Review Request 41463: Fixed documentation by removing duplicate document.

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41463/
---

Review request for mesos, Bernd Mathiske and Neil Conway.


Repository: mesos


Description
---

Removed docs/mesos-documentation-guide.md which got replaced by 
docs/documentation-guide.md.


Diffs
-

  docs/mesos-documentation-guide.md 2a4151ba803df2ab51717ea881a4b343d53c43a1 

Diff: https://reviews.apache.org/r/41463/diff/


Testing
---


Thanks,

Till Toenshoff

Review Request 41461: stout: Added SFINAE-friendly `result_of`.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41461/
---

Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description (updated)
---

VS 2015 won't support C++14 `std::result_of` SFINAE until Update 2, so 
`result_of` must be replaced with `decltype(invoke)`.

Here, we implement SFINAE `result_of` in `stout`.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs
-


Diff: https://reviews.apache.org/r/41461/diff/


Testing (updated)
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Review Request 41462: libprocess: Used SFINAE-friendly `result_of`.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41462/
---

Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description
---

Used the SFINAE `result_of` implemented in 
[r41461](https://reviews.apache.org/r/41461/).

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs
-

  3rdparty/libprocess/include/process/future.hpp 
c9146e3a3ccf09dd37c5a8ac7000fbe84f3c710c 

Diff: https://reviews.apache.org/r/41462/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 41461: stout: Added SFINAE-friendly `result_of`.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41461/
---

(Updated Dec. 16, 2015, 8:03 p.m.)


Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description
---

VS 2015 won't support C++14 `std::result_of` SFINAE until Update 2, so 
`result_of` must be replaced with `decltype(invoke)`.

Here, we implement SFINAE `result_of` in `stout`.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs (updated)
-

  3rdparty/libprocess/3rdparty/stout/include/Makefile.am 
d1ef6f0df82e83d8e0d38f7b8986403702519a7d 
  3rdparty/libprocess/3rdparty/stout/include/stout/result_of.hpp PRE-CREATION 

Diff: https://reviews.apache.org/r/41461/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 41461: stout: Added SFINAE-friendly `result_of`.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41461/
---

(Updated Dec. 16, 2015, 8:05 p.m.)


Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description
---

VS 2015 won't support C++14 `std::result_of` SFINAE until Update 2, so 
`result_of` must be replaced with `decltype(invoke)`.

Here, we implement SFINAE `result_of` in `stout`.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs (updated)
-

  3rdparty/libprocess/3rdparty/stout/include/Makefile.am 
d1ef6f0df82e83d8e0d38f7b8986403702519a7d 
  3rdparty/libprocess/3rdparty/stout/include/stout/result_of.hpp PRE-CREATION 

Diff: https://reviews.apache.org/r/41461/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 41459: Invoked `_Deferred`'s `operator F()` explicitly.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41459/
---

(Updated Dec. 16, 2015, 7:38 p.m.)


Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description (updated)
---

VS 2015 won't support C++14 `std::function` SFINAE until Update 2, so 
converting `_Deferred` to `std::function` must be done by explicitly calling 
`_Deferred`'s conversion function.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs
-

  3rdparty/libprocess/include/process/future.hpp 
c9146e3a3ccf09dd37c5a8ac7000fbe84f3c710c 

Diff: https://reviews.apache.org/r/41459/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 39780: Update OversubscriptionTest to not assume dynamic dlopen search.

[James Peach]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39780/
---

(Updated Dec. 16, 2015, 8:14 p.m.)


Review request for mesos, Benjamin Bannier, Kapil Arya, Niklas Nielsen, and 
Till Toenshoff.


Bugs: MESOS-3725
https://issues.apache.org/jira/browse/MESOS-3725


Repository: mesos


Description
---

Update OversubscriptionTest to not assume dynamic dlopen search.


Diffs (updated)
-

  src/tests/oversubscription_tests.cpp 0333281c247dd182860a49f39be791c00679bf6b 

Diff: https://reviews.apache.org/r/39780/diff/


Testing
---

make check


Thanks,

James Peach

Re: Review Request 41346: WIP: Evict some executors if there are not enough reserved resources.

[Mesos ReviewBot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41346/#review110744
---


Bad review!

Reviews applied: []

Error:
 No reviewers specified. Please find a reviewer by asking on JIRA or the 
mailing list.

- Mesos ReviewBot


On Dec. 16, 2015, 1:28 p.m., Guangya Liu wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41346/
> ---
> 
> (Updated Dec. 16, 2015, 1:28 p.m.)
> 
> 
> Review request for mesos.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Evict some executors if there are not enough reserved resources.
> 
> 
> Diffs
> -
> 
>   src/slave/slave.hpp b7586ce42bfac9d9885a3eb8d82deb94680c236c 
>   src/slave/slave.cpp ef869695ffeb2e6d9ef0a78ddb676b1b7cd19afe 
> 
> Diff: https://reviews.apache.org/r/41346/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Guangya Liu
> 
>

Re: Review Request 41075: Added support for implicit roles.

[Alexander Rukletsov]

> On Dec. 15, 2015, 9:56 a.m., Adam B wrote:
> > src/master/allocator/mesos/hierarchical.cpp, lines 1016-1031
> > 
> >
> > If the allocator's roleSorter doesn't know about the role (i.e. no 
> > frameworks are registered with that role), do we still want to make an 
> > explicit `allocate()` call on a SetQuota request for that role? Will 
> > setting quota on an unregistered role have any impact on the fair shares of 
> > other frameworks?
> 
> Neil Conway wrote:
> It is possible for `roleSorter` to not know about the role but for 
> `quotaRoleSorter` to know about it -- i.e., if the role has a configured 
> quota but no active frameworks. In general, I can imagine circumstances in 
> which we would still want to call `allocate()` here (e.g., if we end up 
> revoking/rescinding offers to non-quota frameworks to preserve resources 
> needed by a quota'd role). Perhaps AlexR can comment?

Once quota is set for a role, the next `allocate()` call will lay away 
resources. I think Adam means that we do not necessarily need to allocate 
straight away because there are no consumers (no active frameworks in the 
quota'ed role) of these resources. Hence though quota on an unregistered role 
(I assume it means a role without any frameworks) does impact other frameworks, 
we should not necessarily rush with `allocate()`.

However, does it make sense to introduce a condition here? What is the 
advantage? One extra `allocate()` call should not be a big deal, since set 
quota operation won't happen that often. So even Adam has the point, I would 
leave the code as is for readability.


> On Dec. 15, 2015, 9:56 a.m., Adam B wrote:
> > src/master/http.cpp, lines 1601-1602
> > 
> >
> > Why do you list unregistered roles with quota configured, if you don't 
> > model/display their quota?
> > If you think quota (and roles configured for quota) should show up in 
> > `/roles`, file a JIRA and we'll do it all right in one pass.
> 
> Neil Conway wrote:
> This was per discussion with AlexR. The idea is that, previously, 
> "/roles" showed all the "potentially interesting" roles: by definition, it 
> will include any role that has a non-default weight or non-default quota. 
> With implicit roles, we want to show the same set of roles that have 
> configured properties.
> 
> Whether we show a role with configured quota is separate from whether we 
> show quota information about any of the roles. For the latter, not sure if we 
> want that information in "/roles", quota-related endpoints, or both. In any 
> case, seems fine to defer this part of it.

That's correct. Let me repeat my argument one more time: once a quota becomes 
"whitelisted" or, I prefer, "visible", we should show it in "/roles" to make 
operator's life easier. If an operator sets a quota for a role, they should see 
this role in "/roles", otherwise it becomes very tedious to track all 
"whitelisted" roles and potential typos.


- Alexander


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41075/#review110449
---


On Dec. 16, 2015, 4:19 a.m., Neil Conway wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41075/
> ---
> 
> (Updated Dec. 16, 2015, 4:19 a.m.)
> 
> 
> Review request for mesos, Adam B, Alexander Rukletsov, Benjamin Hindman, Greg 
> Mann, and Yongqiao Wang.
> 
> 
> Bugs: MESOS-4085
> https://issues.apache.org/jira/browse/MESOS-4085
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Changed the behavior of the master when the `--roles` flag is NOT
> specified. Previously, this would allow only the `*` role to be used. Now,
> omitting `--roles` means that any role can be used. This is called "implicit
> roles". Configuring which principals can perform operations as which roles
> should be done using ACLs in the authorization system.
> 
> Note that this changes the behavior of the system when `--roles` is not
> specified. This is likely acceptable: if the operator didn't specify `--roles`
> in prior versions of Mesos, they were likely not using roles or authorization 
> at
> that time.
> 
> Another minor behavioral change is that the "/roles" endpoint will now only
> return results for currently "active" roles (those with one or more registered
> frameworks).
> 
> The `--roles` flag is now considered deprecated and will be removed in a 
> future
> version of Mesos.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 

Re: Review Request 41075: Added support for implicit roles.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41075/
---

(Updated Dec. 16, 2015, 9:08 p.m.)


Review request for mesos, Adam B, Alexander Rukletsov, Benjamin Hindman, Greg 
Mann, and Yongqiao Wang.


Changes
---

Rebase.


Bugs: MESOS-4085
https://issues.apache.org/jira/browse/MESOS-4085


Repository: mesos


Description
---

Changed the behavior of the master when the `--roles` flag is NOT
specified. Previously, this would allow only the `*` role to be used. Now,
omitting `--roles` means that any role can be used. This is called "implicit
roles". Configuring which principals can perform operations as which roles
should be done using ACLs in the authorization system.

Note that this changes the behavior of the system when `--roles` is not
specified. This is likely acceptable: if the operator didn't specify `--roles`
in prior versions of Mesos, they were likely not using roles or authorization at
that time.

Another minor behavioral change is that the "/roles" endpoint will now only
return results for currently "active" roles (those with one or more registered
frameworks).

The `--roles` flag is now considered deprecated and will be removed in a future
version of Mesos.


Diffs (updated)
-

  include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
  include/mesos/master/allocator.proto 702f56f56c3b1331613cecf26522986f6b572f8c 
  src/master/allocator/mesos/allocator.hpp 
97ee80726ad155917811265a983258b0165d3451 
  src/master/allocator/mesos/hierarchical.hpp 
99c742906874c30c39c159e58a65277ade3c07fd 
  src/master/allocator/mesos/hierarchical.cpp 
5da825a1d578a9ee40b4985378fddb3c5fb3b416 
  src/master/http.cpp 8dfd67f7d51b8395953d6beb77b2d71bc538eacd 
  src/master/master.hpp 9aa548aa6e159046c94e4ec96f631ea8b3bfd5d8 
  src/master/master.cpp 0d1482279c68f2a4a27dabaf28774769a5d515c4 
  src/master/quota_handler.cpp 11167879b2480d9c8dd6398ca39c479089ec2272 
  src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
  src/tests/hierarchical_allocator_tests.cpp 
e239b4746494fcc2b362a83afb634a2ce5e25f9b 
  src/tests/master_quota_tests.cpp 0473869783a714766ed26fff61d7f8c56342df74 

Diff: https://reviews.apache.org/r/41075/diff/


Testing
---

"make check" on OSX 10.10 and Ubuntu 15.10; `--gtest_repeat=1000` for the more 
likely role-related tests.

TODOs:

* Add tests for allocation behavior for weights + implicit roles
* More tests for quota + implicit roles?

Notes:

* There's two places where we use manual `new`/`delete` where a `unique_ptr` 
would probably be nicer. I'm inclined to leave this as-is for now though 
(making use of unique_ptr is a broader issue).


Thanks,

Neil Conway

Re: Review Request 41424: Install picojson.h if we are using the bundled version.

[Joseph Wu]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41424/#review110679
---


Looks like this might be problematic on a clean build (reviewbot error on the 
next review):
```
make[7]: *** No rule to make target `picojson-1.3.0/picojson.h', needed by 
`all-am'.  Stop.
make[7]: *** Waiting for unfinished jobs
```

- Joseph Wu


On Dec. 15, 2015, 4:10 p.m., James Peach wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41424/
> ---
> 
> (Updated Dec. 15, 2015, 4:10 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Ben Mahler, Joseph Wu, and Niklas 
> Nielsen.
> 
> 
> Bugs: MESOS-3909
> https://issues.apache.org/jira/browse/MESOS-3909
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The Stout JSON API depends on picojson.h, so we need to install it
> to be able to build Mesos modules outside the Mesos source tree.
> 
> 
> Diffs
> -
> 
>   3rdparty/libprocess/3rdparty/Makefile.am 
> 04d0fa324bf6eab98a51fc08298b9da2992d29d1 
> 
> Diff: https://reviews.apache.org/r/41424/diff/
> 
> 
> Testing
> ---
> 
> Build an external module that implicitly depends on picojson.h
> 
> 
> Thanks,
> 
> James Peach
> 
>

Re: Review Request 39781: Update ModuleTest to not assume dynamic dlopen search.

[James Peach]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39781/
---

(Updated Dec. 16, 2015, 8:17 p.m.)


Review request for mesos, Benjamin Bannier, Kapil Arya, Niklas Nielsen, and 
Till Toenshoff.


Changes
---

Rebased onto master.


Bugs: MESOS-3725
https://issues.apache.org/jira/browse/MESOS-3725


Repository: mesos


Description
---

Update ModuleTest to not assume dynamic dlopen search.


Diffs (updated)
-

  src/tests/module_tests.cpp a3271a3267647e0964dd3decb3ca8384417dd559 

Diff: https://reviews.apache.org/r/39781/diff/


Testing
---

make check


Thanks,

James Peach

Re: Review Request 39780: Update OversubscriptionTest to not assume dynamic dlopen search.

[James Peach]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39780/
---

(Updated Dec. 16, 2015, 8:17 p.m.)


Review request for mesos, Benjamin Bannier, Kapil Arya, Niklas Nielsen, and 
Till Toenshoff.


Changes
---

Rebased onto master.


Bugs: MESOS-3725
https://issues.apache.org/jira/browse/MESOS-3725


Repository: mesos


Description
---

Update OversubscriptionTest to not assume dynamic dlopen search.


Diffs (updated)
-

  src/tests/oversubscription_tests.cpp 0333281c247dd182860a49f39be791c00679bf6b 

Diff: https://reviews.apache.org/r/39780/diff/


Testing
---

make check


Thanks,

James Peach

Re: Review Request 39782: Add a comment for os::libraries::setPaths.

[James Peach]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39782/
---

(Updated Dec. 16, 2015, 8:17 p.m.)


Review request for mesos, Benjamin Bannier, Kapil Arya, Niklas Nielsen, and 
Till Toenshoff.


Changes
---

Rebased onto master.


