> On Nov. 4, 2014, 12:12 a.m., Vinod Kone wrote:
> > src/launcher/fetcher.cpp, line 230
> > <https://reviews.apache.org/r/27483/diff/3/?file=747858#file747858line230>
> >
> >     Why the fall through to HDFS here?

My reasoning is that if hadoop client is configured with some 
ftp/ftps/http/https credentials that are not in the urls, the hadoop client 
would be able to fetch those resource that the libcurl client can't. I am open 
to changing this if this line of thinking seems confusing.


> On Nov. 4, 2014, 12:12 a.m., Vinod Kone wrote:
> > src/launcher/fetcher.cpp, lines 245-246
> > <https://reviews.apache.org/r/27483/diff/3/?file=747858#file747858line245>
> >
> >     Seems weird to do this after hdfs fetch. Is HDFS local copy as fast as 
> > "cp" ?
> >     
> >     I would change the semantics as follows:
> >     
> >     ```
> >     if (local) {
> >       return fetchWithLocalCopy();
> >     }
> >     
> >     if (http* or ftp*) {
> >       return fetchWithNet();
> >     }
> >     
> >     return fetchWithHadoopClient();
> >     
> >     ```

I am pretty sure that native cp is faster than hadoop client local copy. I can 
make this change.


- Ankur


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/27483/#review59682
-----------------------------------------------------------


On Nov. 4, 2014, 12:02 a.m., Ankur Chauhan wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/27483/
> -----------------------------------------------------------
> 
> (Updated Nov. 4, 2014, 12:02 a.m.)
> 
> 
> Review request for mesos and Timothy St. Clair.
> 
> 
> Bugs: MESOS-1711
>     https://issues.apache.org/jira/browse/MESOS-1711
> 
> 
> Repository: mesos-git
> 
> 
> Description
> -------
> 
> Previously, the fetcher used a hardcoded list of schemes to determine what 
> URIs could be fetched by hadoop (if available). This is now changed such that 
> we first check if hadoop can fetch them for us and then we fallback to the 
> os::net and then a local copy method (same as it used to be). This allows 
> users to fetch artifacts from arbitrary filesystems as long as hadoop is 
> correctly configured (in core-site.xml).
> 
> 
> Diffs
> -----
> 
>   src/hdfs/hdfs.hpp bbfeddef106c598d8379ced085ef0605c4b2f380 
>   src/launcher/fetcher.cpp 9323c28237010fa065ef34d74435c151ded530a8 
> 
> Diff: https://reviews.apache.org/r/27483/diff/
> 
> 
> Testing
> -------
> 
> make check
> sudo bin/mesos-tests.sh --verbose
> support/mesos-style.py
> 
> 
> Thanks,
> 
> Ankur Chauhan
> 
>

Reply via email to