Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 5.3

Install S4

There are 3 2 ways:

  • Download the 0.5.0 release (warning) We recommend getting the "source releaseor download the binary release" release and building it, because some dependencies that may not be available on your machine, but are required for the "binary" release.
  • or checkout from the Apache git repository, by following the instructions. The 0.5.0 tag corresponds to the current release.

...

  1. Compile and install S4 in the local maven repository: (you can also let the tests run , which is currently quite long: we're not yet using mockswithout the -DskipTests option)
    Code Block
    S4:incubator-s4$ ./gradlew install -DskipTests
    .... verbose logs ...
    
  2. Build the startup scripts: 
    Code Block
    S4:incubator-s4$ ./gradlew s4-tools:installApp
    .... verbose logs 
    ...:s4-tools:installApp
    

...

  • HelloApp.java: defines a simple application: exposes an input stream ("names"), connected to the HelloPE. See the event dispatch configuration page for more information about how events are dispatched.
    Code Block
    // App parent class provides integration with the S4 platform
    public class HelloApp extends App {
    
        @Override
        protected void onStart() {
        }
    
        @Override
        protected void onInit() {
            // That's where we define PEs and streams
            // create a prototype
            HelloPE helloPE = createPE(HelloPE.class);
            // Create a stream that listens to the "lines" stream and passes events to the helloPE instance.
            createInputStream("names", new KeyFinder<Event>() {
                    // the KeyFinder is used to identify keys
                @Override
                public List<String> get(Event event) {
                    return Arrays.asList(new String[] { event.get("name") });
                }
            }, helloPE);
        }
    // skipped remaining methods
    

...

  1. Set-up the cluster:
    1. In 2 steps:
      1. Start a Zookeeper server instance:
        Code Block
         S4:incubator-s4$ ./s4 zkServer
        S4:myApp$ calling referenced s4 script : /Users/S4/tmp/incubator-s4/s4
        [main] INFO  org.apache.s4.tools.ZKServer - Starting zookeeper server on port [2181]
        [main] INFO  org.apache.s4.tools.ZKServer - cleaning existing data in [/var/folders/8V/8VdgKWU3HCiy2yV4dzFpDk+++TI/-Tmp-/tmp/zookeeper/data] and [/var/folders/8V/8VdgKWU3HCiy2yV4dzFpDk+++TI/-Tmp-/tmp/zookeeper/log]
      2. Define a new cluster. Say a cluster named "cluster1" with 2 partitions, nodes listening to ports starting from 12000:
        Code Block
        S4:myApp$ ./s4 newCluster -c=cluster1 -nbTasks=2 -flp=12000
        calling referenced s4 script : /Users/S4/tmp/incubator-s4/s4
        [main] INFO  org.apache.s4.tools.DefineCluster - preparing new cluster [cluster1] with [2] node(s)
        [main] INFO  org.apache.s4.tools.DefineCluster - New cluster configuration uploaded into zookeeper
        
    2. Alternatively you may combine these two steps into a single one, by passing the cluster configuration inline with the zkServer command:
      Code Block
       S4:incubator-s4$ ./s4 zkServer -clusters=c=cluster1:flp=12000:nbTasks=2
  2. Start 2 S4 nodes with the default configuration, and attach them to cluster "cluster1" :
    Code Block
    S4:myApp$ ./s4 node -c=cluster1
    calling referenced s4 script : /Users/S4/tmp/incubator-s4/s4
    15:50:18.996 [main] INFO  org.apache.s4.core.Main - Initializing S4 node with :
    - comm module class [org.apache.s4.comm.DefaultCommModule]
    - comm configuration file [default.s4.comm.properties from classpath]
    - core module class [org.apache.s4.core.DefaultCoreModule]
    - core configuration file[default.s4.core.properties from classpath]
    -extra modules: []
    [main] INFO  org.apache.s4.core.Main - Starting S4 node. This node will automatically download applications published for the cluster it belongs to
    
    and again (maybe in another shell):
    Code Block
     S4:myApp$ ./s4 node -c=cluster1
    
  3. Build, package and publish the app to cluster1:
    1. You may do that in a single step (currently, you must use the name of the current project, and you need to specify the gradle build file with a complete path).
      Note that specifying the app class is optional but avoids issues when the scripts tries to guess automatically the app class:
      Code Block
      S4:myApp$ ./s4 deploy -appName=myApp -c=cluster1 -b=`pwd`/build.gradle -a=hello.HelloApp
      .... verbose logs for compiling, building the package, and publishing it to Zookeeper...
      15:46:16.486 [main] INFO  org.apache.s4.tools.Deploy - uploaded application [myApp] to cluster [cluster1], using zookeeper znode [/s4/clusters/cluster1/apps/myApp]
      
