We have an existing spring web app deployed as a WAR file into Amazon Elastic Beanstalk. Currently we load properties files as http resources to give us a single source of p
I found a solution for using spring-cloud-zookeeper without Spring Boot, based on the idea provided here https://wenku.baidu.com/view/493cf9eba300a6c30d229f49.html
It should be easily updated to match your needs and using a Spring Cloud Config Server (the CloudEnvironement class needs to be updated to load the file from the server instead of Zookeeper)
First, create a CloudEnvironement class that will create a PropertySource (ex from Zookeeper) :
CloudEnvironement.java
public class CloudEnvironment extends StandardServletEnvironment {
@Override
protected void customizePropertySources(MutablePropertySources propertySources) {
super.customizePropertySources(propertySources);
try {
propertySources.addLast(initConfigServicePropertySourceLocator(this));
}
catch (Exception ex) {
logger.warn("failed to initialize cloud config environment", ex);
}
}
private PropertySource> initConfigServicePropertySourceLocator(Environment environment) {
ZookeeperConfigProperties configProp = new ZookeeperConfigProperties();
ZookeeperProperties props = new ZookeeperProperties();
props.setConnectString("myzookeeper:2181");
CuratorFramework fwk = curatorFramework(exponentialBackoffRetry(props), props);
ZookeeperPropertySourceLocator propertySourceLocator = new ZookeeperPropertySourceLocator(fwk, configProp);
PropertySource> source= propertySourceLocator.locate(environment);
return source ;
}
private CuratorFramework curatorFramework(RetryPolicy retryPolicy, ZookeeperProperties properties) {
CuratorFrameworkFactory.Builder builder = CuratorFrameworkFactory.builder();
builder.connectString(properties.getConnectString());
CuratorFramework curator = builder.retryPolicy(retryPolicy).build();
curator.start();
try {
curator.blockUntilConnected(properties.getBlockUntilConnectedWait(), properties.getBlockUntilConnectedUnit());
}
catch (InterruptedException e) {
throw new RuntimeException(e);
}
return curator;
}
private RetryPolicy exponentialBackoffRetry(ZookeeperProperties properties) {
return new ExponentialBackoffRetry(properties.getBaseSleepTimeMs(),
properties.getMaxRetries(),
properties.getMaxSleepMs());
}
}
Then create a custom XmlWebApplicationContext class : it will enable to load the PropertySource from Zookeeper when your webapplication start and replace the bootstrap magic of Spring Boot:
MyConfigurableWebApplicationContext.java
public class MyConfigurableWebApplicationContext extends XmlWebApplicationContext {
@Override
protected ConfigurableEnvironment createEnvironment() {
return new CloudEnvironment();
}
}
Last, in your web.xml file add the following context-param for using your MyConfigurableWebApplicationContext class and bootstraping your CloudEnvironement.
contextClass
com.kiabi.config.MyConfigurableWebApplicationContext
If you use a standard property file configurer, it should still be loaded so you can have properties in both a local file and Zookeeper.
For all this to work you need to have spring-cloud-starter-zookeeper-config and curator-framework jar in your classpath with their dependancy, if you use maven you can add the following to your pom.xml
org.springframework.cloud
spring-cloud-zookeeper-dependencies
1.1.1.RELEASE
pom
import
org.springframework.cloud
spring-cloud-starter-zookeeper-config
org.apache.curator
curator-framework