hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sean Busbey (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-11656) Classpath isolation for downstream clients
Date Tue, 25 Apr 2017 01:09:05 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-11656?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15982221#comment-15982221
] 

Sean Busbey commented on HADOOP-11656:
--------------------------------------

I'm in the middle of familial responsibilities, so my apologies in advance if this comes off
as curt [~ctubbsii].

I believe [~sjlee0]'s question above is basically "I didn't realize we'd be publishing shaded
jars into Maven Central, could we skip doing that?" ([~sjlee0], please correct me if I'm wrong.)

The answer to this, as I stated above, is no because then folks relying on the shaded version
could not test as a part of normal maven lifecycle.

If you look at the code as of the 3.0.0-alpha2 release, you will see that we have the following
modules:

* hadoop-client - a neither shaded nor relocated module that claims to expose just those hadoop
and transitive dependency artifacts needed to interact with a client (this has been in the
code base since roughly hadoop 1.0). Anyone wanting non-shaded access to Hadoop and its dependencies
can continue to use this artifact.
* hadoop-minicluster - a neither shaded nor relocated test module that provides tools needed
to test the above (also since roughly Hadoop 1.0)
* hadoop-client-api - a shaded and relocated module containing Hadoop authored classes needed
to compile a client. (If you look at the code or read the relevant jira, it explains that
this is generated via part of the above hadoop-client module and is new for 3.0.0-alpha2)
* hadoop-client-runtime - a companion shaded and relocated  module containing Hadoop and third
party classes needed to run client built using the above api jar. (also generated via part
of the hadoop-cliet module and also new for 3.0.0-alpha2)
* hadoop-client-minicluster - a shaded and relocated  version of the hadoop-minicluster module
above that can be used to test the shade-relocated client modules above
* invariant tests and integration tests for the above generated modules, which require said
modules in order to be checked.

Given the above background information about the old features and the client-side features
under discussion, why would we publish the non-shaded version of any of the three shaded and
relocated jars?

> Classpath isolation for downstream clients
> ------------------------------------------
>
>                 Key: HADOOP-11656
>                 URL: https://issues.apache.org/jira/browse/HADOOP-11656
>             Project: Hadoop Common
>          Issue Type: New Feature
>            Reporter: Sean Busbey
>            Assignee: Sean Busbey
>            Priority: Blocker
>              Labels: classloading, classpath, dependencies, scripts, shell
>         Attachments: HADOOP-11656_proposal.md
>
>
> Currently, Hadoop exposes downstream clients to a variety of third party libraries. As
our code base grows and matures we increase the set of libraries we rely on. At the same time,
as our user base grows we increase the likelihood that some downstream project will run into
a conflict while attempting to use a different version of some library we depend on. This
has already happened with i.e. Guava several times for HBase, Accumulo, and Spark (and I'm
sure others).
> While YARN-286 and MAPREDUCE-1700 provided an initial effort, they default to off and
they don't do anything to help dependency conflicts on the driver side or for folks talking
to HDFS directly. This should serve as an umbrella for changes needed to do things thoroughly
on the next major version.
> We should ensure that downstream clients
> 1) can depend on a client artifact for each of HDFS, YARN, and MapReduce that doesn't
pull in any third party dependencies
> 2) only see our public API classes (or as close to this as feasible) when executing user
provided code, whether client side in a launcher/driver or on the cluster in a container or
within MR.
> This provides us with a double benefit: users get less grief when they want to run substantially
ahead or behind the versions we need and the project is freer to change our own dependency
versions because they'll no longer be in our compatibility promises.
> Project specific task jiras to follow after I get some justifying use cases written in
the comments.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org


Mime
View raw message