> On Feb. 3, 2014, 11:01 a.m., Jiang Yan Xu wrote:
> > src/common/protobuf_utils.hpp, line 105
> > <https://reviews.apache.org/r/17572/diff/1/?file=457210#file457210line105>
> >
> >     I'd like some randomness (here and in master.cpp) too, if a UUID is too 
> > long given that we already have the date info maybe part of  For MESOS-422, 
> > if a master starts twice within a second the `id` would be the same.
> >     
> >     The weird part with date info and random bits is that each new detector 
> > (on master, slave, sched) now generates a different MasterInfo for the same 
> > old contender (which still puts PID in ZK) but I think this is acceptable 
> > because the new master will never use the arbitrarily created 
> > `MasterInfo::id` to get "itself" elected. For other uses only the 
> > `MasterInfo::pid` is looked at so the `id` doesn't matter.

Oh I see that getpid() is used in master.cpp to generate `id` so it is "random" 
and it's OK here.


- Jiang Yan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/17572/#review33466
-----------------------------------------------------------


On Jan. 30, 2014, 4:50 p.m., Vinod Kone wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/17572/
> -----------------------------------------------------------
> 
> (Updated Jan. 30, 2014, 4:50 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Brenden Matthews, Niklas Nielsen, 
> and Jiang Yan Xu.
> 
> 
> Bugs: MESOS-672
>     https://issues.apache.org/jira/browse/MESOS-672
> 
> 
> Repository: mesos-git
> 
> 
> Description
> -------
> 
> This is part 2. 'createMasterInfo()' is used in the subsequent reviews.
> 
> 
> Diffs
> -----
> 
>   src/common/protobuf_utils.hpp c83cbd693abf87089288d03f01ce06200a5d9245 
>   src/common/type_utils.hpp fe6bf71d689f7bfd8b6ae1b8fab9b2e76e28e7a8 
> 
> Diff: https://reviews.apache.org/r/17572/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Vinod Kone
> 
>

Reply via email to