Bugs: MESOS-3725
https://issues.apache.org/jira/browse/MESOS-3725


Repository: mesos


Description
---

Add a comment for os::libraries::setPaths.


Diffs (updated)
-

  3rdparty/libprocess/3rdparty/stout/include/stout/os.hpp 
14fbca6d222bdfc0e8be301050b4ea1a8a6e7758 

Diff: https://reviews.apache.org/r/39782/diff/


Testing
---

No code changes.


Thanks,

James Peach

Re: Review Request 41454: Added initial draft of executor HTTP API user doc.

[Anand Mazumdar]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41454/
---

(Updated Dec. 16, 2015, 7:38 p.m.)


Review request for mesos and Vinod Kone.


Changes
---

Added the corresponding JIRA link


Bugs: MESOS-4177
https://issues.apache.org/jira/browse/MESOS-4177


Repository: mesos


Description
---

This is an initial draft for the user doc of Executor HTTP API. Some details 
might change due to the pending executor library review chain: 
https://reviews.apache.org/r/41275


Diffs
-

  docs/executor-http-api.md PRE-CREATION 
  docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 

Diff: https://reviews.apache.org/r/41454/diff/


Testing
---

https://gist.github.com/hatred/d35fdaa667203bf34a78


Thanks,

Anand Mazumdar

Review Request 41460: Used `is_bind_expression` to SFINAE correctly.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41460/
---

Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description
---

The Standard (C++11 through 17) does not require `std::bind`'s function call 
operator to SFINAE, and VS 2015's doesn't.  `is_bind_expression` can be used to 
manually reroute bind expressions to the 1-arg overload, where (conveniently) 
the argument will be ignored if necessary.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs
-

  3rdparty/libprocess/include/process/future.hpp 
c9146e3a3ccf09dd37c5a8ac7000fbe84f3c710c 

Diff: https://reviews.apache.org/r/41460/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 41461: stout: Added SFINAE-friendly `result_of`.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41461/
---

(Updated Dec. 16, 2015, 7:50 p.m.)


Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Repository: mesos


Description
---

VS 2015 won't support C++14 `std::result_of` SFINAE until Update 2, so 
`result_of` must be replaced with `decltype(invoke)`.

Here, we implement SFINAE `result_of` in `stout`.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs (updated)
-

  3rdparty/libprocess/3rdparty/stout/include/Makefile.am 
d1ef6f0df82e83d8e0d38f7b8986403702519a7d 
  3rdparty/libprocess/3rdparty/stout/include/stout/result_of.hpp PRE-CREATION 

Diff: https://reviews.apache.org/r/41461/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 41461: stout: Added SFINAE-friendly `result_of`.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41461/
---

(Updated Dec. 16, 2015, 7:57 p.m.)


Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Changes
---

Added comments.


Repository: mesos


Description
---

VS 2015 won't support C++14 `std::result_of` SFINAE until Update 2, so 
`result_of` must be replaced with `decltype(invoke)`.

Here, we implement SFINAE `result_of` in `stout`.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs (updated)
-

  3rdparty/libprocess/3rdparty/stout/include/Makefile.am 
d1ef6f0df82e83d8e0d38f7b8986403702519a7d 
  3rdparty/libprocess/3rdparty/stout/include/stout/result_of.hpp PRE-CREATION 

Diff: https://reviews.apache.org/r/41461/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Re: Review Request 41446: Fixed character encoding in docs.

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41446/
---

(Updated Dec. 16, 2015, 8:32 p.m.)


Review request for mesos, Bernd Mathiske and Neil Conway.


Repository: mesos


Description
---

see summary.


Diffs
-

  docs/maintenance.md a5831ffa092a9ea6decbe2e640bae637c759a308 
  docs/networking-for-mesos-managed-containers.md 
594fe2b088ac318a979c86795af6e5db5505499e 
  docs/reconciliation.md 0adfbf1a78c70a12adcbeb3aebd1045452af16ac 
  docs/scheduler-http-api.md 38d3607edb2145b8525268cd56fe4bfc9917f95a 
  docs/testing-patterns.md e5fd989cf910640b70da2ac89bf886a61b940aad 

Diff: https://reviews.apache.org/r/41446/diff/


Testing
---


Thanks,

Till Toenshoff

Re: Review Request 41454: Added initial draft of executor HTTP API user doc.

[Anand Mazumdar]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41454/
---

(Updated Dec. 16, 2015, 8:40 p.m.)


Review request for mesos and Vinod Kone.


Changes
---

Fixed character encoding


Bugs: MESOS-4177
https://issues.apache.org/jira/browse/MESOS-4177


Repository: mesos


Description
---

This is an initial draft for the user doc of Executor HTTP API. Some details 
might change due to the pending executor library review chain: 
https://reviews.apache.org/r/41275


Diffs (updated)
-

  docs/executor-http-api.md PRE-CREATION 
  docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 

Diff: https://reviews.apache.org/r/41454/diff/


Testing
---

https://gist.github.com/hatred/d35fdaa667203bf34a78


Thanks,

Anand Mazumdar

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/#review110749
---

Ship it!


Style-wise, the file should probably be named "provisioner.md" for consistency 
with the other documentation files. But it is fine to commit this RR as-is for 
the time being.

- Neil Conway


On Dec. 16, 2015, 9:15 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41467/
> ---
> 
> (Updated Dec. 16, 2015, 9:15 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/41467/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41460: Used `std::is_bind_expression` to SFINAE correctly.

[Michael Park]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41460/
---

(Updated Dec. 16, 2015, 7:44 p.m.)


Review request for mesos, Alex Clemmer and Joris Van Remoortere.


Summary (updated)
-

Used `std::is_bind_expression` to SFINAE correctly.


Repository: mesos


Description (updated)
---

The Standard (C++11 through 17) does not require `std::bind`'s function call 
operator to SFINAE, and VS 2015's doesn't. `std::is_bind_expression` can be 
used to manually reroute bind expressions to the 1-arg overload, where 
(conveniently) the argument will be ignored if necessary.

Follow-up from [r40114](https://reviews.apache.org/r/40114/).


Diffs
-

  3rdparty/libprocess/include/process/future.hpp 
c9146e3a3ccf09dd37c5a8ac7000fbe84f3c710c 

Diff: https://reviews.apache.org/r/41460/diff/


Testing
---

`make check` on OS X, compiled on Windows.


Thanks,

Michael Park

Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/
---

Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.


Repository: mesos


Description
---

see summary.


Diffs
-

  docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 

Diff: https://reviews.apache.org/r/41467/diff/


Testing
---


Thanks,

Till Toenshoff

Re: Review Request 41075: Added support for implicit roles.

[Neil Conway]

> On Dec. 15, 2015, 9:56 a.m., Adam B wrote:
> > src/master/allocator/mesos/hierarchical.cpp, lines 1016-1031
> > 
> >
> > If the allocator's roleSorter doesn't know about the role (i.e. no 
> > frameworks are registered with that role), do we still want to make an 
> > explicit `allocate()` call on a SetQuota request for that role? Will 
> > setting quota on an unregistered role have any impact on the fair shares of 
> > other frameworks?

It is possible for `roleSorter` to not know about the role but for 
`quotaRoleSorter` to know about it -- i.e., if the role has a configured quota 
but no active frameworks. In general, I can imagine circumstances in which we 
would still want to call `allocate()` here (e.g., if we end up 
revoking/rescinding offers to non-quota frameworks to preserve resources needed 
by a quota'd role). Perhaps AlexR can comment?


> On Dec. 15, 2015, 9:56 a.m., Adam B wrote:
> > src/master/allocator/mesos/hierarchical.cpp, line 1130
> > 
> >
> > If there are no registered frameworks in any role, do we want to 
> > early-exit from the allocate() call?

I think we'll exit quite quickly regardless, so I'd vote against adding an 
early-exit.


- Neil


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41075/#review110449
---


On Dec. 16, 2015, 4:19 a.m., Neil Conway wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41075/
> ---
> 
> (Updated Dec. 16, 2015, 4:19 a.m.)
> 
> 
> Review request for mesos, Adam B, Alexander Rukletsov, Benjamin Hindman, Greg 
> Mann, and Yongqiao Wang.
> 
> 
> Bugs: MESOS-4085
> https://issues.apache.org/jira/browse/MESOS-4085
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Changed the behavior of the master when the `--roles` flag is NOT
> specified. Previously, this would allow only the `*` role to be used. Now,
> omitting `--roles` means that any role can be used. This is called "implicit
> roles". Configuring which principals can perform operations as which roles
> should be done using ACLs in the authorization system.
> 
> Note that this changes the behavior of the system when `--roles` is not
> specified. This is likely acceptable: if the operator didn't specify `--roles`
> in prior versions of Mesos, they were likely not using roles or authorization 
> at
> that time.
> 
> Another minor behavioral change is that the "/roles" endpoint will now only
> return results for currently "active" roles (those with one or more registered
> frameworks).
> 
> The `--roles` flag is now considered deprecated and will be removed in a 
> future
> version of Mesos.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 702f56f56c3b1331613cecf26522986f6b572f8c 
>   src/master/allocator/mesos/allocator.hpp 
> 97ee80726ad155917811265a983258b0165d3451 
>   src/master/allocator/mesos/hierarchical.hpp 
> 99c742906874c30c39c159e58a65277ade3c07fd 
>   src/master/allocator/mesos/hierarchical.cpp 
> 5da825a1d578a9ee40b4985378fddb3c5fb3b416 
>   src/master/http.cpp 37fbcb93074fe189133161afaa28046dc2ad0731 
>   src/master/master.hpp 9aa548aa6e159046c94e4ec96f631ea8b3bfd5d8 
>   src/master/master.cpp 0d1482279c68f2a4a27dabaf28774769a5d515c4 
>   src/master/quota_handler.cpp 11167879b2480d9c8dd6398ca39c479089ec2272 
>   src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
>   src/tests/hierarchical_allocator_tests.cpp 
> e239b4746494fcc2b362a83afb634a2ce5e25f9b 
>   src/tests/master_quota_tests.cpp 0473869783a714766ed26fff61d7f8c56342df74 
> 
> Diff: https://reviews.apache.org/r/41075/diff/
> 
> 
> Testing
> ---
> 
> "make check" on OSX 10.10 and Ubuntu 15.10; `--gtest_repeat=1000` for the 
> more likely role-related tests.
> 
> TODOs:
> 
> * Add tests for allocation behavior for weights + implicit roles
> * More tests for quota + implicit roles?
> 
> Notes:
> 
> * There's two places where we use manual `new`/`delete` where a `unique_ptr` 
> would probably be nicer. I'm inclined to leave this as-is for now though 
> (making use of unique_ptr is a broader issue).
> 
> 
> Thanks,
> 
> Neil Conway
> 
>

Re: Review Request 40995: Added test cases for role behavior.

[Neil Conway]

> On Dec. 15, 2015, 10:18 a.m., Adam B wrote:
> > Thanks for validating the expected `--roles` behavior with tests, but I 
> > think you're mixing up the scheduler reservation/volume API (where only the 
> > framework's role is valid) with the operator API (where any whitelisted 
> > role is valid).

Ah, good point! Turns out that we weren't properly enforcing `--roles` for 
dynamic reservations anyway (see MESOS-4143). I'll remove these tests for now 
and fix 4143 shortly.


- Neil


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40995/#review110460
---


On Dec. 16, 2015, 9:05 p.m., Neil Conway wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40995/
> ---
> 
> (Updated Dec. 16, 2015, 9:05 p.m.)
> 
> 
> Review request for mesos, Adam B, Alexander Rukletsov, Greg Mann, and 
> Yongqiao Wang.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Added test cases for role behavior.
> 
> 
> Diffs
> -
> 
>   src/Makefile.am 8f6b98b5b0ddbfe6b97346704cb96937e0eca02e 
>   src/tests/reservation_tests.cpp f429472e6b93a5d6d8fe6a5f7d5b94fc331f7295 
>   src/tests/role_tests.cpp PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/40995/diff/
> 
> 
> Testing
> ---
> 
> make check.
> 
> 
> Thanks,
> 
> Neil Conway
> 
>

Re: Review Request 41463: Fixed documentation by removing duplicate document.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41463/#review110733
---

Ship it!


Ship It!

- Neil Conway


On Dec. 16, 2015, 7:47 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41463/
> ---
> 
> (Updated Dec. 16, 2015, 7:47 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske and Neil Conway.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Removed docs/mesos-documentation-guide.md which got replaced by 
> docs/documentation-guide.md.
> 
> 
> Diffs
> -
> 
>   docs/mesos-documentation-guide.md 2a4151ba803df2ab51717ea881a4b343d53c43a1 
> 
> Diff: https://reviews.apache.org/r/41463/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41463: Fixed documentation by removing duplicate document.

[Bernd Mathiske]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41463/#review110737
---

Ship it!


Ship It!

- Bernd Mathiske


On Dec. 16, 2015, 11:47 a.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41463/
> ---
> 
> (Updated Dec. 16, 2015, 11:47 a.m.)
> 
> 
> Review request for mesos, Bernd Mathiske and Neil Conway.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Removed docs/mesos-documentation-guide.md which got replaced by 
> docs/documentation-guide.md.
> 
> 
> Diffs
> -
> 
>   docs/mesos-documentation-guide.md 2a4151ba803df2ab51717ea881a4b343d53c43a1 
> 
> Diff: https://reviews.apache.org/r/41463/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41225: Added test cases for implicit roles.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41225/
---

(Updated Dec. 16, 2015, 9:06 p.m.)


Review request for mesos, Adam B, Alexander Rukletsov, and Yongqiao Wang.


Changes
---

Rebase.


Repository: mesos


Description
---

Added test cases for implicit roles.


Diffs (updated)
-

  src/tests/role_tests.cpp PRE-CREATION 

Diff: https://reviews.apache.org/r/41225/diff/


Testing
---


Thanks,

Neil Conway

Re: Review Request 40995: Added test cases for role behavior.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40995/
---

(Updated Dec. 16, 2015, 9:05 p.m.)


Review request for mesos, Adam B, Alexander Rukletsov, Greg Mann, and Yongqiao 
Wang.


Changes
---

Remove tests for explicit role whitelist + dynamic reservations/persistent 
volumes.


Repository: mesos


Description
---

Added test cases for role behavior.


Diffs (updated)
-

  src/Makefile.am 8f6b98b5b0ddbfe6b97346704cb96937e0eca02e 
  src/tests/reservation_tests.cpp f429472e6b93a5d6d8fe6a5f7d5b94fc331f7295 
  src/tests/role_tests.cpp PRE-CREATION 

Diff: https://reviews.apache.org/r/40995/diff/


Testing
---

make check.


