Previously we used Feign to create clients for our own services, which are registered on our Eureka Server using a service name as shown in the previous blog post http://theblasfrompas.blogspot.com.au/2016/11/declarative-rest-client-feign-with_8.html. It's not unusual that you'd want to implement an external rest endpoint, basically an endpoint that's not discoverable by Eureka. In that case, you can use the url property on the @FeignClient annotation,
When deploying PCF, you start by deploying Ops Manager. This is basically a VM that you deploy into your IaaS system of choice and it orchestrates the PCF installation. The installation of PCF is done by you through a web interface that runs on the Ops Manager VM. Into that web interface, you can load various "tiles". Each tile provides a specific set of functionality.
I decided to install PCF 1.8 onto Google Cloud Platform today and I thought the experience was fantastic and very straight forward. The GCP Console is fantastic and very powerful indeed. The steps to install it are as follows
Feign is a declarative web service client. It makes writing web service clients easier. To use Feign create an interface and annotate it. It has pluggable annotation support including Feign annotations and JAX-RS annotations. Feign also supports pluggable encoders and decoders.
In this example I show how to use Spring Cloud / Spring Boot application with Feign. The source code for this is as follows
I was recently asked how you could display all PCF application instances detected buildpack OR docker image being used from the command line. The CF REST API gives you all of this information and more as per the documentation below to list all applications.
In this post I walk through what it would take to access Oracle 12c using a spring boot application deployed to Pivotal Cloud Foundry PCF all from my Macbook Pro. Of course this can be done outside of an isolated laptop like my Macbook Pro but handy while doing DEV/TEST and still being able to use Oracle 12c.
Frequently when developing Spring based applications, I will use the H2 in memory database during your development process. H2 ships with a web based database console, which you can use while your application is under development. It is a convenient way to view the tables created by Hibernate and run queries against the in memory database. In this post I show what is required to set this up as well as what it means to then deploy your Spring Boot applications to Pivotal Cloud Foundry and rely on a database service and hence your application becomes cloud aware.
I recently created a demo which can be used to deploy a basic Spring Boot application on Google Cloud Platform (GCP). There isn't really anything specific in the code to make this work on GCP BUT the maven pom.xml has what is required to make it one simple command to send this app to GCP.
I have known about JHipster for a while and thought I would finally give it a test drive and deploy a simple scaffolded app to Pivotal Cloud Foundry. In this post I will use the p-mysql (Pivotal MySQL) rather then the ClearDB instance.
I have been recently working on a web based interface to Pivotal MySQL Tile which can be deployed to Pivotal Cloud Foundry and bound to a p-mysql service instance for instant access to the MySQL DB. Known as PivotalMySQL*Web this is a browser based SQL tool rendered using Bootstrap UI for MySQL PCF service instances which allows you to run SQL commands and view schema objects from a browser based interface. It includes the following capabilities
Pushed applications to CF or PCF you would of most likely used a manifest.yml file and at some point wanted to use variable substitution. manifest.yml files don't support that and a feature request has been asked for this as follows
With a recent customer we scripted the creation of a manifest.yml file from a Jenkins job which would inject the required ROUTE to the application by creating the manifest.yml through a script as follows as shown below.
The demo below is a full working example of using Spring Boot Application which will build/deploy to Pivotal Cloud Foundry using Concourse. Concourse pipelines can easily be created within your source code as this demo shows and hence created / executed quite easily. Concourse limits itself to three core concepts: tasks, resources, and the jobs that compose them.
PCF Dev is a small footprint distribution of Pivotal Cloud Foundry (PCF) intended to be run locally on a developer machine. It delivers the essential elements of the Pivotal Cloud Foundry experience quickly through a condensed set of components. PCF Dev is ideally suited to developers wanting to explore or evaluate PCF, or those already actively building cloud native applications to be run on PCF.
If you heard of Telstra WIFI API you will know it will allow you to search for WIFI Hotspots within a given radius and can be used after signing in for Telstra.dev account at https://dev.telstra.com/ to obtain the Hotpots within a given Radius and Lat/Long location.
The WIFI API for Telstra is described at the link below.
Pivotal Cloud Foundry (PCF) provides a REST API to provide billing/metering data for application and service usage. Although this usage can we viewed in the applications manager dashboard UI in this post below we will show how to use the REST based API using PCF 1.7.
Below we will show how to use the cf CLI to retrieve information about your app and service instances via the Cloud Controller and Usage service APIs.
In this example I am going to show how to use IntelliJ IDEA 15 to create a service application from the IDE to be consumed by the Service Registry service in Pivotal Cloud Foundry (PCF). For more information on this service view the docs page below.
I created the following demo for PCF using spring boot / PCF. After showing a colleague he decided he would write a Kotlin version of the same application. It's interesting to see how the Kotlin classes differ to those of Java.
