I am trying to get our config server to host plain text files more dynamically. I am currently having a Spring config server and a few services using it to get configuration. I am also have kafka up for Spring Cloud Bus connected to both config server and the clients.
Additionally, I am using config server to server some plain text files (as described in https://cloud.spring.io/spring-cloud-static/spring-cloud.html#_serving_plain_text). I am wishing to get the client to refresh, if the plain text files are updated.
When normal properties are updated, my client through spring cloud bus can receive the refresh request, and perform the refresh, however, when the plain text files are updated, the same thing doesn't happen. The config server logs showed that it registered the changes, but there's no refresh event broadcast to the clients through kafka.
I am wondering if there's a way to achieve this automatic refresh for plain text files.
My current thought:
Have a refresh event listener to catch the kafka message, and do refresh for necessary bean.
example:
#Configuration
public class FileReloadListener implements ApplicationListener<RefreshRemoteApplicationEvent> {
#Override
public void onApplicationEvent(RefreshRemoteApplicationEvent event) {
// refresh the needed beans
}
}
Any idea? Thanks!
Related
I'm working with websocket to push notification using rabbitmq like a message broker. In controller i using SimpMessageTemplate to convertAndSendToUser a message via the nick name of user
#PostMapping("/{nickName}/notification")
public void send(#PathVariable String nickName,
#RequestBody NotificationRes notificationRes) {
messagingTemplate.convertAndSendToUser(nickName, "/exchange/amq.direct/notification", notificationRes);
}
I make some configuration by using JWT to authorize user and set user to StompHeaderAccessor when user start connecting.
Spring security messaging currently convert send to user. With that endpoint spring will convert it to /exchange/amq.direct/notification-user{session_id}. And rabbitmq will create an auto deleted queue. So, if I have 100 connect in the same time. It will create 100 queue. I want to custom this flow to create one queue and spring will using message header for spring can know the user to send message to.
I did some research about UserDestinationResolver but I still don't know how to make spring can handle with header instead of url like ...-user{session_id}. #.#. What should i do for this situatioin?
Is there any way to refresh springboot configuration as soon as we change .properties file?
I came across spring-cloud-config and many articles/blogs suggested to use this for a distributed environment. I have many deployments of my springboot application but they are not related or dependent on one another. I also looked at few solutions where they suggested providing rest endpoints to refresh configs manually without restarting application. But I want to refresh configuration dynamically whenever I change .properties file without manual intervention.
Any guide/suggestion is much appreciated.
Can you just use the Spring Cloud Config "Server" and have it signal to your Spring Cloud client that the properties file changed. See this example:
https://spring.io/guides/gs/centralized-configuration/
Under the covers, it is doing a poll of the underlying resource and then broadcasts it to your client:
#Scheduled(fixedRateString = "${spring.cloud.config.server.monitor.fixedDelay:5000}")
public void poll() {
for (File file : filesFromEvents()) {
this.endpoint.notifyByPath(new HttpHeaders(), Collections
.<String, Object>singletonMap("path", file.getAbsolutePath()));
}
}
If you don't want to use the config server, in your own code, you could use a similar scheduled annotation and monitor your properties file:
#Component
public class MyRefresher {
#Autowired
private ContextRefresher contextRefresher;
#Scheduled(fixedDelay=5000)
public void myRefresher() {
// Code here could potentially look at the properties file
// to see if it changed, and conditionally call the next line...
contextRefresher.refresh();
}
}
I have a Spring Boot app (Jhipster) that uses STOMP over WebSockets to communicate information from the server to users.
I recently added an ActiveMQ server to handle scaling the app horizontally, with an Amazon auto-scaling group / load-balancer.
I make use the convertAndSendToUser() method, which works on single instances of the app to locate the authenticated users' "individual queue" so only they receive the message.
However, when I launch the app behind the load balancer, I am finding that messages are only being sent to the user if the event is generated on the server that their websocket-proxy connection (to the broker) is established on?
How do I ensure the message goes through ActiveMQ to whichever instance of the app that the user is actually "connected too" regardless of which instance receives, say an HTTP Request that executes the convertAndSendToUser() event?
For reference here is my StompBrokerRelayMessageHandler:
#Bean
public AbstractBrokerMessageHandler stompBrokerRelayMessageHandler() {
StompBrokerRelayMessageHandler handler = (StompBrokerRelayMessageHandler) super.stompBrokerRelayMessageHandler();
handler.setTcpClient(new Reactor2TcpClient<>(
new StompTcpFactory(orgProperties.getAws().getAmazonMq().getStompRelayHost(),
orgProperties.getAws().getAmazonMq().getStompRelayPort(), orgProperties.getAws().getAmazonMq
().getSsl())
));
return handler;
}
#Override
public void configureMessageBroker(MessageBrokerRegistry config) {
config.enableStompBrokerRelay("/queue", "/topic")
.setSystemLogin(orgProperties.getAws().getAmazonMq().getStompRelayHostUser())
.setSystemPasscode(orgProperties.getAws().getAmazonMq().getStompRelayHostPass())
.setClientLogin(orgProperties.getAws().getAmazonMq().getStompRelayHostUser())
.setClientPasscode(orgProperties.getAws().getAmazonMq().getStompRelayHostPass());
config.setApplicationDestinationPrefixes("/app");
}
I have found the name corresponding to the queue that is generated on ActiveMQ by examining the headers in the SessionSubscribeEvent, that is generated in the listener when a user subscribes to a user-queue, as simpSessionId.