Thanks,

Neil Conway

Re: Review Request 41425: Fix the Mesos pkg-config.

[James Peach]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41425/
---

(Updated Dec. 16, 2015, 5:26 p.m.)


Review request for mesos, Benjamin Hindman, Ben Mahler, Joseph Wu, and Niklas 
Nielsen.


Changes
---

Rebased onto master.


Bugs: MESOS-3909
https://issues.apache.org/jira/browse/MESOS-3909


Repository: mesos


Description
---

The pkg-config file needs to explicitly set the -I path, and also
set and preprocessor definitions that Mesos assumes will be set for
picojson.h.


Diffs (updated)
-

  mesos.pc.in bc4f7e778dd8e5e2bd5e8f4181d62934ac423135 

Diff: https://reviews.apache.org/r/41425/diff/


Testing
---

Check various pkg-config invocations and make sure they are sane:

$ PKG_CONFIG_PATH=/opt/mesos/lib/pkgconfig pkg-config --libs mesos
-L/opt/mesos/lib -lmesos
$ PKG_CONFIG_PATH=/opt/mesos/lib/pkgconfig pkg-config --cflags mesos
-DPICOJSON_USE_INT64 -D__STDC_FORMAT_MACROS -I/opt/mesos/include
$ PKG_CONFIG_PATH=/opt/mesos/lib/pkgconfig pkg-config --cflags-only-I mesos
-I/opt/mesos/include


Thanks,

James Peach

Re: Review Request 41411: Added flags to update summary / desc for reviews.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41411/#review110690
---

Ship it!


Ship It!

- Neil Conway


On Dec. 15, 2015, 10:38 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41411/
> ---
> 
> (Updated Dec. 15, 2015, 10:38 p.m.)
> 
> 
> Review request for mesos and Neil Conway.
> 
> 
> Bugs: MESOS-4180
> https://issues.apache.org/jira/browse/MESOS-4180
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Previously if you amended a commit message and posted it to reviewboard
> via the support/post_reviews.py script, then the summary / description
> on reviewboard would not be updated. The only way to update these was to
> modify them directly on the webpage. However, with a simple directive in the
> .reviewboardrc file we can force the review's summary / description to
> be updated to the text in the commit message. This commit adds these
> directives to the default .reviewboardrc.
> 
> Review: https://reviews.apache.org/r/41411
> 
> 
> Diffs
> -
> 
>   support/reviewboardrc 56bd8a59b1f3bed85241bdc133cc1127d4bff94d 
> 
> Diff: https://reviews.apache.org/r/41411/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Till Toenshoff]

> On Dec. 16, 2015, 9:19 p.m., Neil Conway wrote:
> > Style-wise, the file should probably be named "provisioner.md" for 
> > consistency with the other documentation files. But it is fine to commit 
> > this RR as-is for the time being.

Thanks a bunch - do you think my summary (used behind the link) is fine?


- Till


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/#review110749
---


On Dec. 16, 2015, 9:15 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41467/
> ---
> 
> (Updated Dec. 16, 2015, 9:15 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/41467/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41075: Added support for implicit roles.

[Adam B]

> On Dec. 15, 2015, 1:56 a.m., Adam B wrote:
> > src/master/allocator/mesos/hierarchical.cpp, lines 1016-1031
> > 
> >
> > If the allocator's roleSorter doesn't know about the role (i.e. no 
> > frameworks are registered with that role), do we still want to make an 
> > explicit `allocate()` call on a SetQuota request for that role? Will 
> > setting quota on an unregistered role have any impact on the fair shares of 
> > other frameworks?
> 
> Neil Conway wrote:
> It is possible for `roleSorter` to not know about the role but for 
> `quotaRoleSorter` to know about it -- i.e., if the role has a configured 
> quota but no active frameworks. In general, I can imagine circumstances in 
> which we would still want to call `allocate()` here (e.g., if we end up 
> revoking/rescinding offers to non-quota frameworks to preserve resources 
> needed by a quota'd role). Perhaps AlexR can comment?
> 
> Alexander Rukletsov wrote:
> Once quota is set for a role, the next `allocate()` call will lay away 
> resources. I think Adam means that we do not necessarily need to allocate 
> straight away because there are no consumers (no active frameworks in the 
> quota'ed role) of these resources. Hence though quota on an unregistered role 
> (I assume it means a role without any frameworks) does impact other 
> frameworks, we should not necessarily rush with `allocate()`.
> 
> However, does it make sense to introduce a condition here? What is the 
> advantage? One extra `allocate()` call should not be a big deal, since set 
> quota operation won't happen that often. So even Adam has the point, I would 
> leave the code as is for readability.

"Adam means that we do not necessarily need to allocate straight away because 
there are no consumers"
That is indeed what I meant.
"One extra allocate() call should not be a big deal... I would leave the code 
as is for readability."
SGTM. Dropping the issue.


> On Dec. 15, 2015, 1:56 a.m., Adam B wrote:
> > src/master/http.cpp, line 316
> > 
> >
> > I'm in agreement with Yong. This form of `model(role, name, weight)` 
> > isn't used anywhere, and is confusing alongside `modelRole(roleName, 
> > weight, Option)`. Remove it and rename the other one to fit the 
> > `model(...)` pattern.
> 
> Neil Conway wrote:
> Whoops, this version of `model(role, name, weight)` was included 
> accidentally (it is from an old version of the patch). Will remove.
> 
> I'm not sure that changing `modelRole()` to use the `model(...)` naming 
> pattern is the right thing to do: the pattern is that `model(X)` returns a 
> JSON object modeled after `X`. In this case, the parameters will be `string 
> name, double weight, Option`, so (a) it doesn't quite fit (b) from the 
> function name + arguments, it is a little unclear that `model(string name, 
> double weight, Option)` is supposed to do.

Although precedent exists, I would agree that it is also unclear what the 
following existing `model(...)` call is supposed to do:
JSON::Object model(
const TaskInfo& task,
const FrameworkID& frameworkId,
const TaskState& state,
const std::vector& statuses);


> On Dec. 15, 2015, 1:56 a.m., Adam B wrote:
> > src/master/http.cpp, lines 1601-1602
> > 
> >
> > Why do you list unregistered roles with quota configured, if you don't 
> > model/display their quota?
> > If you think quota (and roles configured for quota) should show up in 
> > `/roles`, file a JIRA and we'll do it all right in one pass.
> 
> Neil Conway wrote:
> This was per discussion with AlexR. The idea is that, previously, 
> "/roles" showed all the "potentially interesting" roles: by definition, it 
> will include any role that has a non-default weight or non-default quota. 
> With implicit roles, we want to show the same set of roles that have 
> configured properties.
> 
> Whether we show a role with configured quota is separate from whether we 
> show quota information about any of the roles. For the latter, not sure if we 
> want that information in "/roles", quota-related endpoints, or both. In any 
> case, seems fine to defer this part of it.
> 
> Alexander Rukletsov wrote:
> That's correct. Let me repeat my argument one more time: once a quota 
> becomes "whitelisted" or, I prefer, "visible", we should show it in "/roles" 
> to make operator's life easier. If an operator sets a quota for a role, they 
> should see this role in "/roles", otherwise it becomes very tedious to track 
> all "whitelisted" roles and potential typos.

Fair enough. Dropping this issue.
Just make sure that ROLES_HELP accurately explains this. I'll open a separate 
issue for that.


- Adam


---
This is an 

Review Request 41472: Disallowed dynamic reservations for roles not on the role whitelist.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41472/
---

Review request for mesos, Adam B and Michael Park.


Bugs: MESOS-4143
https://issues.apache.org/jira/browse/MESOS-4143


Repository: mesos


Description
---

Also added a test that dynamic reservations via the "/reserve" endpoint are
allowed when using implicit roles.


Diffs
-

  src/master/http.cpp 8dfd67f7d51b8395953d6beb77b2d71bc538eacd 
  src/tests/reservation_endpoints_tests.cpp 
b8edd6fafedd4c2221a8d19c1ebc71254071a8c7 

Diff: https://reviews.apache.org/r/41472/diff/


Testing
---

make check


Thanks,

Neil Conway

Re: Review Request 41361: Wrote cmd to create WANdiscoSVN file for centos.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41361/#review110787
---

Ship it!


Seems fine as far as it goes, but it would be great to find a more standard 
source for the packages than WANdisco. For example, 'devtoolset-3' appears to 
have what we need.

- Neil Conway


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41361/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Previously we just told users to create this file and then add the
> contents to it. Now they can just copy and paste a single command to
> create the file and fill the contents automatically.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41361/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Neil Conway]

> On Dec. 16, 2015, 9:19 p.m., Neil Conway wrote:
> > Style-wise, the file should probably be named "provisioner.md" for 
> > consistency with the other documentation files. But it is fine to commit 
> > this RR as-is for the time being.
> 
> Till Toenshoff wrote:
> Thanks a bunch - do you think my summary (used behind the link) is fine?

"for supporting provisioning" is a bit awkward; I'd just say "for 
provisioning". Otherwise LGTM.


- Neil


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/#review110749
---


On Dec. 16, 2015, 9:15 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41467/
> ---
> 
> (Updated Dec. 16, 2015, 9:15 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/41467/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/#review110757
---

Ship it!


Ship It!

- Gilbert Song


On Dec. 16, 2015, 1:27 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41467/
> ---
> 
> (Updated Dec. 16, 2015, 1:27 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/41467/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41364: Added NOTE to Examples in Getting Started on prod.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41364/#review110790
---

Ship it!


Ship It!


docs/getting-started.md (line 194)


Can we capitalize "Mesos" and "IP"?


- Neil Conway


On Dec. 16, 2015, 11:08 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41364/
> ---
> 
> (Updated Dec. 16, 2015, 11:08 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The current instructions assume you are launching mesos on your local
> machine. This could be confusing for someone trying to run mesos on
> (e.g. AWS) and not understanding why they can't access the mesos
> web page. This note adds a comment to try and hinder such confusion.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41364/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41362: Added note in docs for installing wget, tar, git.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41362/#review110788
---

Ship it!


Ship It!

- Neil Conway


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41362/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The 'Downloading Mesos' section of the getting started guide
> assumes that you already have the 'wget', 'tar', and 'git' utilities
> installed on your system before running through the 'System
> Requirements' section below it. This commit adds a note to first run
> through the 'System Requirments' section if problems are encountered
> with downloading mesos via the provided instructions.
> 
> It also adds the 'wget', 'tar', and 'git' utilities to the list of
> requirements for all systems in the 'System Requirements' section.
> The two CentOS sections already had instructions for 'wget' and 'tar',
> but now 'git' has been added to the list.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41362/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41363: Reworded NOTE in Examples for Getting Started doc.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41363/#review110789
---

Ship it!


Ship It!

- Neil Conway


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41363/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The previous wording was a bit confusing. The new text makes it clear
> that the test-framework binaries have been generated by a call to
> `make check` as described in the 'Building Mesos' section.
> 
> Also, I changed the text from being a footnote to being actual text
> inline with the rest of the text in this section. The note is relevant
> enough to the text above it to warrant such a change.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41363/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41476: Updated getting started for 0.26 release.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41476/#review110799
---

Ship it!


Ship It!

- Neil Conway


On Dec. 16, 2015, 11:51 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41476/
> ---
> 
> (Updated Dec. 16, 2015, 11:51 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Artem Harutyunyan, Neil Conway, 
> and Till Toenshoff.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The old release in the documentation was 0.24. Looks like we just jumped
> over 0.25...
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41476/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41360: Aligned the code blocks in Getting Started better.

[Kevin Klues]

> On Dec. 16, 2015, 11:36 p.m., Neil Conway wrote:
> > Looks good! While we're at it, can we update the example version in the 
> > first `wget` command to be 0.26?
> 
> Kevin Klues wrote:
> I'll do that, but as a separate commit.

https://reviews.apache.org/r/41476/


- Kevin


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41360/#review110793
---


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41360/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Now each line starts at the beginning of the left hand margin, so that
> multiple lines can be copied without extra space at the beginning of
> each line. This is important for multiline commands (e.g. printing
> multi-line strings to a file) so they can easily be copy and pasted
> without having to modify the lines.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41360/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41075: Added support for implicit roles.

[Neil Conway]

> On Dec. 15, 2015, 9:56 a.m., Adam B wrote:
> > src/master/http.cpp, line 316
> > 
> >
> > I'm in agreement with Yong. This form of `model(role, name, weight)` 
> > isn't used anywhere, and is confusing alongside `modelRole(roleName, 
> > weight, Option)`. Remove it and rename the other one to fit the 
> > `model(...)` pattern.
> 
> Neil Conway wrote:
> Whoops, this version of `model(role, name, weight)` was included 
> accidentally (it is from an old version of the patch). Will remove.
> 
> I'm not sure that changing `modelRole()` to use the `model(...)` naming 
> pattern is the right thing to do: the pattern is that `model(X)` returns a 
> JSON object modeled after `X`. In this case, the parameters will be `string 
> name, double weight, Option`, so (a) it doesn't quite fit (b) from the 
> function name + arguments, it is a little unclear that `model(string name, 
> double weight, Option)` is supposed to do.
> 
> Adam B wrote:
> Although precedent exists, I would agree that it is also unclear what the 
> following existing `model(...)` call is supposed to do:
> JSON::Object model(
> const TaskInfo& task,
> const FrameworkID& frameworkId,
> const TaskState& state,
> const std::vector& statuses);

Sure, I renamed `modelRole()` -> `model()`.