    2. You may also do that in 2 separate steps:
      1. Create an s4r archive. The following creates an archive named myApp.s4r (here you may specify an arbitrary name) in build/libs.
        Again specifying the app class is optional :
        Code Block
        ./s4 s4r -a=hello.HelloApp -b=`pwd`/build.gradle myApp
      2. Publish the s4r archive (you may first copy it to a more adequate place). The name of the app is arbitrary:
        Code Block
        ./s4 deploy -s4r=`pwd`/build/libs/myApp.s4r -c=cluster1 -appName=myApp
        (grey lightbulb) You can follow this method for a distributed deployment (by copying the s4r to a shared location on a distributed file system)
  4. S4 nodes will detect the new application, download it, load it and start it. You will get something like:
    Code Block
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s.d.DistributedDeploymentManager - Detected new application(s) to deploy {}[myApp]
    [ZkClient-EventThread-15-localhost:2181] INFO  org.apache.s4.core.Server - Local app deployment: using s4r file name [myApp] as application name
    [ZkClient-EventThread-15-localhost:2181] INFO  org.apache.s4.core.Server - App class name is: hello.HelloApp
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s4.comm.topology.ClusterFromZK - Changing cluster topology to { nbNodes=0,name=unknown,mode=unicast,type=,nodes=[]} from null
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s4.comm.topology.ClusterFromZK - Adding topology change listener:org.apache.s4.comm.tcp.TCPEmitter@79b2591c
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s.comm.topology.AssignmentFromZK - New session:87684175268872203; state is : SyncConnected
    [ZkClient-EventThread-19-localhost:2181] INFO  o.a.s4.comm.topology.ClusterFromZK - Changing cluster topology to { nbNodes=1,name=cluster1,mode=unicast,type=,nodes=[{partition=0,port=12000,machineName=myMachine.myNetwork,taskId=Task-0}]} from { nbNodes=0,name=unknown,mode=unicast,type=,nodes=[]}
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s.comm.topology.AssignmentFromZK - Successfully acquired task:Task-1 by myMachine.myNetwork
    [ZkClient-EventThread-19-localhost:2181] INFO  o.a.s4.comm.topology.ClusterFromZK - Changing cluster topology to { nbNodes=2,name=cluster1,mode=unicast,type=,nodes=[{partition=0,port=12000,machineName=myMachine.myNetwork,taskId=Task-0}, {partition=1,port=12001,machineName=myMachine.myNetwork,taskId=Task-1}]} from { nbNodes=1,name=cluster1,mode=unicast,type=,nodes=[{partition=0,port=12000,machineName=myMachine.myNetwork,taskId=Task-0}]}
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s4.comm.topology.ClustersFromZK - New session:87684175268872205
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s4.comm.topology.ClustersFromZK - Detected new stream [names]
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s4.comm.topology.ClustersFromZK - New session:87684175268872206
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s4.comm.topology.ClusterFromZK - Changing cluster topology to { nbNodes=2,name=cluster1,mode=unicast,type=,nodes=[{partition=0,port=12000,machineName=myMachine.myNetwork,taskId=Task-0}, {partition=1,port=12001,machineName=myMachine.myNetwork,taskId=Task-1}]} from null
    [ZkClient-EventThread-15-localhost:2181] INFO  org.apache.s4.core.Server - Loaded application from file /tmp/deploy-test/cluster1/myApp.s4r
    [ZkClient-EventThread-15-localhost:2181] INFO  o.a.s.d.DistributedDeploymentManager - Successfully installed application myApp
    [ZkClient-EventThread-15-localhost:2181] DEBUG o.a.s.c.g.OverloadDispatcherGenerator - Dumping generated overload dispatcher class for PE of class [class hello.HelloPE]
    [ZkClient-EventThread-15-localhost:2181] DEBUG o.a.s4.comm.topology.ClustersFromZK - Adding input stream [names] for app [-1] in cluster [cluster1]
    [ZkClient-EventThread-15-localhost:2181] INFO  org.apache.s4.core.App - Init prototype [hello.HelloPE].
    

...

You may also customize the communication and the core layers of S4 by tweaking configuration files and modules.

Last, the javadoc will help you when writing applications.

We hope this will help you start rapidly, and remember: In conclusion, edges are still a bit rough, more aspects need to be documented, and this is not a final version, but that should let you start, and we're happy to help!