CacheFactory | useful synchronous key-value store | Storage library

You cannot. The error message is right.

You promised to return a derived type and yet you return a base type, knowing that it will be right, because that is what you instantiated. That is not how a compiler works. result needs to be of type CACHE_TYPE.

You basically need to configure the list of locators used by the members so they can "see" each other, locators are mandatory even when using a peer to peer topology. Please have a look at Peer-to-Peer Configuration for further details.

Cheers

This "refused connection: exceeded max-connections" error happens because the clients are trying to create more connections to the servers than the max-connections that you set.

The clients use a connection pool, so the number of concurrent connections from a single client is proportional to the number of threads that are currently doing operations from the client.

The next step is to figure out how many threads your load generation is actually creating. Theoretically, each thread should only be using a single connection at a time, so worst case for max-connections should be around num_threads * num_clients if the load is very uneven and all of the clients are targeting the same server at the same time.

If you are unable to control the number of threads on the client, the pool itself has tuning parameters such as PoolFactory.setMaxConnections which will limit the number of connections from a single client.

You are initializing memberCache in putInCache() method instead of initializing in constructor. With your design, you need to put something in the cache before you can retrieve something. So either initialize it in the constructor or put something in it.

Also note that by default Guice returns a new instance each time it supplies a value. So you may be using one instance while putting stuff into the cache, hence initializing memberCache, but using another instance while retrieving data, in which memberCache is null. You can use @Singleton annotation to inject the same instance of DtoPlan everywhere.

And looks like you are creating a new instance of cache every time you need to put something in it. So you lose the previously stored values. This whole code looks flawed.

The dispatcher will process a batch of messages from the queue when either the batch size or the time interval is reached. In your particular case, you have configured the batch interval as 2 milliseconds, that's probably the reason why the events get processed almost inmediatly.

Hope this helps. Cheers.

In order for the region configuration to be persisted across restarts you need to use the Cluster Configuration Service, please have a look at Configuring and Running a Cluster for further details about this. Unfortunately in GemFire 8.2.X there was no public API to manually use this configuration service (the only way to modify cluster configuration is through internal methods or manually editing the configuration xml file while the cluster is offline) so your only option would be to use GemFire SHell. Newer versions of GemFire also have this limitation, but there's an enhancement currently being worked to make this configurable through he API directly (please have a look at Cluster Management Service for further details).

As a side note, and considering that you're using Spring Boot already, Spring Boot for Apache Geode & Pivotal GemFire has made huge improvements around usability and configurability, it can certainly give you more flexibility and help you achieve your goal here (specifically the @EnableClusterConfiguration annotation).

Hope this helps. Cheers.

Actually, it sounds like exactly the issue which is already tracked in the StackExchange.Redis repository. So, I suggest that you could update your nuget version and restart it.

For more details, you could refer to this similar issue.

It shouldn't. You're only actually using the passed Context if the Spanned doesn't already exist. The real performance issue would be in creating the Typeface itself, which is done only once.

Since you're caching the Spanned and not the Context parameter, you shouldn't have a problem.

Edit:

If you really want to be safe about it, pass context.getApplicationContext() when calling the method. That Context never "expires" as long as your app is running, and will be GCed as soon as it stops running.

There is a error while parsing JSON, your JSON is not correct if you wants to you can check your JSON format at JSON Viewer else you can check your json format by installing json viewer chrome extension