application properties.
This is how the spring.devtools.remote.secret property is read and passed to the server for authentication.
|
It is always advisable to use https:// as the connection protocol, so that traffic is encrypted and passwords cannot be intercepted.
|
If you need to use a proxy to access the remote application, configure the spring.devtools.remote.proxy.host and spring.devtools.remote.proxy.port properties.
|
8.6.2. Remote Update
The remote client monitors your application classpath for changes in the same way as the local restart. Any updated resource is pushed to the remote application and (if required) triggers a restart. This can be helpful if you iterate on a feature that uses a cloud service that you do not have locally. Generally, remote updates and restarts are much quicker than a full rebuild and deploy cycle.
On a slower development environment, it may happen that the quiet period is not enough, and the changes in the classes may be split into batches. The server is restarted after the first batch of class changes is uploaded. The next batch can’t be sent to the application, since the server is restarting.
This is typically manifested by a warning in the RemoteSpringApplication
logs about failing to upload some of the classes, and a consequent retry.
But it may also lead to application code inconsistency and failure to restart after the first batch of changes is uploaded.
If you observe such problems constantly, try increasing the spring.devtools.restart.poll-interval
and spring.devtools.restart.quiet-period
parameters to the values that fit your development environment.
See the Configuring File System Watcher section for configuring these properties.
Files are only monitored when the remote client is running. If you change a file before starting the remote client, it is not pushed to the remote server. |
9. Packaging Your Application for Production
Executable jars can be used for production deployment. As they are self-contained, they are also ideally suited for cloud-based deployment.
For additional “production ready” features, such as health, auditing, and metric REST or JMX end-points, consider adding spring-boot-actuator
.
See actuator.html for details.
10. What to Read Next
You should now understand how you can use Spring Boot and some best practices that you should follow. You can now go on to learn about specific Spring Boot features in depth, or you could skip ahead and read about the “production ready” aspects of Spring Boot.