#Override
#EventListener({SessionSubscribeEvent.class})
public void onSessionSubscribeEvent(SessionSubscribeEvent event) {
log.debug("Session Subscribe Event:" +
"{}", event.getMessage().getHeaders().toString());
}
Corresponding queues' can be found in ActiveMQ, in the format: {simpDestination}-user{simpSessionId}
Could I save the sessionId in a key-value pair and just push messages onto that topic channel?
I also found some possibilities of setting ActiveMQ specific STOMP properties in the CONNECT/SUBSCRIBE frame to create durable subscribers if I set these properties will Spring than understand the routing?
client-id & subcriptionName
Modifying the MessageBrokerReigstry config resolved the issue:
config.enableStompBrokerRelay("/queue", "/topic")
.setUserDestinationBroadcast("/topic/registry.broadcast")
Based on this paragraph in the documentation section 4.4.13:
In a multi-application server scenario a user destination may remain
unresolved because the user is connected to a different server. In
such cases you can configure a destination to broadcast unresolved
messages to so that other servers have a chance to try. This can be
done through the userDestinationBroadcast property of the
MessageBrokerRegistry in Java config and the
user-destination-broadcast attribute of the message-broker element in
XML
I did not see any documentation on "why" /topic/registry.broadcast was the correct "topic" destination, but I am finding various iterations of it:
websocket sessions sample doesn't cluster.. spring-session-1.2.2
What is MultiServerUserRegistry in spring websocket?
Spring websocket - sendToUser from a cluster does not work from backup server
We have a spring application where redis cache has been implemented along with the database MySQL. Here we are using redis cache to store the temporary values for the server validations instead of hitting the database every time, hence hitting the database calls every time gets reduces system performance.
Now i explain my problem while hitting the spring boot action endpoints,
if suddenly my redis cache server stops, we would like to know how to get the notification that my redis cache server is down. So we need solution / example java application to get the notification using redis cache listener context or anything like that.
Redis doesn't work that way. In fact, no remote service will notify your application that it's down. Usually, it's the other way round: If the service you're consuming is accessed with a more or less sophisticated client, you might take advantage of the client's features.
Asynchronous clients that run I/O, or monitoring threads can help here. More specific, it depends on the client you're using with Spring Boot and Redis. Jedis is a plain client that reacts on a request basis. Lettuce allows you to register a RedisConnectionStateListener that is called on specific connection events, such as connected/disconnected:
RedisClient redisClient = …;
redisClient.addListener(new RedisConnectionStateListener() {
#Override
public void onRedisConnected(RedisChannelHandler<?, ?> redisChannelHandler) {
}
#Override
public void onRedisDisconnected(RedisChannelHandler<?, ?> redisChannelHandler) {
}
#Override
public void onRedisExceptionCaught(RedisChannelHandler<?, ?> redisChannelHandler, Throwable throwable) {
}
});
When using Spring Data Redis, retrieving the RedisClient from LettuceConnectionFactory might be a bit tricky as it is a private field. Hence it requires reflection.
Spring cloud config client helps to change the properties in run time. Below are 2 ways to do that
Update GIT repository and hit /refresh in the client application to get the latest values
Update the client directly by posting the update to /env and then /refresh
Problem here in both the approaches is that there could be multiple instances of client application running in cloud foundry and above rest calls will reach any one of the instances leaving application in inconsistent state
Eg. POST to /env could hit instance 1 and leaves instance 2 with old data.
One solution I could think of is to continuously hit these end points "n" times using for loop just to make sure all instance will be updated but it is a crude solution. Do any body have better solution for this?
Note: We are deploying our application in private PCF environment.
The canonical solution for that problem is the Spring Cloud Bus. If your apps are bound to a RabbitMQ service and they have the bus on the classpath there will be additional endpoints /bus/env and /bus/refresh that broadcast the messages to all instances. See docs for more details.
Spring Cloud Config Server Not Refreshing
see org.springframework.cloud.bootstrap.config.RefreshEndpoint code here：
public synchronized String[] refresh() {
Map<String, Object> before = extract(context.getEnvironment()
.getPropertySources());
addConfigFilesToEnvironment();
Set<String> keys = changes(before,
extract(context.getEnvironment().getPropertySources())).keySet();
scope.refreshAll();
if (keys.isEmpty()) {
return new String[0];
}
context.publishEvent(new EnvironmentChangeEvent(keys));
return keys.toArray(new String[keys.size()]);
}
that means /refresh endpoint pull git first and then refresh catch,and public a environmentChangeEvent,so we can customer the code like this.