- Neil


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41075/#review110449
---


On Dec. 16, 2015, 10:43 p.m., Neil Conway wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41075/
> ---
> 
> (Updated Dec. 16, 2015, 10:43 p.m.)
> 
> 
> Review request for mesos, Adam B, Alexander Rukletsov, Benjamin Hindman, Greg 
> Mann, and Yongqiao Wang.
> 
> 
> Bugs: MESOS-4085
> https://issues.apache.org/jira/browse/MESOS-4085
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Changed the behavior of the master when the `--roles` flag is NOT
> specified. Previously, this would allow only the `*` role to be used. Now,
> omitting `--roles` means that any role can be used. This is called "implicit
> roles". Configuring which principals can perform operations as which roles
> should be done using ACLs in the authorization system.
> 
> Note that this changes the behavior of the system when `--roles` is not
> specified. This is likely acceptable: if the operator didn't specify `--roles`
> in prior versions of Mesos, they were likely not using roles or authorization 
> at
> that time.
> 
> Another minor behavioral change is that the "/roles" endpoint will now only
> return results for currently "active" roles (those with one or more registered
> frameworks).
> 
> The `--roles` flag is now considered deprecated and will be removed in a 
> future
> version of Mesos.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 702f56f56c3b1331613cecf26522986f6b572f8c 
>   src/master/allocator/mesos/allocator.hpp 
> 97ee80726ad155917811265a983258b0165d3451 
>   src/master/allocator/mesos/hierarchical.hpp 
> 99c742906874c30c39c159e58a65277ade3c07fd 
>   src/master/allocator/mesos/hierarchical.cpp 
> 5da825a1d578a9ee40b4985378fddb3c5fb3b416 
>   src/master/http.cpp 8dfd67f7d51b8395953d6beb77b2d71bc538eacd 
>   src/master/master.hpp 9aa548aa6e159046c94e4ec96f631ea8b3bfd5d8 
>   src/master/master.cpp 0d1482279c68f2a4a27dabaf28774769a5d515c4 
>   src/master/quota_handler.cpp 11167879b2480d9c8dd6398ca39c479089ec2272 
>   src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
>   src/tests/hierarchical_allocator_tests.cpp 
> e239b4746494fcc2b362a83afb634a2ce5e25f9b 
>   src/tests/master_quota_tests.cpp 0473869783a714766ed26fff61d7f8c56342df74 
> 
> Diff: https://reviews.apache.org/r/41075/diff/
> 
> 
> Testing
> ---
> 
> "make check" on OSX 10.10 and Ubuntu 15.10; `--gtest_repeat=1000` for the 
> more likely role-related tests.
> 
> TODOs:
> 
> * Add tests for allocation behavior for weights + implicit roles
> * More tests for quota + implicit roles?
> 
> Notes:
> 
> * There's two places where we use manual `new`/`delete` where a `unique_ptr` 
> would probably be nicer. I'm inclined to leave this as-is for now though 
> (making use of unique_ptr is a broader issue).
> 
> 
> Thanks,
> 
> Neil Conway
> 
>

Re: Review Request 41360: Aligned the code blocks in Getting Started better.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41360/#review110793
---

Ship it!


Looks good! While we're at it, can we update the example version in the first 
`wget` command to be 0.26?

- Neil Conway


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41360/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Now each line starts at the beginning of the left hand margin, so that
> multiple lines can be copied without extra space at the beginning of
> each line. This is important for multiline commands (e.g. printing
> multi-line strings to a file) so they can easily be copy and pasted
> without having to modify the lines.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41360/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/
---

(Updated Dec. 16, 2015, 9:27 p.m.)


Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.


Changes
---

Fixed link description wording.


Repository: mesos


Description
---

see summary.


Diffs (updated)
-

  docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 

Diff: https://reviews.apache.org/r/41467/diff/


Testing
---


Thanks,

Till Toenshoff

Review Request 41468: Removed MesosCon Europe from site index page.

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41468/
---

Review request for mesos, Bernd Mathiske and Neil Conway.


Bugs: MESOS-3742
https://issues.apache.org/jira/browse/MESOS-3742


Repository: mesos


Description
---

see summary.


Diffs
-

  site/source/index.html.md c5b56417ee8ac9aa5ca41da9bc2a2396edcb5f1a 

Diff: https://reviews.apache.org/r/41468/diff/


Testing
---


Thanks,

Till Toenshoff

Re: Review Request 41463: Fixed documentation by removing duplicate document.

[Guangya Liu]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41463/#review110785
---

Ship it!


Ship It!

- Guangya Liu


On Dec. 16, 2015, 7:47 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41463/
> ---
> 
> (Updated Dec. 16, 2015, 7:47 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske and Neil Conway.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Removed docs/mesos-documentation-guide.md which got replaced by 
> docs/documentation-guide.md.
> 
> 
> Diffs
> -
> 
>   docs/mesos-documentation-guide.md 2a4151ba803df2ab51717ea881a4b343d53c43a1 
> 
> Diff: https://reviews.apache.org/r/41463/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41278: Added note to site-docker doc about tunneling.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41278/#review110784
---

Ship it!


Ship It!

- Neil Conway


On Dec. 15, 2015, 1:06 a.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41278/
> ---
> 
> (Updated Dec. 15, 2015, 1:06 a.m.)
> 
> 
> Review request for mesos and Neil Conway.
> 
> 
> Bugs: MESOS-4134
> https://issues.apache.org/jira/browse/MESOS-4134
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> If we are running the site-docker container on a remote machine, we
> should set up a tunnel to localhost to view the site locally. This
> commit adds a note about how to do so.
> 
> 
> Diffs
> -
> 
>   support/site-docker/README.md 80d34f33a5089214522b9c936710ccd29038a91a 
> 
> Diff: https://reviews.apache.org/r/41278/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41371: Fixed getting started instructions for centos 7.1.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41371/#review110796
---

Ship it!


Ship It!

- Gilbert Song


On Dec. 16, 2015, 3:06 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41371/
> ---
> 
> (Updated Dec. 16, 2015, 3:06 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Gilbert Song, Jojy Varghese, and 
> Neil Conway.
> 
> 
> Bugs: MESOS-3842
> https://issues.apache.org/jira/browse/MESOS-3842
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The current instructions are broken for installing the development tools
> on centos 7.1 after updating the WANdisco SVN repo file. Specifically,
> the updated svn installation requires libserf-1, which is not part of
> the base yum repository.  We need to install the epel repository to get
> access to libserf-1.
> 
> As part of this, I also update the instructions to install
> java-1.8.0-openjdk-devel instead of java-1.7.0-openjdk-devel. This is
> necessary because our instructions have us installing apache-maven
> BEFORE installing java-1.7.0-openjdk-devel. As such, installing maven
> installs the default jre from the base repository (java-1.8.0-openjdk)
> as one of its dependencies. This causes problems with finding the proper
> java.home in our mesos/configure script because of the mismatch between
> the most up to date jre (1.8.0) and the most up to date development
> tools (1.7.0). Alternative solutions to resolve this problem probably
> exist (i.e. move our dependence on maven until AFTER our installation of
> java-1.7.0-openjdk-devel), but this would likely break if running on a
> system that already had the 1.8.0 jre installed for other purposes. Best
> to just update to the newwest default version of the java development
> tools.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41371/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/
---

(Updated Dec. 16, 2015, 9:40 p.m.)


Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.


Repository: mesos


Description
---

see summary.


Diffs
-

  docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 

Diff: https://reviews.apache.org/r/41467/diff/


Testing
---


Thanks,

Till Toenshoff

Re: Review Request 39456: Documentation: added containerizer internals

[Jojy Varghese]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39456/
---

(Updated Dec. 16, 2015, 9:48 p.m.)


Review request for mesos, Jie Yu and Timothy Chen.


Changes
---

addressed reviews.


Repository: mesos


Description
---

Documentation: added containerizer internals


Diffs (updated)
-

  docs/containerizer-internals.md PRE-CREATION 
  docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 

Diff: https://reviews.apache.org/r/39456/diff/


Testing
---


Thanks,

Jojy Varghese

Re: Review Request 41467: Added docs/mesos-provisioner.md to docs/home.md

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41467/#review110772
---

Ship it!


Ship It!

- Timothy Chen


On Dec. 16, 2015, 9:40 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41467/
> ---
> 
> (Updated Dec. 16, 2015, 9:40 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Neil Conway, and Timothy Chen.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/41467/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41468: Removed MesosCon Europe from site index page.

[Neil Conway]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41468/#review110773
---

Ship it!


I didn't preview with mesos-website-containerizer -- someone should do that at 
some point. But otherwise LGTM.

- Neil Conway


On Dec. 16, 2015, 9:38 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41468/
> ---
> 
> (Updated Dec. 16, 2015, 9:38 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske and Neil Conway.
> 
> 
> Bugs: MESOS-3742
> https://issues.apache.org/jira/browse/MESOS-3742
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   site/source/index.html.md c5b56417ee8ac9aa5ca41da9bc2a2396edcb5f1a 
> 
> Diff: https://reviews.apache.org/r/41468/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41468: Removed MesosCon Europe from site index page.

[Till Toenshoff]

> On Dec. 16, 2015, 10:34 p.m., Neil Conway wrote:
> > I didn't preview with mesos-website-containerizer -- someone should do that 
> > at some point. But otherwise LGTM.

Yes, I am in the process of upgrading the site using the mesos-site docker 
container from `support/site-docker/`. Will do further checking and land these 
and the other site related changes to come afterwards.


- Till


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41468/#review110773
---


On Dec. 16, 2015, 9:38 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41468/
> ---
> 
> (Updated Dec. 16, 2015, 9:38 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske and Neil Conway.
> 
> 
> Bugs: MESOS-3742
> https://issues.apache.org/jira/browse/MESOS-3742
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   site/source/index.html.md c5b56417ee8ac9aa5ca41da9bc2a2396edcb5f1a 
> 
> Diff: https://reviews.apache.org/r/41468/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41383: Added ubuntu dependence for libsasl2-modules.

[Jojy Varghese]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41383/#review110800
---

Ship it!


Ship It!

- Jojy Varghese


On Dec. 16, 2015, 11:06 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41383/
> ---
> 
> (Updated Dec. 16, 2015, 11:06 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Gilbert Song, Jojy Varghese, and 
> Neil Conway.
> 
> 
> Bugs: MESOS-3305
> https://issues.apache.org/jira/browse/MESOS-3305
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This commit updates the Getting Started guide for ubuntu 14.04 to
> explicitly list libsasl2-modules as a dependence for its system
> requirements. It's possible that the stock ubuntu image comes with this
> package automatically installed (e.g. the default ubuntu/trusty64 image
> on atlas has it pre-installed). Regardless, we need to ensure that
> it is installed before we build mesos, so we list it here.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41383/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41372: Added update for systemd in centos Getting Started.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41372/#review110797
---

Ship it!


Ship It!

- Gilbert Song


On Dec. 16, 2015, 3:07 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41372/
> ---
> 
> (Updated Dec. 16, 2015, 3:07 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Gilbert Song, Jojy Varghese, 
> Joris Van Remoortere, and Neil Conway.
> 
> 
> Bugs: MESOS-3954
> https://issues.apache.org/jira/browse/MESOS-3954
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Several bugs related to older versions of systemd have caused failures
> on centos 7.1 (e.g. MESOS-3296, MESOS-3352, MESOS-3352). We now add
> instructions to explicity update systemd as part of the system
> requirements.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41372/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Review Request 41476: Updated getting started for 0.26 release.

[Kevin Klues]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41476/
---

Review request for mesos, Benjamin Hindman, Artem Harutyunyan, Neil Conway, and 
Till Toenshoff.


Repository: mesos


Description
---

The old release in the documentation was 0.24. Looks like we just jumped
over 0.25...


Diffs
-

  docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 

Diff: https://reviews.apache.org/r/41476/diff/


Testing
---


Thanks,

Kevin Klues

Re: Review Request 39456: Documentation: added containerizer internals

[Jie Yu]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/39456/#review110764
---

Ship it!



docs/containerizer-internals.md (lines 54 - 55)


Do you need to remove the second half of this sentense?


- Jie Yu


On Dec. 16, 2015, 9:48 p.m., Jojy Varghese wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/39456/
> ---
> 
> (Updated Dec. 16, 2015, 9:48 p.m.)
> 
> 
> Review request for mesos, Jie Yu and Timothy Chen.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Documentation: added containerizer internals
> 
> 
> Diffs
> -
> 
>   docs/containerizer-internals.md PRE-CREATION 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/39456/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Jojy Varghese
> 
>

Re: Review Request 41462: libprocess: Used SFINAE-friendly `result_of`.

[Mesos ReviewBot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41462/#review110780
---


Patch looks great!

Reviews applied: [41459, 41460, 41461, 41462]

Passed command: export OS=ubuntu:14.04;export CONFIGURATION="--verbose";export 
COMPILER=gcc; ./support/docker_build.sh

- Mesos ReviewBot


On Dec. 16, 2015, 7:47 p.m., Michael Park wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41462/
> ---
> 
> (Updated Dec. 16, 2015, 7:47 p.m.)
> 
> 
> Review request for mesos, Alex Clemmer and Joris Van Remoortere.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Used the SFINAE `result_of` implemented in 
> [r41461](https://reviews.apache.org/r/41461/).
> 
> Follow-up from [r40114](https://reviews.apache.org/r/40114/).
> 
> 
> Diffs
> -
> 
>   3rdparty/libprocess/include/process/future.hpp 
> c9146e3a3ccf09dd37c5a8ac7000fbe84f3c710c 
> 
> Diff: https://reviews.apache.org/r/41462/diff/
> 
> 
> Testing
> ---
> 
> `make check` on OS X, compiled on Windows.
> 
> 
> Thanks,
> 
> Michael Park
> 
>

Re: Review Request 40431: Move RoleInfo message out of allocator.proto

[Adam B]

> On Nov. 24, 2015, 6:48 p.m., Klaus Ma wrote:
> > include/mesos/role/role.proto, line 38
> > 
> >
> > I'm thinke whether we need to enhance weight to `int`. Any case that we 
> > has to use `double` as weight? If we want to control the scope, TODO is OK.
> 
> Yongqiao Wang wrote:
> In mesos, weight can be any value greater than -1, I think it's make 
> sence. It does not need to change it's type to int if there is no specified 
> requirement.

Current share is divided by weight, then compared against other roles' share, 
so we're already going to be comparing floating point numbers. We don't 
directly compare weights or generate units based on these values, so I don't 
think floating point imprecision is a concern.
Besides, since the default weight is 1.0, adding a new low-priority role with 
lower-than-average weight is impossible without using floating point or 
simultaneously increasing all other weights proportionately. It's valuable to 
be able to set a 0.5 weight.


- Adam


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40431/#review107903
---


On Dec. 7, 2015, 9:20 p.m., Yongqiao Wang wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40431/
> ---
> 
> (Updated Dec. 7, 2015, 9:20 p.m.)
> 
> 
> Review request for mesos, Adam B, Guangya Liu, Qian Zhang, and Jian Qiu.
> 
> 
> Bugs: MESOS-3944
> https://issues.apache.org/jira/browse/MESOS-3944
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Currently role protobuf is defined in allocator.proto due to only the 
> traditional DRF allocator uses roles as it’s first level of hierarchy, I 
> think we should move it out and define it in a separated file as quota had in 
> dynamic roles project, because role protobuf will also be used by master to 
> persist.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 702f56f56c3b1331613cecf26522986f6b572f8c 
>   include/mesos/role/role.hpp PRE-CREATION 
>   include/mesos/role/role.proto PRE-CREATION 
>   src/CMakeLists.txt c0d77c745eb5b12dd6d9d7afaba7e820f8d848ef 
>   src/Makefile.am e96e0ec41e11acff00fbfb3e86427b48a0625bd2 
>   src/master/allocator/mesos/allocator.hpp 
> 97ee80726ad155917811265a983258b0165d3451 
>   src/master/allocator/mesos/hierarchical.hpp 
> 99c742906874c30c39c159e58a65277ade3c07fd 
>   src/master/allocator/mesos/hierarchical.cpp 
> 5da825a1d578a9ee40b4985378fddb3c5fb3b416 
>   src/master/master.hpp 4683fa542a740f9a0b80fff7fbe0e63ec66266f2 
>   src/master/master.cpp 953fa4f14929581b226a7e27d30aea7a5aa1fd7c 
>   src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
>   src/tests/hierarchical_allocator_tests.cpp 
> fb214a829a57529d3f5c49730ae9733f53e622ca 
> 
> Diff: https://reviews.apache.org/r/40431/diff/
> 
> 
> Testing
> ---
> 
> 1. Make Check successfully;
> 
> 2. $ curl http://9.110.48.168:5050/roles
> {"roles":[{"frameworks":[],"name":"*","resources":{"cpus":0,"disk":0,"mem":0},"weight":1.0}]}
> 
> 
> Thanks,
> 
> Yongqiao Wang
> 
>

Re: Review Request 41383: Added ubuntu dependence for libsasl2-modules.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41383/#review110792
---

Ship it!


Totally agreed with ensuring this package installed, because of different 
versions of ubuntu14.04.

- Gilbert Song


On Dec. 16, 2015, 3:06 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41383/
> ---
> 
> (Updated Dec. 16, 2015, 3:06 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske, Gilbert Song, Jojy Varghese, and 
> Neil Conway.
> 
> 
> Bugs: MESOS-3305
> https://issues.apache.org/jira/browse/MESOS-3305
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This commit updates the Getting Started guide for ubuntu 14.04 to
> explicitly list libsasl2-modules as a dependence for its system
> requirements. It's possible that the stock ubuntu image comes with this
> package automatically installed (e.g. the default ubuntu/trusty64 image
> on atlas has it pre-installed). Regardless, we need to ensure that
> it is installed before we build mesos, so we list it here.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41383/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Review Request 41474: Added documentation for `defer` in libprocess README.

[Greg Mann]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41474/
---

Review request for mesos, Benjamin Hindman, Ben Mahler, Joris Van Remoortere, 
and Neil Conway.


Bugs: MESOS-3996
https://issues.apache.org/jira/browse/MESOS-3996


Repository: mesos


Description
---

Added documentation for `defer` in libprocess README.


Diffs
-

  3rdparty/libprocess/README.md c3f309a7dc1c94882c4cc97eeaf0736c2fca0ba5 

Diff: https://reviews.apache.org/r/41474/diff/


Testing
---

Viewed the markdown in a branch on GitHub: 
https://github.com/mesosphere/mesos/tree/defer_docs/3rdparty/libprocess


Thanks,

Greg Mann

Re: Review Request 41360: Aligned the code blocks in Getting Started better.

[Kevin Klues]

> On Dec. 16, 2015, 11:36 p.m., Neil Conway wrote:
> > Looks good! While we're at it, can we update the example version in the 
> > first `wget` command to be 0.26?

I'll do that, but as a separate commit.


- Kevin


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41360/#review110793
---


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41360/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Now each line starts at the beginning of the left hand margin, so that
> multiple lines can be copied without extra space at the beginning of
> each line. This is important for multiline commands (e.g. printing
> multi-line strings to a file) so they can easily be copy and pasted
> without having to modify the lines.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41360/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41446: Fixed character encoding in docs.

[Mesos ReviewBot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41446/#review110803
---


Patch looks great!

Reviews applied: [41446]

Passed command: export OS=ubuntu:14.04;export CONFIGURATION="--verbose";export 
COMPILER=gcc; ./support/docker_build.sh

- Mesos ReviewBot


On Dec. 16, 2015, 8:32 p.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41446/
> ---
> 
> (Updated Dec. 16, 2015, 8:32 p.m.)
> 
> 
> Review request for mesos, Bernd Mathiske and Neil Conway.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   docs/maintenance.md a5831ffa092a9ea6decbe2e640bae637c759a308 
>   docs/networking-for-mesos-managed-containers.md 
> 594fe2b088ac318a979c86795af6e5db5505499e 
>   docs/reconciliation.md 0adfbf1a78c70a12adcbeb3aebd1045452af16ac 
>   docs/scheduler-http-api.md 38d3607edb2145b8525268cd56fe4bfc9917f95a 
>   docs/testing-patterns.md e5fd989cf910640b70da2ac89bf886a61b940aad 
> 
> Diff: https://reviews.apache.org/r/41446/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 40167: [2/7] Added ACL protobuf messages 'CreateVolume' and 'DestroyVolume'.

[Greg Mann]

> On Dec. 16, 2015, 2:35 p.m., Alexander Rukletsov wrote:
> >

Thanks for the reviews, AlexR!!


> On Dec. 16, 2015, 2:35 p.m., Alexander Rukletsov wrote:
> > include/mesos/authorizer/authorizer.proto, line 106
> > 
> >
> > created the volume?

:facepalm:

Thanks AlexR :-)


> On Dec. 16, 2015, 2:35 p.m., Alexander Rukletsov wrote:
> > include/mesos/authorizer/authorizer.proto, lines 100-101
> > 
> >
> > Why did you wrap the comment this way?

This comment is just over 80 characters long, and by wrapping this way I reduce 
the overall "jaggedness". I find this a bit easier to read than a two-line 
comment that has just one or two words on the second line.


- Greg


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40167/#review110641
---


On Dec. 17, 2015, 12:03 a.m., Greg Mann wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40167/
> ---
> 
> (Updated Dec. 17, 2015, 12:03 a.m.)
> 
> 
> Review request for mesos, Jie Yu, Michael Park, and Neil Conway.
> 
> 
> Bugs: MESOS-4178
> https://issues.apache.org/jira/browse/MESOS-4178
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Added ACL protobuf messages 'CreateVolume' and 'DestroyVolume'.
> 
> 
> Diffs
> -
> 
>   include/mesos/authorizer/authorizer.proto 
> 74fcc86d3c92cb3aa27e45b647b1653705b3201c 
> 
> Diff: https://reviews.apache.org/r/40167/diff/
> 
> 
> Testing
> ---
> 
> This is the second in a chain of 7 patches. `make check` was used to test 
> after all patches were applied.
> 
> Note that this chain of patches touches many of the same files as another 
> chain beginning with Review #39985 and ending with Review #39989, which is 
> currently in review as well. To avoid conflicts, the beginning of this chain 
> begins on top of Review #39989.
> 
> 
> Thanks,
> 
> Greg Mann
> 
>

Re: Review Request 40431: Move RoleInfo message out of allocator.proto

[Adam B]

> On Dec. 10, 2015, 2:40 a.m., Adam B wrote:
> > include/mesos/master/allocator.proto, line 19
> > 
> >
> > Shouldn't this file have `java_package` and `java_outer_classname` just 
> > like the other protos?
> > Looks like isolator.proto and oversubscription.proto are missing it 
> > too. Would you mind creating a separate patch to fix that?
> 
> Yongqiao Wang wrote:
> I am not sure if we need to add java_package and java_outer_classname in 
> those proto files, can you please clarify a little more about why we need to 
> do this?

Nevermind. That's only necessary for the scheduler/executor API protobufs, 
since they may need to be consumed by Java processes.
https://developers.google.com/protocol-buffers/docs/javatutorial
Dropping the issue.


> On Dec. 10, 2015, 2:40 a.m., Adam B wrote:
> > include/mesos/role/role.proto, line 21
> > 
> >
> > Why change the package? Couldn't this still be in `mesos.master`? Then 
> > you wouldn't have to change all the other files.
> 
> Yongqiao Wang wrote:
> Like other feature(such as quota), I also think role manamgnet is a 
> seprated function, so I define role protobuf in a separated package rather 
> than define it in mesos.proto.

Ok, I just thought you could reduce code churn by keeping the package name the 
same, even if you create a new proto file.
This 'role' proto package is still only useful on the master, right? 
mesos.scheduler.role and mesos.agent.role would have different messages/fields.
But this is moot after the implicit roles changes in 
https://reviews.apache.org/r/41075/ which removes the original RoleInfo.


- Adam


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40431/#review109704
---


On Dec. 7, 2015, 9:20 p.m., Yongqiao Wang wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40431/
> ---
> 
> (Updated Dec. 7, 2015, 9:20 p.m.)
> 
> 
> Review request for mesos, Adam B, Guangya Liu, Qian Zhang, and Jian Qiu.
> 
> 
> Bugs: MESOS-3944
> https://issues.apache.org/jira/browse/MESOS-3944
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Currently role protobuf is defined in allocator.proto due to only the 
> traditional DRF allocator uses roles as it’s first level of hierarchy, I 
> think we should move it out and define it in a separated file as quota had in 
> dynamic roles project, because role protobuf will also be used by master to 
> persist.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 702f56f56c3b1331613cecf26522986f6b572f8c 
>   include/mesos/role/role.hpp PRE-CREATION 
>   include/mesos/role/role.proto PRE-CREATION 
>   src/CMakeLists.txt c0d77c745eb5b12dd6d9d7afaba7e820f8d848ef 
>   src/Makefile.am e96e0ec41e11acff00fbfb3e86427b48a0625bd2 
>   src/master/allocator/mesos/allocator.hpp 
> 97ee80726ad155917811265a983258b0165d3451 
>   src/master/allocator/mesos/hierarchical.hpp 
> 99c742906874c30c39c159e58a65277ade3c07fd 
>   src/master/allocator/mesos/hierarchical.cpp 
> 5da825a1d578a9ee40b4985378fddb3c5fb3b416 
>   src/master/master.hpp 4683fa542a740f9a0b80fff7fbe0e63ec66266f2 
>   src/master/master.cpp 953fa4f14929581b226a7e27d30aea7a5aa1fd7c 
>   src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
>   src/tests/hierarchical_allocator_tests.cpp 
> fb214a829a57529d3f5c49730ae9733f53e622ca 
> 
> Diff: https://reviews.apache.org/r/40431/diff/
> 
> 
> Testing
> ---
> 
> 1. Make Check successfully;
> 
> 2. $ curl http://9.110.48.168:5050/roles
> {"roles":[{"frameworks":[],"name":"*","resources":{"cpus":0,"disk":0,"mem":0},"weight":1.0}]}
> 
> 
> Thanks,
> 
> Yongqiao Wang
> 
>

Re: Review Request 41188: Providing JSON bindings to so that mesos modules can expose DiscoveryInfo protobuf messages to HTTP endpoints

[Anand Mazumdar]

> On Dec. 17, 2015, 12:12 a.m., Anand Mazumdar wrote:
> > src/tests/common/http_tests.cpp, lines 130-140
> > 
> >
> > Can you confirm that other objects that have labels that are exposed 
> > via `/state` endpoint also have a nested `labels` mapping ?
> > 
> > AFAICT, this is an artifact of us using the Protobuf to JSON converters 
> > when we should have just used the `model(...)` functions.
> 
> Avinash sridharan wrote:
> I am trying to understand this comment. Is it that applications don't 
> expect nested labels (for ports) when they read state.json? 
> 
> In terms of exposing labels in state.json I can see from 
> TEST_F(SlaveTest, TaskLabels) (slave_tests.cpp) the labels are nested within 
> tasks. Since, DiscoveryInfo also has labels and Port has labels unless we 
> nest them, the relationship would not be explicit if we do not replicate the 
> protobuf right?

Let me give you an example:

Here is how the Task Labels JSON returned from this test looks like:

```
"tasks": [
{
  "executor_id": "default",
  "framework_id": "5452180f-fa83-4d25-802f-8a0c974dfd21-",
  "id": "1",
  "labels": [
{
  "key": "foo",
  "value": "bar"
},
{
  "key": "bar",
  "value": "baz"
},
{
  "key": "bar",
  "value": "qux"
}
  ],
  "name": "",
  "resources": {
"cpus": 2,
"disk": 1024,
"mem": 1024,
"ports": "[31000-32000]"
  },
```

If you notice, it does not have a "labels" field nested inside a "labels" field 
as we have when we do automated protobuf to JSON conversions. Do we want to 
preserve this behavior ?


- Anand


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41188/#review110804
---


On Dec. 15, 2015, 9:13 p.m., Avinash sridharan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41188/
> ---
> 
> (Updated Dec. 15, 2015, 9:13 p.m.)
> 
> 
> Review request for mesos, Adam B and Anand Mazumdar.
> 
> 
> Bugs: MESOS-3962
> https://issues.apache.org/jira/browse/MESOS-3962
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Providing JSON bindings to so that mesos modules can expose DiscoveryInfo 
> protobuf messages to HTTP endpoints
> 
> 
> Diffs
> -
> 
>   src/slave/http.cpp d1b1158c0a80d72c32c9b28977043b6be2295239 
>   src/tests/common/http_tests.cpp 3aca5b0437a012664f58ff331cc7cf682d442699 
>   src/tests/slave_tests.cpp 4975bea8a7a701e0414426760692720f73dea7f5 
> 
> Diff: https://reviews.apache.org/r/41188/diff/
> 
> 
> Testing
> ---
> 
> make check.
> Added Unit tests to verify setting of DiscoveryInfo in state.json in slave. 
> Also added Unit test to test that DiscoveryInfo gets exposed in master when 
> TaskInfo protobuf is converted to JSON objects.
> 
> 
> Thanks,
> 
> Avinash sridharan
> 
>

Re: Review Request 40169: [4/7] Added 'Master::authorize{Destroy, Create}Volume' to create/destroy persistent volumes.

[Greg Mann]

> On Dec. 16, 2015, 2:36 p.m., Alexander Rukletsov wrote:
> > src/master/master.cpp, lines 2870-2871
> > 
> >
> > Looks like a copy-paste-induced typo slipped in : ), I believe you mean 
> > "volume is created".

:double facepalm:

Thanks!! :-D


- Greg


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40169/#review110656
---


On Dec. 17, 2015, 12:29 a.m., Greg Mann wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40169/
> ---
> 
> (Updated Dec. 17, 2015, 12:29 a.m.)
> 
> 
> Review request for mesos, Jie Yu, Michael Park, and Neil Conway.
> 
> 
> Bugs: MESOS-4179
> https://issues.apache.org/jira/browse/MESOS-4179
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Added 'Master::authorize{Destroy,Create}Volume' to create/destroy persistent 
> volumes.
> 
> 
> Diffs
> -
> 
>   src/master/master.hpp 9aa548aa6e159046c94e4ec96f631ea8b3bfd5d8 
>   src/master/master.cpp 0d1482279c68f2a4a27dabaf28774769a5d515c4 
> 
> Diff: https://reviews.apache.org/r/40169/diff/
> 
> 
> Testing
> ---
> 
> This is the fourth in a chain of 7 patches. `make check` was used to test 
> after all patches were applied.
> 
> 
> Thanks,
> 
> Greg Mann
> 
>

Re: Review Request 41372: Added update for systemd in centos Getting Started.

