This Source-to-Image Builder let’s you create projects using Vert.x 3 and built with:
-
maven
-
gradle
This builder expect the build process to generate a (unique) "fat" jar (with a file ending with .jar).
|
Note
|
If a project has a pom.xml and a build.gradle, maven will take precedence |
-
BUILDER_ARGS: Allows you to specify options to pass to maven or gradle
-
OUTPUT_DIR: The directory where the "fat-jar" is located. It can be adjusted when building a multi-module projects.
If you do not specify any BUILDER_ARGS, by default the s2i image will use the following:
-
Maven
BUILDER_ARGS="package -Popenshift -DskipTests" OUTPUT_DIR=target
-
Gradle
BUILDER_ARGS="build -x test" OUTPUT_DIR=build/libs
-
First load all the needed resources in a project.
oc create -f https://raw.githubusercontent.com/vert-x3/vertx-openshift-s2i/master/vertx-s2i-all.json
-
Once the builder vertx-s2i has finished building, you can create an app with:
-
Instant app already provided as template
-
Using the vertx-s2i builder image using a regular Git repository
-
-
APP_OPTIONS: let you pass options to the application, such as
-cluster -cp ./cluster.xml…
APP_OPTIONS are set using a .s2i/environment file in your source code:
APP_OPTIONS=-cluster -cp .:./cluster.xml
This S2I builds the fat jar of your application, but also let you copy additional files. You can use this feature to copy configuration files (cluster.xml, json files). These files are copied in the deploy directory (/opt/openshift) and are copied from:
-
./src/env -
OUTPUT_DIR/env
The first location is used for mono-module project and files that does not require processing (at built time). The second location is used for multi-module projects, or for files having being processed during the packaging (filtering, computation…).
Be aware that in the second case you would need to configure your build process to copy the files to the right location.
To run clustered vert.x applications (with Hazelcast), you would need:
-
some dependencies in your build descriptor to use the Hazelcast Kubernetes discovery
-
a specific service for the event bus
-
a specific cluster configuration
-
set
APP_OPTIONSto use this specific cluster configuration
Hazelcast 3.6+ propose a SPI to extend / customize the cluster member discovery. They provide a discovery extension for Kubernetes. To use it, you need to add the related artifact in your build descriptor. For Maven, do it as follows:
<dependency>
<groupId>com.hazelcast</groupId>
<artifactId>hazelcast</artifactId>
<version>3.6.1</version>
</dependency>
<dependency>
<groupId>io.vertx</groupId>
<artifactId>vertx-hazelcast</artifactId>
</dependency>
<dependency>
<groupId>com.noctarius.discovery</groupId>
<artifactId>hazelcast-kubernetes-discovery</artifactId>
<version>0.9.2</version>
</dependency>You should update the version numbers to latest.
When the application is packaged (as a fat jar), it embeds the kubernetes discovery.
To work, the hazelcast-kubernetes-discovery needs a known service. Let’s call this service vertx-eventbus. In a Openshift template (used to deploy a vert.x application), add:
{
"kind" : "Service",
"apiVersion" : "v1",
"metadata" : {
"name": "vertx-eventbus",
"labels" : {
"component" : "vertx-eventbus"
}
},
"spec" : {
"clusterIP" : "None",
"ports" : [ {
"port" : 5701,
"protocol" : "TCP",
"targetPort" : 5701
} ],
"selector": {
"deploymentconfig": "YOUR_APPLICATION_NAME"
}
}
}This service is headless. Your container should also declare the port 5701. For example:
"ports": [
{
"containerPort": 8080,
"protocol": "TCP"
},
{
"containerPort" : 5701,
"protocol" : "TCP"
}
],IMPORTANT: do not use the port 5701 for liveness.
Only one of your application would publish this service in a given namespace / project.
Your vert.x application needs to use specific cluster.xml file. Create in the src/env directory, a cluster.xml file with the following content:
<?xml version="1.0" encoding="UTF-8"?>
<hazelcast xsi:schemaLocation="http://www.hazelcast.com/schema/config hazelcast-config-3.6.xsd"
xmlns="http://www.hazelcast.com/schema/config"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<properties>
<property name="hazelcast.mancenter.enabled">false</property>
<property name="hazelcast.memcache.enabled">false</property>
<property name="hazelcast.rest.enabled">false</property>
<property name="hazelcast.wait.seconds.before.join">0</property>
<property name="hazelcast.logging.type">jdk</property>
<!-- at the moment the discovery needs to be activated explicitly -->
<property name="hazelcast.discovery.enabled">true</property>
</properties>
<network>
<port auto-increment="true" port-count="10000">5701</port>
<outbound-ports>
<ports>0</ports>
</outbound-ports>
<join>
<multicast enabled="false"/>
<tcp-ip enabled="false"/>
<discovery-strategies>
<discovery-strategy enabled="true"
class="com.noctarius.hazelcast.kubernetes.HazelcastKubernetesDiscoveryStrategy">
<properties>
<!-- configure discovery headless service lookup -->
<property name="service-dns">vertx-eventbus.vertx-demo-cluster.svc.cluster.local</property>
</properties>
</discovery-strategy>
</discovery-strategies>
</join>
<interfaces enabled="false">
<interface>10.10.1.*</interface>
</interfaces>
<ssl enabled="false"/>
<socket-interceptor enabled="false"/>
</network>
<partition-group enabled="false"/>
<executor-service name="default">
<pool-size>16</pool-size>
<queue-capacity>0</queue-capacity>
</executor-service>
<map name="__vertx.subs">
<backup-count>1</backup-count>
<time-to-live-seconds>0</time-to-live-seconds>
<max-idle-seconds>0</max-idle-seconds>
<eviction-policy>NONE</eviction-policy>
<max-size policy="PER_NODE">0</max-size>
<eviction-percentage>25</eviction-percentage>
<merge-policy>com.hazelcast.map.merge.LatestUpdateMapMergePolicy</merge-policy>
</map>
<semaphore name="__vertx.*">
<initial-permits>1</initial-permits>
</semaphore>
</hazelcast>You need to edit the service-dns property value. The value is computed as follows:
service-name (vertx-eventbus).project name.svc.cluster.local.
This cluster.xml file will be placed in the same directory as your application. The source to image is copying the file there.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent