mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jan Schlicht <...@mesosphere.io>
Subject Re: Review Request 61271: Implemented HTTP connection handling for the resource provider driver.
Date Thu, 03 Aug 2017 12:43:54 GMT


> On Aug. 3, 2017, noon, Benjamin Bannier wrote:
> > include/mesos/v1/resource_provider.hpp
> > Lines 81 (patched)
> > <https://reviews.apache.org/r/61271/diff/3/?file=1788172#file1788172line82>
> >
> >     Passing an `Option` here makes sense to me, but this is not done in e.g., the
scheduler, https://github.com/apache/mesos/blob/382f526ee2c13df063e17d8346915f3716fe6d21/include/mesos/scheduler.hpp#L364-L366
and I am not sure I follow their reasoning (we also already assume that e.g., libprocess can
be used in this interface).

`mesos/scheduler.hpp` isn't the right place to look, it's actually `mesos/v1/scheduler.hpp`.
There, an `Option` is used as a parameter in the constructor. Maybe stout isn't visible in
`mesos/scheduler.hpp` because it's a public interface that is used by other language bindings.


> On Aug. 3, 2017, noon, Benjamin Bannier wrote:
> > src/Makefile.am
> > Lines 1095 (patched)
> > <https://reviews.apache.org/r/61271/diff/3/?file=1788173#file1788173line1095>
> >
> >     We need this in the cmake setup as well.

I couldn't find a similar section in any of the `CMakeLists.txt`, that's why I didn't add
anything there.


> On Aug. 3, 2017, noon, Benjamin Bannier wrote:
> > src/resource_provider/driver.cpp
> > Lines 43 (patched)
> > <https://reviews.apache.org/r/61271/diff/3/?file=1788174#file1788174line45>
> >
> >     As we only have a single user below, I feel that it would be fine to use a lambda
at the use site instead.

The use site is a constructor's initializer list. While I could use a lambda there, IMHO it
doesn't look very nice and quite distracting.


- Jan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/61271/#review182082
-----------------------------------------------------------


On Aug. 2, 2017, 2:38 p.m., Jan Schlicht wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/61271/
> -----------------------------------------------------------
> 
> (Updated Aug. 2, 2017, 2:38 p.m.)
> 
> 
> Review request for mesos, Benjamin Bannier and Jie Yu.
> 
> 
> Bugs: MESOS-7469
>     https://issues.apache.org/jira/browse/MESOS-7469
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Similar to the existing HTTP connection handling of schedulers and
> executors, the resource provider driver will create two connections
> with the resource provider manager, one for streaming events and another
> one for sending calls. This connection handling has been generalized as
> a 'HttpConnectionProcess' and can be reused in other cases.
> 
> 
> Diffs
> -----
> 
>   include/mesos/v1/resource_provider.hpp 88b606212ea57fee1c1ea522d2dc7f8124a9adef 
>   src/Makefile.am 5712bad2acc4cf0f8ec9b7febffcdb0fa77578c9 
>   src/resource_provider/driver.cpp 6778ec9c863022446f141ee88f70eb563178ea05 
>   src/resource_provider/http_connection.hpp PRE-CREATION 
>   src/resource_provider/storage/provider.cpp 4c39312be5e4a6d783df3d385a66be6b3dcf8603

> 
> 
> Diff: https://reviews.apache.org/r/61271/diff/3/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Jan Schlicht
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message