[Kevin Klues]

> On Dec. 17, 2015, 12:09 a.m., Jojy Varghese wrote:
> > docs/getting-started.md, line 146
> > 
> >
> > I would emphaise versions with backtics.

This doesn't appear to add any special empahsis (in the markdown that is), 
because it is embedded in a code block.  I can put ticks around it, but I'd 
rather use a standard apostrophe like what is used elsewhere in the comments.


- Kevin


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41372/#review110805
---


On Dec. 16, 2015, 11:07 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41372/
> ---
> 
> (Updated Dec. 16, 2015, 11:07 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Gilbert Song, Jojy Varghese, 
> Joris Van Remoortere, and Neil Conway.
> 
> 
> Bugs: MESOS-3954
> https://issues.apache.org/jira/browse/MESOS-3954
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Several bugs related to older versions of systemd have caused failures
> on centos 7.1 (e.g. MESOS-3296, MESOS-3352, MESOS-3352). We now add
> instructions to explicity update systemd as part of the system
> requirements.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41372/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41003: Logger Module: Add the SandboxContainerLogger, the default ContainerLogger implementation.

[Benjamin Hindman]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41003/#review110827
---

Ship it!



src/slave/container_loggers/sandbox.hpp (line 44)


Can we leave a comment of what this provides? And that it's the default?



src/slave/container_loggers/sandbox.cpp (line 55)


Is this comment useful? Is there something else here you want to capture?



src/slave/container_loggers/sandbox.cpp (line 72)


Can we put a newline before this? Or:

info.environment = environment.get({});


- Benjamin Hindman


On Dec. 15, 2015, 8:39 p.m., Joseph Wu wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41003/
> ---
> 
> (Updated Dec. 15, 2015, 8:39 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Cody Maloney, and Artem 
> Harutyunyan.
> 
> 
> Bugs: MESOS-4087
> https://issues.apache.org/jira/browse/MESOS-4087
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This implementation mirrors how executor/task stdout/stderr is currently 
> saved to plain files.
> 
> 
> Diffs
> -
> 
>   src/slave/container_loggers/sandbox.hpp PRE-CREATION 
>   src/slave/container_loggers/sandbox.cpp PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/41003/diff/
> 
> 
> Testing
> ---
> 
> This is added to the Makefile later in the review chain.
> 
> 
> Thanks,
> 
> Joseph Wu
> 
>

Re: Review Request 41468: [site] Removed MesosCon Europe from index page.

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41468/
---

(Updated Dec. 17, 2015, 1:22 a.m.)


Review request for mesos, Bernd Mathiske and Neil Conway.


Changes
---

Updated to correct layout.


Bugs: MESOS-3742
https://issues.apache.org/jira/browse/MESOS-3742


Repository: mesos


Description
---

see summary.


Diffs (updated)
-

  site/source/index.html.md c5b56417ee8ac9aa5ca41da9bc2a2396edcb5f1a 

Diff: https://reviews.apache.org/r/41468/diff/


Testing (updated)
---

mesos-site rake.


Thanks,

Till Toenshoff

Re: Review Request 40431: Move RoleInfo message out of allocator.proto

[Yongqiao Wang]

> On Dec. 10, 2015, 10:40 a.m., Adam B wrote:
> > include/mesos/master/allocator.proto, line 19
> > 
> >
> > Shouldn't this file have `java_package` and `java_outer_classname` just 
> > like the other protos?
> > Looks like isolator.proto and oversubscription.proto are missing it 
> > too. Would you mind creating a separate patch to fix that?
> 
> Yongqiao Wang wrote:
> I am not sure if we need to add java_package and java_outer_classname in 
> those proto files, can you please clarify a little more about why we need to 
> do this?
> 
> Adam B wrote:
> Nevermind. That's only necessary for the scheduler/executor API 
> protobufs, since they may need to be consumed by Java processes.
> https://developers.google.com/protocol-buffers/docs/javatutorial
> Dropping the issue.

So can I log another JIRA ticket to fix this issue?


> On Dec. 10, 2015, 10:40 a.m., Adam B wrote:
> > include/mesos/role/role.proto, line 21
> > 
> >
> > Why change the package? Couldn't this still be in `mesos.master`? Then 
> > you wouldn't have to change all the other files.
> 
> Yongqiao Wang wrote:
> Like other feature(such as quota), I also think role manamgnet is a 
> seprated function, so I define role protobuf in a separated package rather 
> than define it in mesos.proto.
> 
> Adam B wrote:
> Ok, I just thought you could reduce code churn by keeping the package 
> name the same, even if you create a new proto file.
> This 'role' proto package is still only useful on the master, right? 
> mesos.scheduler.role and mesos.agent.role would have different 
> messages/fields.
> But this is moot after the implicit roles changes in 
> https://reviews.apache.org/r/41075/ which removes the original RoleInfo.

OK, we can defer this patch until implicit roles chagnes finalize.


- Yongqiao


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40431/#review109704
---


On Dec. 8, 2015, 5:20 a.m., Yongqiao Wang wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40431/
> ---
> 
> (Updated Dec. 8, 2015, 5:20 a.m.)
> 
> 
> Review request for mesos, Adam B, Guangya Liu, Qian Zhang, and Jian Qiu.
> 
> 
> Bugs: MESOS-3944
> https://issues.apache.org/jira/browse/MESOS-3944
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Currently role protobuf is defined in allocator.proto due to only the 
> traditional DRF allocator uses roles as it’s first level of hierarchy, I 
> think we should move it out and define it in a separated file as quota had in 
> dynamic roles project, because role protobuf will also be used by master to 
> persist.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 702f56f56c3b1331613cecf26522986f6b572f8c 
>   include/mesos/role/role.hpp PRE-CREATION 
>   include/mesos/role/role.proto PRE-CREATION 
>   src/CMakeLists.txt c0d77c745eb5b12dd6d9d7afaba7e820f8d848ef 
>   src/Makefile.am e96e0ec41e11acff00fbfb3e86427b48a0625bd2 
>   src/master/allocator/mesos/allocator.hpp 
> 97ee80726ad155917811265a983258b0165d3451 
>   src/master/allocator/mesos/hierarchical.hpp 
> 99c742906874c30c39c159e58a65277ade3c07fd 
>   src/master/allocator/mesos/hierarchical.cpp 
> 5da825a1d578a9ee40b4985378fddb3c5fb3b416 
>   src/master/master.hpp 4683fa542a740f9a0b80fff7fbe0e63ec66266f2 
>   src/master/master.cpp 953fa4f14929581b226a7e27d30aea7a5aa1fd7c 
>   src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
>   src/tests/hierarchical_allocator_tests.cpp 
> fb214a829a57529d3f5c49730ae9733f53e622ca 
> 
> Diff: https://reviews.apache.org/r/40431/diff/
> 
> 
> Testing
> ---
> 
> 1. Make Check successfully;
> 
> 2. $ curl http://9.110.48.168:5050/roles
> {"roles":[{"frameworks":[],"name":"*","resources":{"cpus":0,"disk":0,"mem":0},"weight":1.0}]}
> 
> 
> Thanks,
> 
> Yongqiao Wang
> 
>

Re: Review Request 41364: Added NOTE to Examples in Getting Started on prod.

[Till Toenshoff]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41364/#review110845
---

Ship it!


Ship It!

- Till Toenshoff


On Dec. 17, 2015, 1:26 a.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41364/
> ---
> 
> (Updated Dec. 17, 2015, 1:26 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Bugs: MESOS-3844
> https://issues.apache.org/jira/browse/MESOS-3844
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The current instructions assume you are launching mesos on your local
> machine. This could be confusing for someone trying to run mesos on
> (e.g. AWS) and not understanding why they can't access the Mesos
> web page. This note adds a comment to try and hinder such confusion.
> 
> Review: https://reviews.apache.org/r/41364
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41364/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41011: Unified Container: Implemented interface for runtime configure to be returned from provisioner to mesos containerizer.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41011/
---

(Updated Dec. 16, 2015, 6:38 p.m.)


Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-2980
https://issues.apache.org/jira/browse/MESOS-2980


Repository: mesos


Description
---

Unified Container: Implemented interface for runtime configure to be returned 
from provisioner to mesos containerizer.


Diffs (updated)
-

  src/slave/containerizer/mesos/isolators/filesystem/linux.hpp 
b8d47e8250a892fa333a0a966a0f38fe1f2816f2 
  src/slave/containerizer/mesos/isolators/filesystem/linux.cpp 
8442e9c30612fa04f34130b9f967cb1414880ca6 
  src/slave/containerizer/mesos/provisioner/provisioner.hpp 
b5d40b9428c9aa2f397540ef275cca7a7a335e43 
  src/slave/containerizer/mesos/provisioner/provisioner.cpp 
8a2c9bbf7e01a62ad92ba19216b32f6e813cd736 
  src/slave/containerizer/mesos/provisioner/store.hpp 
8d1493856420dee3210af79b628c8c770c5c8550 
  src/tests/containerizer/provisioner_appc_tests.cpp 
9c5d91ba1fd49602418a0dd7c26b3132d3ac1a68 

Diff: https://reviews.apache.org/r/41011/diff/


Testing
---

make check (ubuntu14.04 + clang-3.6)


Thanks,

Gilbert Song

Re: Review Request 41194: Unified Container: Cleanup uneccessary JSON parse in local puller.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41194/
---

(Updated Dec. 16, 2015, 6:38 p.m.)


Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-2980
https://issues.apache.org/jira/browse/MESOS-2980


Repository: mesos


Description
---

Unified Container: Cleanup uneccessary JSON parse in local puller.


Diffs (updated)
-

  src/slave/containerizer/mesos/provisioner/docker/local_puller.cpp 
f8856664b39de0a07a3c3d7de8e03658acba9631 

Diff: https://reviews.apache.org/r/41194/diff/


Testing
---

make check(ubuntu14.04 + clang-3.6)


Thanks,

Gilbert Song

Review Request 41487: Provisioner: Changed docker v2 manifest naming.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41487/
---

Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-4186
https://issues.apache.org/jira/browse/MESOS-4186


Repository: mesos


Description
---

Provisioner: Changed docker v2 manifest naming.


Diffs
-

  src/slave/containerizer/mesos/provisioner/docker/message.proto 
5c032701671b275d86c6d9276791a46df814396c 
  src/slave/containerizer/mesos/provisioner/docker/registry_client.hpp 
9b02b6ff6dc5c6e8aabdc4ac0aa4df337764ef30 
  src/slave/containerizer/mesos/provisioner/docker/registry_client.cpp 
e69bab43d5f5359cfd8eb4cb7c5ad4a1d22c4e05 
  src/slave/containerizer/mesos/provisioner/docker/registry_puller.cpp 
89f61c20e52e5eff8d8e92748f03b3b461516cd2 
  src/slave/containerizer/mesos/provisioner/docker/spec.hpp 
e674b6f2e705f8b3e49770560eeab4c127473f94 
  src/slave/containerizer/mesos/provisioner/docker/spec.cpp 
1f05c75dc9473bd5e4c0d3f74fa0ef996b96a84e 
  src/tests/containerizer/provisioner_docker_tests.cpp 
3f1717b770e139c3759aab0aeda9dbcf5029b0c2 

Diff: https://reviews.apache.org/r/41487/diff/


Testing
---

make check


Thanks,

Gilbert Song

Review Request 41491: Unified Container: Implemented passing entrypoint in runtime config.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41491/
---

Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-2980
https://issues.apache.org/jira/browse/MESOS-2980


Repository: mesos


Description
---

Unified Container: Implemented passing entrypoint in runtime config.


Diffs
-

  src/slave/containerizer/mesos/provisioner/docker/store.cpp 
1ad7b67a94b1d9367afcb7c30a6d01fdf6b8ab6c 
  src/slave/containerizer/mesos/provisioner/store.hpp 
8d1493856420dee3210af79b628c8c770c5c8550 
  src/tests/containerizer/provisioner_docker_tests.cpp 
3f1717b770e139c3759aab0aeda9dbcf5029b0c2 

Diff: https://reviews.apache.org/r/41491/diff/


Testing
---

make check


Thanks,

Gilbert Song

Review Request 41489: Provisioner: Implemented docker v1 parse serialization method in spec.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41489/
---

Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-4186
https://issues.apache.org/jira/browse/MESOS-4186


Repository: mesos


Description
---

Provisioner: Implemented docker v1 parse serialization method in spec.


Diffs
-

  src/slave/containerizer/mesos/provisioner/docker/registry_client.cpp 
e69bab43d5f5359cfd8eb4cb7c5ad4a1d22c4e05 
  src/slave/containerizer/mesos/provisioner/docker/spec.hpp 
e674b6f2e705f8b3e49770560eeab4c127473f94 
  src/slave/containerizer/mesos/provisioner/docker/spec.cpp 
1f05c75dc9473bd5e4c0d3f74fa0ef996b96a84e 
  src/tests/containerizer/provisioner_docker_tests.cpp 
3f1717b770e139c3759aab0aeda9dbcf5029b0c2 

Diff: https://reviews.apache.org/r/41489/diff/


Testing
---

make check


Thanks,

Gilbert Song

Review Request 41490: Provisioner: Added test case for docker v1 manifest serialization.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41490/
---

Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-4186
https://issues.apache.org/jira/browse/MESOS-4186


Repository: mesos


Description
---

Provisioner: Added test case for docker v1 manifest serialization.


Diffs
-

  src/tests/containerizer/provisioner_docker_tests.cpp 
3f1717b770e139c3759aab0aeda9dbcf5029b0c2 

Diff: https://reviews.apache.org/r/41490/diff/


Testing
---

make check


Thanks,

Gilbert Song

Re: Review Request 41407: Unified Container: Save all docker image information on disk instead of rootfs only.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41407/
---

(Updated Dec. 16, 2015, 6:38 p.m.)


Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-2980
https://issues.apache.org/jira/browse/MESOS-2980


Repository: mesos


Description
---

Unified Container: Save all docker image information on disk instead of rootfs 
only.


