> On Dec. 31, 2014, 7:11 a.m., Adam B wrote:
> > include/mesos/mesos.proto, lines 393-395
> > <https://reviews.apache.org/r/28697/diff/3-4/?file=803136#file803136line393>
> >
> > I like the concise form. Now let's name each:
> > ("*", STATIC) - Unreserved
> > (R, STATIC) - Statically Reserved
> > (R, DYNAMIC) - Dynamically Reserved
> >
> > As an aside, why do we even care what `reservation_type` is set to when
> > `role`=="*"? If we just ignore it, then all states are "valid".
We talked about how reserving for role=R within resources with role=R would
effectively be a no-op but we would error to be clear to the framework writers
that what they think they're doing isn't what's going on. This is similar in
that if we let `("*", DYNAMIC)` be valid, framework writers could think they
can dynamically reserve resources with role="*".
> On Dec. 31, 2014, 7:11 a.m., Adam B wrote:
> > include/mesos/mesos.proto, lines 399-400
> > <https://reviews.apache.org/r/28697/diff/3-4/?file=803136#file803136line399>
> >
> > Reworded with the terms defined above:
> > Reservation: ("*", STATIC) => (R, DYNAMIC)
> > "An unreserved resource can become dynamically reserved for a role R."
> >
> > Unreservation: (R, DYNAMIC) => ("*", STATIC)
> > "A dynamically reserved resource can become unreserved."
> >
> > The implied disallowed transitions:
> > "Static reservations cannot be set or changed by a framework. These are
> > set by a cluster administrator on slave start."
Awesome. Thanks for the new wording!
- Michael
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/28697/#review66449
-----------------------------------------------------------
On Dec. 30, 2014, 8:42 p.m., Michael Park wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/28697/
> -----------------------------------------------------------
>
> (Updated Dec. 30, 2014, 8:42 p.m.)
>
>
> Review request for mesos, Adam B, Benjamin Hindman, Ben Mahler, Jie Yu, and
> Vinod Kone.
>
>
> Bugs: MESOS-2137
> https://issues.apache.org/jira/browse/MESOS-2137
>
>
> Repository: mesos-git
>
>
> Description
> -------
>
> Adding new protobuf messages necessary to support dynamic reservations.
>
>
> Diffs
> -----
>
> include/mesos/mesos.proto 540071db64961466eb75c779b3ea6863f4594437
>
> Diff: https://reviews.apache.org/r/28697/diff/
>
>
> Testing
> -------
>
> make check
>
>
> Thanks,
>
> Michael Park
>
>
