-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/60351/#review178702
-----------------------------------------------------------
I wonder if we should just change the definition of `operator<<` for `RepeatedPtrField<Resource>`.
The current behavior is to silently omit printing invalid resources, which seems very misleading.
This would also avoid the risk of random `CHECK` failures if a code path attempts to print
a resource before upgrading it.
- Neil Conway
On June 22, 2017, 12:08 a.m., Michael Park wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/60351/
> -----------------------------------------------------------
>
> (Updated June 22, 2017, 12:08 a.m.)
>
>
> Review request for mesos and Neil Conway.
>
>
> Repository: mesos
>
>
> Description
> -------
>
> During the authorization phase, the resources have not been validated
> not converted to the "post-reservation-refinement" format.
> We can't rely on any operations that require valid resources and/or
> "post-reservation-refinement" format. `operator<<` is one of those
> functions, and here we print out the JSON representation of the
> resources instead.
>
>
> Diffs
> -----
>
> src/master/master.cpp ec594a8f4fa95e77fc38103c5561d1797fe2b133
>
>
> Diff: https://reviews.apache.org/r/60351/diff/1/
>
>
> Testing
> -------
>
>
> Thanks,
>
> Michael Park
>
>
|