Diffs (updated)
-

  src/slave/containerizer/mesos/provisioner/docker/puller.cpp 
4aa4a9c4074d96c30c3bceea59d071feeecae2ea 
  src/slave/containerizer/mesos/provisioner/docker/store.cpp 
1ad7b67a94b1d9367afcb7c30a6d01fdf6b8ab6c 
  src/tests/containerizer/provisioner_docker_tests.cpp 
3f1717b770e139c3759aab0aeda9dbcf5029b0c2 

Diff: https://reviews.apache.org/r/41407/diff/


Testing
---

make check


Thanks,

Gilbert Song

Re: Review Request 41406: Unified Container: Added path helper method to get image layer manifest on local store.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41406/
---

(Updated Dec. 16, 2015, 6:38 p.m.)


Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-2980
https://issues.apache.org/jira/browse/MESOS-2980


Repository: mesos


Description
---

Unified Container: Added path helper method to get image layer manifest on 
local store.


Diffs (updated)
-

  src/slave/containerizer/mesos/provisioner/docker/paths.hpp 
7e5122d956c82e6cbd62fc217158205ea05f4290 
  src/slave/containerizer/mesos/provisioner/docker/paths.cpp 
9c1f6a19c59b12f58b0550d022f9128e7015a4e9 

Diff: https://reviews.apache.org/r/41406/diff/


Testing
---

make check


Thanks,

Gilbert Song

Review Request 41488: Provisioner: Added docker v1 manifest protobuf message.

[Gilbert Song]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41488/
---

Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.


Bugs: MESOS-4186
https://issues.apache.org/jira/browse/MESOS-4186


Repository: mesos


Description
---

Provisioner: Added docker v1 manifest protobuf message.


Diffs
-

  src/slave/containerizer/mesos/provisioner/docker/message.proto 
5c032701671b275d86c6d9276791a46df814396c 

Diff: https://reviews.apache.org/r/41488/diff/


Testing
---

make check


Thanks,

Gilbert Song

Re: Review Request 41361: Wrote cmd to create WANdiscoSVN file for centos.

[Till Toenshoff]

> On Dec. 16, 2015, 11:22 p.m., Neil Conway wrote:
> > Seems fine as far as it goes, but it would be great to find a more standard 
> > source for the packages than WANdisco. For example, 'devtoolset-3' appears 
> > to have what we need.

Yes, overall it seems we need to revisit this entire guide again to update / 
streamline things -- I also believe there is a lot of room for improvement on 
e.g. the Centos7 part.


- Till


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41361/#review110787
---


On Dec. 16, 2015, 11:09 p.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41361/
> ---
> 
> (Updated Dec. 16, 2015, 11:09 p.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Previously we just told users to create this file and then add the
> contents to it. Now they can just copy and paste a single command to
> create the file and fill the contents automatically.
> 
> 
> Diffs
> -
> 
>   docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 
> 
> Diff: https://reviews.apache.org/r/41361/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41362: Added note in docs for installing wget, tar, git.

[Kevin Klues]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41362/
---

(Updated Dec. 17, 2015, 2:44 a.m.)


Review request for mesos, Benjamin Hindman, Joseph Wu, and Neil Conway.


Bugs: MESOS-3844
https://issues.apache.org/jira/browse/MESOS-3844


Repository: mesos


Description (updated)
---

The 'Downloading Mesos' section of the getting started guide
assumes that you already have the 'wget', 'tar', and 'git' utilities
installed on your system before running through the 'System
Requirements' section below it. This commit adds a note to first run
through the 'System Requirments' section if problems are encountered
with downloading mesos via the provided instructions.

It also adds the 'wget', 'tar', and 'git' utilities to the list of
requirements for all systems in the 'System Requirements' section.
The two CentOS sections already had instructions for 'wget' and 'tar',
but now 'git' has been added to the list.

Review: https://reviews.apache.org/r/41362


Diffs (updated)
-

  docs/getting-started.md 0225c9db985261b9c3057b2b376f29b1c80784fb 

Diff: https://reviews.apache.org/r/41362/diff/


Testing
---


Thanks,

Kevin Klues

Re: Review Request 41075: Added support for implicit roles.

[Neil Conway]

> On Dec. 17, 2015, 2:54 a.m., Yongqiao Wang wrote:
> > src/master/http.cpp, line 1543
> > 
> >
> > I am not sure /roles will return the reserved resources of role, as far 
> > as I know, it will return the used resources of each registered framework.

Ah, right -- I called this "total allocated resources" (since it includes both 
offered and used).


- Neil


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41075/#review110867
---


On Dec. 17, 2015, 2:01 a.m., Neil Conway wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41075/
> ---
> 
> (Updated Dec. 17, 2015, 2:01 a.m.)
> 
> 
> Review request for mesos, Adam B, Alexander Rukletsov, Benjamin Hindman, Greg 
> Mann, and Yongqiao Wang.
> 
> 
> Bugs: MESOS-4085
> https://issues.apache.org/jira/browse/MESOS-4085
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Changed the behavior of the master when the `--roles` flag is NOT
> specified. Previously, this would allow only the `*` role to be used. Now,
> omitting `--roles` means that any role can be used. This is called "implicit
> roles". Configuring which principals can perform operations as which roles
> should be done using ACLs in the authorization system.
> 
> Note that this changes the behavior of the system when `--roles` is not
> specified. This is likely acceptable: if the operator didn't specify `--roles`
> in prior versions of Mesos, they were likely not using roles or authorization 
> at
> that time.
> 
> Another minor behavioral change is that the "/roles" endpoint will now only
> return results for currently "active" roles (those with one or more registered
> frameworks).
> 
> The `--roles` flag is now considered deprecated and will be removed in a 
> future
> version of Mesos.
> 
> 
> Diffs
> -
> 
>   include/mesos/master/allocator.hpp 619ef01c3a7d640560653cfc1838dd09046d1da0 
>   include/mesos/master/allocator.proto 
> 702f56f56c3b1331613cecf26522986f6b572f8c 
>   src/master/allocator/mesos/allocator.hpp 
> 97ee80726ad155917811265a983258b0165d3451 
>   src/master/allocator/mesos/hierarchical.hpp 
> 99c742906874c30c39c159e58a65277ade3c07fd 
>   src/master/allocator/mesos/hierarchical.cpp 
> 5da825a1d578a9ee40b4985378fddb3c5fb3b416 
>   src/master/http.cpp 8dfd67f7d51b8395953d6beb77b2d71bc538eacd 
>   src/master/master.hpp 9aa548aa6e159046c94e4ec96f631ea8b3bfd5d8 
>   src/master/master.cpp 0d1482279c68f2a4a27dabaf28774769a5d515c4 
>   src/master/quota_handler.cpp 11167879b2480d9c8dd6398ca39c479089ec2272 
>   src/tests/allocator.hpp c7670525765491fe931a4ee38446fa7e9d79af42 
>   src/tests/hierarchical_allocator_tests.cpp 
> e239b4746494fcc2b362a83afb634a2ce5e25f9b 
>   src/tests/master_quota_tests.cpp 0473869783a714766ed26fff61d7f8c56342df74 
> 
> Diff: https://reviews.apache.org/r/41075/diff/
> 
> 
> Testing
> ---
> 
> "make check" on OSX 10.10 and Ubuntu 15.10; `--gtest_repeat=1000` for the 
> more likely role-related tests.
> 
> TODOs:
> 
> * Add tests for allocation behavior for weights + implicit roles
> * More tests for quota + implicit roles?
> 
> Notes:
> 
> * There's two places where we use manual `new`/`delete` where a `unique_ptr` 
> would probably be nicer. I'm inclined to leave this as-is for now though 
> (making use of unique_ptr is a broader issue).
> 
> 
> Thanks,
> 
> Neil Conway
> 
>

Re: Review Request 41454: Added initial draft of executor HTTP API user doc.

[Mesos ReviewBot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41454/#review110876
---


Patch looks great!

Reviews applied: [41454]

Passed command: export OS=ubuntu:14.04;export CONFIGURATION="--verbose";export 
COMPILER=gcc; ./support/docker_build.sh

- Mesos ReviewBot


On Dec. 17, 2015, 1:07 a.m., Anand Mazumdar wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41454/
> ---
> 
> (Updated Dec. 17, 2015, 1:07 a.m.)
> 
> 
> Review request for mesos and Vinod Kone.
> 
> 
> Bugs: MESOS-4177
> https://issues.apache.org/jira/browse/MESOS-4177
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This is an initial draft for the user doc of Executor HTTP API. Some details 
> might change due to the pending executor library review chain: 
> https://reviews.apache.org/r/41275
> 
> 
> Diffs
> -
> 
>   docs/executor-http-api.md PRE-CREATION 
>   docs/home.md a01612ec8347eb8a7a9277b829365b6c3a1fe9e3 
> 
> Diff: https://reviews.apache.org/r/41454/diff/
> 
> 
> Testing
> ---
> 
> https://gist.github.com/hatred/d35fdaa667203bf34a78
> 
> 
> Thanks,
> 
> Anand Mazumdar
> 
>

Re: Review Request 41011: Unified Container: Implemented interface for runtime configure to be returned from provisioner to mesos containerizer.

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41011/#review110875
---

Ship it!


Ship It!

- Timothy Chen


On Dec. 17, 2015, 2:38 a.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41011/
> ---
> 
> (Updated Dec. 17, 2015, 2:38 a.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Implemented interface for runtime configure to be returned 
> from provisioner to mesos containerizer.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/isolators/filesystem/linux.hpp 
> b8d47e8250a892fa333a0a966a0f38fe1f2816f2 
>   src/slave/containerizer/mesos/isolators/filesystem/linux.cpp 
> 8442e9c30612fa04f34130b9f967cb1414880ca6 
>   src/slave/containerizer/mesos/provisioner/provisioner.hpp 
> b5d40b9428c9aa2f397540ef275cca7a7a335e43 
>   src/slave/containerizer/mesos/provisioner/provisioner.cpp 
> 8a2c9bbf7e01a62ad92ba19216b32f6e813cd736 
>   src/slave/containerizer/mesos/provisioner/store.hpp 
> 8d1493856420dee3210af79b628c8c770c5c8550 
>   src/tests/containerizer/provisioner_appc_tests.cpp 
> 9c5d91ba1fd49602418a0dd7c26b3132d3ac1a68 
> 
> Diff: https://reviews.apache.org/r/41011/diff/
> 
> 
> Testing
> ---
> 
> make check (ubuntu14.04 + clang-3.6)
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 41188: Providing JSON bindings to so that mesos modules can expose DiscoveryInfo protobuf messages to HTTP endpoints

[Avinash sridharan]

> On Dec. 17, 2015, 12:12 a.m., Anand Mazumdar wrote:
> > src/tests/common/http_tests.cpp, lines 130-140
> > 
> >
> > Can you confirm that other objects that have labels that are exposed 
> > via `/state` endpoint also have a nested `labels` mapping ?
> > 
> > AFAICT, this is an artifact of us using the Protobuf to JSON converters 
> > when we should have just used the `model(...)` functions.
> 
> Avinash sridharan wrote:
> I am trying to understand this comment. Is it that applications don't 
> expect nested labels (for ports) when they read state.json? 
> 
> In terms of exposing labels in state.json I can see from 
> TEST_F(SlaveTest, TaskLabels) (slave_tests.cpp) the labels are nested within 
> tasks. Since, DiscoveryInfo also has labels and Port has labels unless we 
> nest them, the relationship would not be explicit if we do not replicate the 
> protobuf right?
> 
> Anand Mazumdar wrote:
> Let me give you an example:
> 
> Here is how the Task Labels JSON returned from this test looks like:
> 
> ```
> "tasks": [
> {
>   "executor_id": "default",
>   "framework_id": "5452180f-fa83-4d25-802f-8a0c974dfd21-",
>   "id": "1",
>   "labels": [
> {
>   "key": "foo",
>   "value": "bar"
> },
> {
>   "key": "bar",
>   "value": "baz"
> },
> {
>   "key": "bar",
>   "value": "qux"
> }
>   ],
>   "name": "",
>   "resources": {
> "cpus": 2,
> "disk": 1024,
> "mem": 1024,
> "ports": "[31000-32000]"
>   },
> ```
> 
> If you notice, it does not have a "labels" field nested inside a "labels" 
> field as we have when we do automated protobuf to JSON conversions. Do we 
> want to preserve this behavior ?

I get your point . You are right this is an artifact of using the 
JSON::protobuf compred to explicit model functions . 

Is this going to be an issue? Since the dependency is reflected in the protobuf 
definition.
One reason I can of think of keeping it as is, is that the protobuf is a 
contract between the framework and service discovery, so they might actually 
expect this.


- Avinash


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41188/#review110804
---


On Dec. 17, 2015, 3:10 a.m., Avinash sridharan wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41188/
> ---
> 
> (Updated Dec. 17, 2015, 3:10 a.m.)
> 
> 
> Review request for mesos, Adam B and Anand Mazumdar.
> 
> 
> Bugs: MESOS-3962
> https://issues.apache.org/jira/browse/MESOS-3962
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Providing JSON bindings to so that mesos modules can expose DiscoveryInfo 
> protobuf messages to HTTP endpoints
> 
> 
> Diffs
> -
> 
>   src/slave/http.cpp d1b1158c0a80d72c32c9b28977043b6be2295239 
>   src/tests/common/http_tests.cpp 3aca5b0437a012664f58ff331cc7cf682d442699 
>   src/tests/slave_tests.cpp 4975bea8a7a701e0414426760692720f73dea7f5 
> 
> Diff: https://reviews.apache.org/r/41188/diff/
> 
> 
> Testing
> ---
> 
> make check.
> Added Unit tests to verify setting of DiscoveryInfo in state.json in slave. 
> Also added Unit test to test that DiscoveryInfo gets exposed in master when 
> TaskInfo protobuf is converted to JSON objects.
> 
> 
> Thanks,
> 
> Avinash sridharan
> 
>

Re: Review Request 41485: [site] Updated downloads-page for 0.26.0.

[Till Toenshoff]

> On Dec. 17, 2015, 2:54 a.m., Kevin Klues wrote:
> > site/source/downloads.html.md, line 75
> > 
> >
> > Is this link supposed to trigger a download?  It just seems to take me 
> > to a mirrors site.

