beam-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stephen Sisk (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (BEAM-2031) Hadoop FileSystem needs to receive Hadoop Configuration through PipelineOptions
Date Tue, 25 Apr 2017 23:28:04 GMT

    [ https://issues.apache.org/jira/browse/BEAM-2031?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15983814#comment-15983814
] 

Stephen Sisk edited comment on BEAM-2031 at 4/25/17 11:27 PM:
--------------------------------------------------------------

where/how would we do that? There's no way right now for a user to specify any config for
a FileSystem except via PipelineOptions - they're instantiated behind the scenes via the registrars.

In my previous discussion on BEAM-2005 (https://issues.apache.org/jira/browse/BEAM-2005?focusedCommentId=15977497&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15977497)
I brought up that it'd be good to allow multiple configs and that it'd be nice to specify
configuration information on the transform that gets passed to the FileSystem when doing a
read/write, but it's not currently supported as far as I know.  (cc [~dhalperi@google.com]
since this impacts the future of FBS/the sources reading. )

I assume we won't be attempting to do that for the first stable release, and thus PipelineOptions
is the only way.


was (Author: sisk):
where/how would we do that? There's no way right now for a user to specify any config for
a FileSystem except via PipelineOptions - they're instantiated behind the scenes via the registrars.

In my previous discussion on BEAM-2005 I brought up that it'd be good to allow multiple configs
and that it'd be nice to specify configuration information on the transform that gets passed
to the FileSystem when doing a read/write, but it's not currently supported as far as I know.
 (cc [~dhalperi@google.com] since this impacts the future of FBS/the sources reading. )

I assume we won't be attempting to do that for the first stable release, and thus PipelineOptions
is the only way.

> Hadoop FileSystem needs to receive Hadoop Configuration through PipelineOptions
> -------------------------------------------------------------------------------
>
>                 Key: BEAM-2031
>                 URL: https://issues.apache.org/jira/browse/BEAM-2031
>             Project: Beam
>          Issue Type: Sub-task
>          Components: sdk-java-extensions
>            Reporter: Stephen Sisk
>            Assignee: Stephen Sisk
>             Fix For: First stable release
>
>
> Since Beam FileSystem objects are configured via PipelineOptions, we need to pass a Hadoop
Configuration through PipelineOptions. I think that's very solvable, but it does seem semi-complicated.
> cc [~peihe0@gmail.com] I believe you mentioned in the past that you had an answer to
this - is that written down anywhere?



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message