It is often a problem especially for bigger companies, where many new projects are introduced into production, and operations teams do not really know what exactly all those new applications are doing. All they have are the log files, and if they are lucky, they can find some information there, provided developers did their job well, and applications log all their activities properly. In most cases they get incident tickets from the customers, stating they experience troubles using application XYZ. Application administrators try to search in logs, find nothing at all or nothing they can explain, send those logs to developers, wait for few days, and finally end up with the statement from developers, they cannot find anything in the logs, so there is nothing for them to fix

For managers, even if they realize that the vendor of the application XYZ is delivering very bad quality, they simply cannot afford to change the vendor, because it is extremely difficult for anybody else to continue developing that application. So they must go on with that vendor, well knowing all they can expect from them is more troubles

So you have to live with the applications that are very difficult to operate, not easy to fix, and almost impossible to improve

Flink-ESB solves that situation in a very elegant way. It gives developers ability to see applications with the eyes of administrators, and administrators can see everything that developers see. Even people that do not work with the application will be able to say in 5 minutes whether the application is performing good, and if not, where is the problem

It is not just a bare statement. Everything in Flink-ESB is done by "Routes". All Flink-ESB applications are collections of "Routes" and "Beans", that the routes micht use. If the application is listening on HTTP URI "/processRequest" and does some processing with every received message - it is done by the route. If the application processes some batch data on regular basis and stores something in the database - it is done by the route

Every route consists of "Route Nodes", that are "Linked" with their neighbours. Every route node is responsible for some simple activity within the route: convert the message, split the message into array of smaller messages, decide where to send the message next (based on content of the message), send HTTP request to remote server, execute SQL statement etc. Links are responsible for passing message to the next route node, sometimes they also convert the message payload to some other format

So how all this information helps solving the problem described here?

Now imagine, you have possibility to see all the routes application consists of in a graphical form. For every route you see the performance statistic: how many times it was called, how many time it took to process the route, how many errors it has thrown, what kind of errors, when these errors were thrown, what were the messages that led to the errors

For every route you see what are the route nodes, how they are linked to each other, what every route node is doing. You can see performance statistic per route node. So you see where exactly within a route an error was thrown, what that particular route node is doing, what was an error, when it was thrown, and what was the message that led to an error

You cannot hide anything in Flink-ESB application. It should take no more then 5 minutes for anybody to check the application on all cluster instances and to say whether it is performing well. You usually do not even need to search within log files

It is easy for the developers to continue working on Flink-ESB application, or improving existing version, even if the application is very complicated and was initially created by another vendor

Flink-ESB Editor is the only tool you need to develop different kinds of applications. You don't need to install any other software, like Web- or Application Servers

Many components are available out of the box, allowing you to design applications with processes of any complexity. In most situations no coding at all (and hence no programming experience) is needed to create an application and design all its routes

You can develop multiple projects at the same time using the same Editor. You can start them, check the functionality, try integration with already installed applications

Flink-ESB Editor allows you to test every part of the project

If you code java classes and define them as beans in the project, you can start the project and call every method of such beans, try different arguments, and see whether they return what you expect

Try calling Routes using different payloads, properties and headers

Flink-ESB Editor comes with preinstalled Flink-ESB Admin Console. As soon as you start one or multiple projects in the Editor, you can use Admin Console to verify behaviour of different parts of the projects

Projects started in the Editor behave exactly as applications installed in Flink-ESB Runtime Environment. Test interaction with other applications. See how application reacts on different internal or external events or outages. Simulate outages or mistakes to verify application's behaviour

Installing Flink-ESB application into Flink-ESB Runtime Environment is very quick and simple task, and can be achieved in different ways

Just drop an application jar into a certain folder, where Flink-ESB Runtime will pick it up and install

or copy it to any folder you wish and use SSH console to deploy it from there

You can also use Flink-ESB Admin Console to administer installation tasks

Flink-ESB offers API to get different monitoring statistics from the running instances. This API is accessible to authorised users over HTTP from any machine. Format of output is either JSON or CSV, which allows to extract portion of data you would like to use in your custom monitoring script

Flink-ESB outputs statistics about all instances running in the cluster, or just instances / applications / routes you would like to have statistics for

You can also use Flink-ESB Admin Console to have all performance, error and runtime statistics in user friendly graphical format

Flink-ESB protocols every call. The longer you use it, the more information it will show you on strong and weak spots of your application

You can see what errors were thrown, what kind of errors and on what places within your business processes exactly

Flink-ESB collects also performance statistics and shows you this information on Admin Console. You can see exactly what processes are to slow, on what place they are slow, and what exactly is done on that place

Just let Flink-ESB collect the information, and check it on Admin Console every now and then

Use Flink-ESB Admin Console or SSH Console to redeploy updated application

No restart of the process instance is needed for this