[
https://issues.apache.org/jira/browse/BEAM-4130?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16583602#comment-16583602
]
Maximilian Michels edited comment on BEAM-4130 at 8/21/18 1:02 PM:
-------------------------------------------------------------------
For local runs, where Flink is bootstrapped inside the Docker container, we
need to default to run the SDK harness in the same container
{{InProcessEnvironmentFactory}}. This is necessary because we can't start
Docker containers inside the Docker container.
For remote Flink clusters, we can use the regular {{DockerEnvironmentFactory}}.
edit: The InProcess SDK harness is a separate project:
https://issues.apache.org/jira/browse/BEAM-5187
As discussed on the mailing list, we can create sibling containers for embedded
runs by mounting the Docker binaries and the Docker socket inside the Job
Server container.
was (Author: mxm):
For local runs, where Flink is bootstrapped inside the Docker container, we
need to default to run the SDK harness in the same container
{{InProcessEnvironmentFactory}}. This is necessary because we can't start
Docker containers inside the Docker container.
For remote Flink clusters, we can use the regular {{DockerEnvironmentFactory}}.
> Portable Flink runner JobService entry point in a Docker container
> ------------------------------------------------------------------
>
> Key: BEAM-4130
> URL: https://issues.apache.org/jira/browse/BEAM-4130
> Project: Beam
> Issue Type: New Feature
> Components: runner-flink
> Reporter: Ben Sidhom
> Assignee: Maximilian Michels
> Priority: Minor
> Time Spent: 4h 50m
> Remaining Estimate: 0h
>
> The portable Flink runner exists as a Job Service that runs somewhere. We
> need a main entry point that itself spins up the job service (and artifact
> staging service). The main program itself should be packaged into an uberjar
> such that it can be run locally or submitted to a Flink deployment via `flink
> run`.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)