Appears to be an upstream problem - other projects use the same scheme for 
getting a dynamically assigned mirror in a link. e.g. 
http://www.apache.org/dyn/mirrors/mirrors.cgi/ws/axis2/tools/1_2/axis2-eclipse-service-archiver-wizard-1.2.zip
 or 
http://www.apache.org/dyn/mirrors/mirrors.cgi/flex/falcon/0.0.2/binaries/apache-flex-falconjx-0.0.2-bin.zip


- Till


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41485/#review110872
---


On Dec. 17, 2015, 1:31 a.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41485/
> ---
> 
> (Updated Dec. 17, 2015, 1:31 a.m.)
> 
> 
> Review request for mesos and Bernd Mathiske.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   site/source/downloads.html.md dbbeebe81939a9e73ceecb84e0dd426ec0f900f2 
> 
> Diff: https://reviews.apache.org/r/41485/diff/
> 
> 
> Testing
> ---
> 
> mesos site rake.
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41406: Unified Container: Added path helper method to get image layer manifest on local store.

[Gilbert Song]

> On Dec. 16, 2015, 9:38 p.m., Timothy Chen wrote:
> > src/slave/containerizer/mesos/provisioner/docker/paths.hpp, line 40
> > 
> >
> > VERSION This is only relevant for local puller right?

Currently yes. I am following the way docker save info in each layer. We may 
need this information in the future. So we may want to keep it(only the size of 
`float` number).


- Gilbert


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41406/#review110901
---


On Dec. 16, 2015, 6:38 p.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41406/
> ---
> 
> (Updated Dec. 16, 2015, 6:38 p.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Added path helper method to get image layer manifest on 
> local store.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/docker/paths.hpp 
> 7e5122d956c82e6cbd62fc217158205ea05f4290 
>   src/slave/containerizer/mesos/provisioner/docker/paths.cpp 
> 9c1f6a19c59b12f58b0550d022f9128e7015a4e9 
> 
> Diff: https://reviews.apache.org/r/41406/diff/
> 
> 
> Testing
> ---
> 
> make check
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 41487: Provisioner: Changed docker v2 manifest naming.

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41487/#review110909
---

Ship it!


Ship It!


src/tests/containerizer/provisioner_docker_tests.cpp (line 303)


Let's make the naming consistent -> SerializeV2DockerManifest



src/tests/containerizer/provisioner_docker_tests.cpp (line 484)


ditto


- Timothy Chen


On Dec. 17, 2015, 2:38 a.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41487/
> ---
> 
> (Updated Dec. 17, 2015, 2:38 a.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-4186
> https://issues.apache.org/jira/browse/MESOS-4186
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Provisioner: Changed docker v2 manifest naming.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/docker/message.proto 
> 5c032701671b275d86c6d9276791a46df814396c 
>   src/slave/containerizer/mesos/provisioner/docker/registry_client.hpp 
> 9b02b6ff6dc5c6e8aabdc4ac0aa4df337764ef30 
>   src/slave/containerizer/mesos/provisioner/docker/registry_client.cpp 
> e69bab43d5f5359cfd8eb4cb7c5ad4a1d22c4e05 
>   src/slave/containerizer/mesos/provisioner/docker/registry_puller.cpp 
> 89f61c20e52e5eff8d8e92748f03b3b461516cd2 
>   src/slave/containerizer/mesos/provisioner/docker/spec.hpp 
> e674b6f2e705f8b3e49770560eeab4c127473f94 
>   src/slave/containerizer/mesos/provisioner/docker/spec.cpp 
> 1f05c75dc9473bd5e4c0d3f74fa0ef996b96a84e 
>   src/tests/containerizer/provisioner_docker_tests.cpp 
> 3f1717b770e139c3759aab0aeda9dbcf5029b0c2 
> 
> Diff: https://reviews.apache.org/r/41487/diff/
> 
> 
> Testing
> ---
> 
> make check
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 40167: [2/7] Added ACL protobuf messages 'CreateVolume' and 'DestroyVolume'.

[Greg Mann]

> On Dec. 16, 2015, 2:35 p.m., Alexander Rukletsov wrote:
> > include/mesos/authorizer/authorizer.proto, line 96
> > 
> >
> > I see that you ensure this when you create requests in 
> > `authorizeCreateVolume`, but I'm not sure it's validated for ACLs. Do you 
> > think it makes sense to add validation in the 
> > `LocalAuthorizer::initialize()`?
> > 
> > After a second thought, this note relates to the default 
> > implementation, because the master does not really validate it, right? 
> > Which means a 3rdparty authorizer may react to particular types.

You're correct that a 3rd party authorizer could handle values other than ANY 
and NONE, but the current default implementation enforces this constraint by 
always setting `volume_types` equal to ANY in the request ACL. I altered this 
comment to clarify that point a bit; feel free to re-open the issue if you 
think we should do more.


- Greg


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/40167/#review110641
---


On Dec. 17, 2015, 6:58 a.m., Greg Mann wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/40167/
> ---
> 
> (Updated Dec. 17, 2015, 6:58 a.m.)
> 
> 
> Review request for mesos, Jie Yu, Michael Park, and Neil Conway.
> 
> 
> Bugs: MESOS-4178
> https://issues.apache.org/jira/browse/MESOS-4178
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Added ACL protobuf messages 'CreateVolume' and 'DestroyVolume'.
> 
> 
> Diffs
> -
> 
>   include/mesos/authorizer/authorizer.proto 
> 74fcc86d3c92cb3aa27e45b647b1653705b3201c 
> 
> Diff: https://reviews.apache.org/r/40167/diff/
> 
> 
> Testing
> ---
> 
> This is the second in a chain of 7 patches. `make check` was used to test 
> after all patches were applied.
> 
> Note that this chain of patches touches many of the same files as another 
> chain beginning with Review #39985 and ending with Review #39989, which is 
> currently in review as well. To avoid conflicts, the beginning of this chain 
> begins on top of Review #39989.
> 
> 
> Thanks,
> 
> Greg Mann
> 
>

Re: Review Request 41244: Using 'git rev-parse --git-dir' in post-reviews.py.

[Till Toenshoff]

> On Dec. 17, 2015, 3:43 a.m., Till Toenshoff wrote:
> > Ship It!

I noted some of your recent reviews contain the review-request link in the 
description shown on RB (this one doesnt) - can you please make sure its not 
anything your own recent changes are related to, thanks.


- Till


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41244/#review110884
---


On Dec. 17, 2015, 3:44 a.m., Kevin Klues wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41244/
> ---
> 
> (Updated Dec. 17, 2015, 3:44 a.m.)
> 
> 
> Review request for mesos, Benjamin Hindman, Artem Harutyunyan, Till 
> Toenshoff, and Vinod Kone.
> 
> 
> Bugs: MESOS-4125
> https://issues.apache.org/jira/browse/MESOS-4125
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> The post-reviews.py script has similar problems to the bootstrap script
> in assuming that .git is a folder in the top-level directory of the git
> repo.  This may not be true when mesos is included as a submodule in
> another project. Use 'git rev-parse --git-dir' instead.
> 
> 
> Diffs
> -
> 
>   support/post-reviews.py 170be83aa6dca6e8175292169d78e8f7915f7e6e 
> 
> Diff: https://reviews.apache.org/r/41244/diff/
> 
> 
> Testing
> ---
> 
> I have been posting all of my reviews with this script since first creating 
> this commit. Tested both in an environment with and without mesos embedded as 
> a submodule inside another git repository.
> 
> 
> Thanks,
> 
> Kevin Klues
> 
>

Re: Review Request 41485: [site] Updated downloads-page for 0.26.0.

[Mesos ReviewBot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41485/#review110890
---


Patch looks great!

Reviews applied: [41467, 41468, 41479, 41480, 41485]

Passed command: export OS=ubuntu:14.04;export CONFIGURATION="--verbose";export 
COMPILER=gcc; ./support/docker_build.sh

- Mesos ReviewBot


On Dec. 17, 2015, 1:31 a.m., Till Toenshoff wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41485/
> ---
> 
> (Updated Dec. 17, 2015, 1:31 a.m.)
> 
> 
> Review request for mesos and Bernd Mathiske.
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> see summary.
> 
> 
> Diffs
> -
> 
>   site/source/downloads.html.md dbbeebe81939a9e73ceecb84e0dd426ec0f900f2 
> 
> Diff: https://reviews.apache.org/r/41485/diff/
> 
> 
> Testing
> ---
> 
> mesos site rake.
> 
> 
> Thanks,
> 
> Till Toenshoff
> 
>

Re: Review Request 41194: Unified Container: Cleanup uneccessary JSON parse in local puller.

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41194/#review110899
---

Ship it!


Ship It!

- Timothy Chen


On Dec. 17, 2015, 2:38 a.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41194/
> ---
> 
> (Updated Dec. 17, 2015, 2:38 a.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Cleanup uneccessary JSON parse in local puller.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/docker/local_puller.cpp 
> f8856664b39de0a07a3c3d7de8e03658acba9631 
> 
> Diff: https://reviews.apache.org/r/41194/diff/
> 
> 
> Testing
> ---
> 
> make check(ubuntu14.04 + clang-3.6)
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 41125: Unified Container: Changed docker store and appc store to support passing runtime config to provisioner.

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41125/#review110898
---

Ship it!


Ship It!

- Timothy Chen


On Dec. 17, 2015, 2:38 a.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41125/
> ---
> 
> (Updated Dec. 17, 2015, 2:38 a.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Changed docker store and appc store to support passing 
> runtime config to provisioner.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/appc/store.hpp 
> b2a77802f49c76f36f49282c7c77dfa51a602a9a 
>   src/slave/containerizer/mesos/provisioner/appc/store.cpp 
> 439ae0eb804f8d50081015165e6f2eca28df520f 
>   src/slave/containerizer/mesos/provisioner/docker/store.hpp 
> e246a1460ecdf9855ccafb43379572425de4ffd3 
>   src/slave/containerizer/mesos/provisioner/docker/store.cpp 
> 1ad7b67a94b1d9367afcb7c30a6d01fdf6b8ab6c 
>   src/slave/containerizer/mesos/provisioner/provisioner.hpp 
> b5d40b9428c9aa2f397540ef275cca7a7a335e43 
>   src/slave/containerizer/mesos/provisioner/provisioner.cpp 
> 8a2c9bbf7e01a62ad92ba19216b32f6e813cd736 
>   src/slave/containerizer/mesos/provisioner/store.hpp 
> 8d1493856420dee3210af79b628c8c770c5c8550 
>   src/tests/containerizer/provisioner_appc_tests.cpp 
> 9c5d91ba1fd49602418a0dd7c26b3132d3ac1a68 
>   src/tests/containerizer/provisioner_docker_tests.cpp 
> 3f1717b770e139c3759aab0aeda9dbcf5029b0c2 
>   src/tests/containerizer/store.hpp 0d2879602f01e3dfcc20d7f2272ea01e47729574 
> 
> Diff: https://reviews.apache.org/r/41125/diff/
> 
> 
> Testing
> ---
> 
> make check (ubuntu14.04 + clang-3.6)
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 41406: Unified Container: Added path helper method to get image layer manifest on local store.

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41406/#review110900
---

Ship it!


Ship It!

- Timothy Chen


On Dec. 17, 2015, 2:38 a.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41406/
> ---
> 
> (Updated Dec. 17, 2015, 2:38 a.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Added path helper method to get image layer manifest on 
> local store.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/docker/paths.hpp 
> 7e5122d956c82e6cbd62fc217158205ea05f4290 
>   src/slave/containerizer/mesos/provisioner/docker/paths.cpp 
> 9c1f6a19c59b12f58b0550d022f9128e7015a4e9 
> 
> Diff: https://reviews.apache.org/r/41406/diff/
> 
> 
> Testing
> ---
> 
> make check
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 41407: Unified Container: Save all docker image information on disk instead of rootfs only.

[Timothy Chen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41407/#review110908
---

Ship it!


Ship It!

- Timothy Chen


On Dec. 17, 2015, 2:38 a.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41407/
> ---
> 
> (Updated Dec. 17, 2015, 2:38 a.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Save all docker image information on disk instead of 
> rootfs only.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/docker/puller.cpp 
> 4aa4a9c4074d96c30c3bceea59d071feeecae2ea 
>   src/slave/containerizer/mesos/provisioner/docker/store.cpp 
> 1ad7b67a94b1d9367afcb7c30a6d01fdf6b8ab6c 
>   src/tests/containerizer/provisioner_docker_tests.cpp 
> 3f1717b770e139c3759aab0aeda9dbcf5029b0c2 
> 
> Diff: https://reviews.apache.org/r/41407/diff/
> 
> 
> Testing
> ---
> 
> make check
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>

Re: Review Request 41346: WIP: Evict some executors if there are not enough reserved resources.

[Guangya Liu]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41346/
---

(Updated Dec. 17, 2015, 6:28 a.m.)


Review request for mesos and Jian Qiu.


Repository: mesos


Description
---

Evict some executors if there are not enough reserved resources.


Diffs (updated)
-

  src/slave/slave.hpp b7586ce42bfac9d9885a3eb8d82deb94680c236c 
  src/slave/slave.cpp ef869695ffeb2e6d9ef0a78ddb676b1b7cd19afe 

Diff: https://reviews.apache.org/r/41346/diff/


Testing
---


Thanks,

Guangya Liu

Re: Review Request 41407: Unified Container: Save all docker image information on disk instead of rootfs only.

[Gilbert Song]

> On Dec. 16, 2015, 10:03 a.m., Jie Yu wrote:
> > src/slave/containerizer/mesos/provisioner/docker/puller.cpp, lines 183-184
> > 
> >
> > s/stagingLayerPath/layerPath/

I am totally agreed with this change. But we have another layerPath above in 
this method (not causing any error thougg), which is confusing. Do you think it 
is good to change:
s/layerPath/layerTarPath above on Line145 & Line180


- Gilbert


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/41407/#review110705
---


On Dec. 16, 2015, 6:38 p.m., Gilbert Song wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/41407/
> ---
> 
> (Updated Dec. 16, 2015, 6:38 p.m.)
> 
> 
> Review request for mesos, Artem Harutyunyan, Jie Yu, and Timothy Chen.
> 
> 
> Bugs: MESOS-2980
> https://issues.apache.org/jira/browse/MESOS-2980
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Unified Container: Save all docker image information on disk instead of 
> rootfs only.
> 
> 
> Diffs
> -
> 
>   src/slave/containerizer/mesos/provisioner/docker/puller.cpp 
> 4aa4a9c4074d96c30c3bceea59d071feeecae2ea 
>   src/slave/containerizer/mesos/provisioner/docker/store.cpp 
> 1ad7b67a94b1d9367afcb7c30a6d01fdf6b8ab6c 
>   src/tests/containerizer/provisioner_docker_tests.cpp 
> 3f1717b770e139c3759aab0aeda9dbcf5029b0c2 
> 
> Diff: https://reviews.apache.org/r/41407/diff/
> 
> 
> Testing
> ---
> 
> make check
> 
> 
> Thanks,
> 
> Gilbert Song
> 
>