Permalink
165 lines (97 sloc) 4.41 KB

StreamSets Data Collector Edge (SDC Edge)

Minimum Requirements

  • Java >= 8u101 (for supporting Let's Encrypt SSL Certificates as used by gopkg.in)

Clone Repository

Building a distribution archive for a specific platform

./gradlew <platform>DistTar | <platform>DistZip

Where platform is one of:

  • darwinAmd64
  • linuxAmd64
  • linuxArm
  • windowsAmd64

The zip variant is used for Windows targets, and tar for all other targets.

e.g. ./gradlew darwinAmd64DistTar

Building distribution archives for all platforms

./gradlew clean dist

Install an unarchived distribution into the dist folder

./gradlew install<platform>

Where platform is one of:

  • DarwinAmd64
  • LinuxAmd64
  • LinuxArm
  • WindowsAmd64

e.g. ./gradlew installDarwinAmd64

Publishing Binaries to Maven Repo for all platforms

./gradlew publish

Building Docker image

docker build -t streamsets/datacollector-edge .

Building Docker image for alternate platform

docker build --build-arg base=<target image> --build-arg platform=<platform>

Where platform is one of:

  • DarwinAmd64
  • LinuxAmd64
  • LinuxArm
  • WindowsAmd64

e.g. docker build --build-arg base=arm32v6/alpine --build-arg platform=LinuxArm

Run tests

./gradlew test

Run coverage

./gradlew coverage

Run checks (test, fmt and vet)

./gradlew check

Running

cd dist
bin/edge

To start pipeline on SDC Edge start

bin/edge -start=<pipelineId>

To pass runtime parameters

bin/edge -start=tailFileToHttp -runtimeParameters='{"filePath":"/tmp/sds.log","httpUrl":"http://localhost:9999","sdcAppId":"sde"}'

To enable DEBUG Log Level

bin/edge -debug -start=tailFileToHttp

To include TensorFlow Processor in the Edge binary

  • Install TensorFlow for Go - https://www.tensorflow.org/install/install_go

  • Build Edge using: ./gradlew install -PincludeStage="kafka javascript tensorflow"

    Where platform is one of:

    • DarwinAmd64
    • LinuxAmd64
    • LinuxArm
    • WindowsAmd64

    e.g. ./gradlew installDarwinAmd64 -PincludeStage="kafka javascript tensorflow"

or use TensorFlow Docker build

> resources/docker/tensorflow/build.sh
> docker run --publish 18633:18633 --name edge --rm streamsets/datacollector-edge:tf

REST API

curl -X GET http://localhost:18633/rest/v1/pipeline/:pipelineId/status
curl -X POST http://localhost:18633/rest/v1/pipeline/:pipelineId/start
curl -X POST http://localhost:18633/rest/v1/pipeline/:pipelineId/stop
curl -X POST http://localhost:18633/rest/v1/pipeline/:pipelineId/resetOffset
curl -X GET http://localhost:18633/rest/v1/pipeline/:pipelineId/metrics

To pass runtime parameters during start

curl -X POST http://localhost:18633/rest/v1/pipeline/tailFileToHttp/start -H 'Content-Type: application/json;charset=UTF-8' --data-binary '{"filePath":"/tmp/sds.log","httpUrl":"http://localhost:9999","sdcAppId":"sde"}'

Docker run the image

To run a container from the resulting image:

docker run --publish 18633:18633 --name edge --rm streamsets/datacollector-edge:latest

Getting inside the container

docker exec -it datacollector-edge /bin/sh

Release

./gradlew publish -Prelease

CPU & Heap Profile

curl http://localhost:18633/debug/pprof/profile > cpu.pb.gz
curl http://localhost:18633/debug/pprof/heap > heap.pb.gz
curl http://localhost:18633/debug/pprof/goroutine > goroutine.pb.gz
curl http://localhost:18633/debug/pprof/block > block.pb.gz

Running StreamSets Data Collector Edge as service

####( Currently supports Windows XP+, Linux/(systemd | Upstart | SysV), and OSX/Launchd.)

Service Name - datacollector-edge
Command Description
bin/edge -service install Install Data Collector Edge as a service
bin/edge -service uninstall Uninstall the Data Collector Edge service
bin/edge -service start Start the Data Collector Edge service
bin/edge -service stop Stop the Data Collector Edge service
bin/edge -service restart Restart the Data Collector Edge service