Apigee and Pivotal partnered to provide comprehensive API management capabilities that expedite the scalable delivery of apps on the powerful Pivotal Cloud Foundry platform. Apigee Edge is available for rapid deployment as a partner service in the Pivotal Network
The following link talks about this service in detail
Working with a customer last week we determined that Spring Boot applications using Liberty Buildpack failed to allow the "Auto-Scale" service to show up "Throughput" metrics and essentially mean we couldn't scale out or down application instances using Throughput metrics.
With Bluemix Public you can view the status page which details all the runtimes and services and thier current status on all 3 PUBLIC regions. Those customers with Bluemix Dedicated or Local get a status page which includes a column on the status of thier Dedicated or Local instance.
To navigate to it perform the following steps:
1. Log into your Bluemix dedicated or local instance web console
2. Click on the Status link which is accessed through the profile icon on the top right hand corner
OpenWhisk is a new event-driven platform that lets developers quickly and easily build feature-rich apps that automatically trigger responses to events. To read more about it view the link below. In this simple example we will explore it it's use from IBM Bluemix by returning Todays date.
I almost always use the tomcat buildpack within IBM Bluemix for my Java based applications. By default IBM bluemix will use the IBM Liberty buildpack for java apps unless you specify otherwise. The buildpacks on Bluemix can be viewed using "cf buildpacks" and the tomcat buildpack is referred to as "java_buildpack"
So to use the tomcat buildpack in a manifest.yml you would target it as follows
I rarely need to save state within the Servlet Context via an application scope, but recently I did and here is what your controller class would look like to get access to the ServletConext with Spring. I was using Spring Boot 1.3.2.RELEASE.
In short you implement the "org.springframework.web.context.ServletContextAware" interface as shown below. In this example we retrieve an application scope attribute.
There is now a new command line plugin for IBM containers on Bluemix so you can push and run docker images using CF CLI itself. The steps below show you how to set this up and I use a basic spring boot application as a docker image to test this out.
Take a note of the docker local host IP. In this example it was as follows, as I test my docker image on my laptop prior to pushing it to Bluemix.
-> docker is configured to use the default machine with IP 192.168.99.100
1. Install the latest CF command line, I used the following version.
I created this simple step by step guide on how to use the IBM Secure Gateway Service. Very simple demo which shows how easy it is to set this up and open a world of possibilities from on premise resources directly exposed via Bluemix whether it's PUBLIC or Dedicated/LOcal Instances
The IBM Bluemix Dedicated/Local Administration Console also includes support for CLI to enable common management tasks to be performed using the Cloud Foundry Command Line by adding a plugin to enable support for it's commands.
Note: The Bluemix Admin CLI plugin requires version 6.11.2 or later.
Here is the updated demo with Bootstrap added to the UI pages. You can deploy this to Bluemix using the "Deploy to Bluemix" directly from GitHub and it will as you to Sing into IBM devOps prior to deploying it directly into your Bluemix Environment.
I am more often then not using spring boot applications on IBM Bluemix and most of what I need is packaged with the application from JPA or JDBC, drivers, Rest etc. Of course with IBM Bluemix we can specify which build pack we wish to use but by default for java applications LIberty is used.
When a stand-alone application is deployed, a default Liberty configuration is provided for the application. The default configuration enables the following Liberty features:
If the application requires backing services they can automatically be created using the"declared-services" tag as shown below. Declared services are a manifest extension, which creates or looks for the required or optional services that
Not before time I finally added my first "Deploy to Bluemix" button on my GitHub projects for Bluemix applications. The screen shot below shows this for the Spring Session - Spring Boot Portable Cloud Ready HTTP Session demo.
The demo below is a simple Spring Boot / Thymeleaf/ Bootstrap application to test Session replication using Spring Session - Spring Boot within IBM Bluemix. Same demo will run on Pivotal Cloud Foundry as well.
In the post below I will show how I integrated Telstra public SMS Api into my Bluemix catalog to be consumed as a service. This was all done from Public Bluemix using the Cloud Integration Service. Step 1 - Create a T.DEV account
In order to get started you need to create an account on http://dev.telstra.com in order to be granted access to the SMS API. Once access is granted you need to create an application which enables you to add/manage Telstra API keys as shown below.
Just upgraded to to Intellij IDEA 14.1.4 and found that the Spring Initializr web page for quickly creating spring boot applications has been added to the New Project dialog. The web site I normally drive new spring boot applications from as follows, is now part of IntelliJ IDEA which is great. http://start.spring.io/
The latest Liberty biuldpack update now adds the ability to switch to JRE 1.8 runtime. Given I am now switching to JDK 1.8 for all my projects this has come in very handy, and of course Spring Boot favors JDK 1.8 as well. The default is still JRE 1.7 so to switch to 1.8 set an ENV variable as follows in your manifest.yml as shown below.