> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > Looks like you forgot to add stout/proc.hpp?

gah. sorry. done


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/slave/reaper.hpp, line 47
> > <https://reviews.apache.org/r/8570/diff/3/?file=249260#file249260line47>
> >
> >     Are you able to programmatically enforce that one of these holds?

not easily, no. esp. 1 and 2, in posix compliant way.

more importantly, what do you mean by "enforce"? what do you propose we do when 
one of those 3 conditions aren't satisfied? log a warning?


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/slave/reaper.hpp, line 51
> > <https://reviews.apache.org/r/8570/diff/3/?file=249260#file249260line51>
> >
> >     I think this is more succinct without losing clarity:
> >     s/addProcess/monitor

I like addProcess because of its similarity to addListener. But, I could be 
convinced otherwise.


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/slave/reaper.hpp, line 61
> > <https://reviews.apache.org/r/8570/diff/3/?file=249260#file249260line61>
> >
> >     Why do we use a set for the listeners?
> >     Can you please change this to a list?
> >     
> >     std::set is ordered, and our listeners do not provide comparison 
> > operations

good catch. done


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/slave/reaper.cpp, line 96
> > <https://reviews.apache.org/r/8570/diff/3/?file=249261#file249261line96>
> >
> >     But what about condition 3 you listed, where we are a privileged user?

Then we souldn't enter this if loop.


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/tests/reaper_tests.cpp, line 68
> > <https://reviews.apache.org/r/8570/diff/3/?file=249263#file249263line68>
> >
> >     Don't use PLOG within a child! We may deadlock, right?

fair enough. using EXIT instead.


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/tests/reaper_tests.cpp, line 76
> > <https://reviews.apache.org/r/8570/diff/3/?file=249263#file249263line76>
> >
> >     kill newline

I like to have a new line between 2 comment blocks if the first one is more 
general and applies to more than the 2nd block. makes sense?


> On Jan. 16, 2013, 11:07 p.m., Ben Mahler wrote:
> > src/tests/reaper_tests.cpp, line 80
> > <https://reviews.apache.org/r/8570/diff/3/?file=249263#file249263line80>
> >
> >     ditto, let's not use any glog stuff in child processes

using EXIT


- Vinod


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/8570/#review15415
-----------------------------------------------------------


On Jan. 16, 2013, 9:54 p.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/8570/
> -----------------------------------------------------------
> 
> (Updated Jan. 16, 2013, 9:54 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman and Ben Mahler.
> 
> 
> Description
> -------
> 
> Needed this to properly monitor the exit status of re-connected executors, as 
> they will be parented by INIT. 
> 
> 
> Diffs
> -----
> 
>   src/Makefile.am c94736df660a25b58dc47c07d9c56c3c26152a66 
>   src/linux/proc.hpp 6c84a73ab1c4d6893390f96eae15422c21001328 
>   src/linux/proc.cpp 99df77447ddfb07a5febec9ebfed396778665041 
>   src/slave/reaper.hpp b9aa62daa42bdaa736ade43884982529ba3d4bb1 
>   src/slave/reaper.cpp c0ee4b4c07fd792bcb39455b666808b712eb32c2 
>   src/tests/master_tests.cpp 948ab5dff34eeba1f3ce593a864ddf282c8b19ed 
>   src/tests/reaper_tests.cpp PRE-CREATION 
>   src/tests/utils.hpp be457117515ee727af101370b26bf9188afb8f45 
>   third_party/libprocess/Makefile.am 6b68c69b07b831255d34a68d7b9b4b37eee55167 
> 
> Diff: https://reviews.apache.org/r/8570/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Reply via email to