Re: Review Request 70585: Test for a race between resubscriptions with different principals - WIP.

[Mesos Reviewbot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70585/#review215043
---



Bad review!

Reviews applied: []

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

- Mesos Reviewbot


On May 2, 2019, 6:47 p.m., Andrei Sekretenko wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70585/
> ---
> 
> (Updated May 2, 2019, 6:47 p.m.)
> 
> 
> Review request for mesos.
> 
> 
> Bugs: MESOS-9763
> https://issues.apache.org/jira/browse/MESOS-9763
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Test for a race between resubscriptions with different principals.
> 
> This is a work in progress: currently I'm lucky enough that this test fails 
> EVERY time, but it is necessary to add something to reproduce this race 
> reliably.
> 
> 
> Diffs
> -
> 
>   src/tests/http_fault_tolerance_tests.cpp 
> a9e8d0c8962d9c978e0488e16317117d55d214fa 
> 
> 
> Diff: https://reviews.apache.org/r/70585/diff/1/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Andrei Sekretenko
> 
>

Re: Review Request 70534: Added tests for UPDATE_FRAMEWORK.

[Mesos Reviewbot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70534/#review215042
---



Patch looks great!

Reviews applied: [70583, 70531, 70530, 70532, 70533, 70534]

Passed command: export OS='ubuntu:14.04' BUILDTOOL='autotools' COMPILER='gcc' 
CONFIGURATION='--verbose --disable-libtool-wrappers 
--disable-parallel-test-execution' ENVIRONMENT='GLOG_v=1 MESOS_VERBOSE=1'; 
./support/docker-build.sh

- Mesos Reviewbot


On May 2, 2019, 4:06 p.m., Andrei Sekretenko wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70534/
> ---
> 
> (Updated May 2, 2019, 4:06 p.m.)
> 
> 
> Review request for mesos and Benjamin Mahler.
> 
> 
> Bugs: MESOS-7258
> https://issues.apache.org/jira/browse/MESOS-7258
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This patch adds tests for the UPDATE_FRAMEWORK call.
> 
> The reason behind implementing the MasterAPISubscriber class with mock 
> methods is the need to wait for interesting API events while skipping 
> uninteresting ones. 
> This is needed to properly test the FRAMEWORK_UPDATED event sent by the 
> master (for that I need to wait for AGENT_ADDED and FRAMEWORK_UPDATED, and to 
> skip everything else).
> 
> I'm in doubts about the proper location for the MasterAPISubscriber. On one 
> hand, it is used only by these tests and the implementation is by no means 
> complete; on the other hand, this might be useful in other tests of the 
> master API behaviour. 
> Probably, turning the MasterAPISubscriber into a general-purpose mock (and 
> moving it into another file at that point) should be a separate task.
> 
> 
> Diffs
> -
> 
>   src/Makefile.am 5f97523fbe2d80733fbdcc7706f2761f5a071f9f 
>   src/tests/CMakeLists.txt e6b1d8a097246f0921467d7f778034d6754fcff6 
>   src/tests/update_framework_tests.cpp PRE-CREATION 
> 
> 
> Diff: https://reviews.apache.org/r/70534/diff/3/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Andrei Sekretenko
> 
>

Re: Review Request 70587: Fixed unguarded calls to `Option::get()` in the master.

[Gastón Kleiman]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70587/#review215039
---


Ship it!




Ship It!

- Gastón Kleiman


On May 2, 2019, 3 p.m., Greg Mann wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70587/
> ---
> 
> (Updated May 2, 2019, 3 p.m.)
> 
> 
> Review request for mesos, Benno Evers, Gastón Kleiman, Joseph Wu, and Till 
> Toenshoff.
> 
> 
> Bugs: MESOS-9698
> https://issues.apache.org/jira/browse/MESOS-9698
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This patch avoids making unguarded calls to `Option::get()`
> in `Master::updateOperationStatus()`. During agent reregistration, it's
> possible that a `ReregisterSlaveMessage` from the agent can race with a
> `SlaveReregisteredMessage` from the master, leading to multiple rounds
> of master/agent operation reconciliation. The duplicate operation status
> updates which occur as a result would crash the master before this fix.
> 
> 
> Diffs
> -
> 
>   src/master/master.cpp 7dcdc9ab62a46638a027eb9a54c1dff173785927 
>   src/tests/agent_operation_feedback_tests.cpp 
> e427441b3ef702acf0fba52adf7ba027ea6bc508 
> 
> 
> Diff: https://reviews.apache.org/r/70587/diff/1/
> 
> 
> Testing
> ---
> 
> `make check`
> `bin/mesos-tests.sh --gtest_filter="*DroppedOperationDuplicateStatusUpdate*" 
> --gtest_repeat=-1 --gtest_break_on_failure`
> 
> 
> Thanks,
> 
> Greg Mann
> 
>

Re: Review Request 70570: Logged headroom related info in the allocator.

[Mesos Reviewbot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70570/#review215038
---



Patch looks great!

Reviews applied: [70570]

Passed command: export OS='ubuntu:14.04' BUILDTOOL='autotools' COMPILER='gcc' 
CONFIGURATION='--verbose --disable-libtool-wrappers 
--disable-parallel-test-execution' ENVIRONMENT='GLOG_v=1 MESOS_VERBOSE=1'; 
./support/docker-build.sh

- Mesos Reviewbot


On April 30, 2019, 1 a.m., Meng Zhu wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70570/
> ---
> 
> (Updated April 30, 2019, 1 a.m.)
> 
> 
> Review request for mesos and Benjamin Mahler.
> 
> 
> Bugs: MESOS-9759
> https://issues.apache.org/jira/browse/MESOS-9759
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Logged headroom related info in the allocator.
> 
> 
> Diffs
> -
> 
>   src/master/allocator/mesos/hierarchical.cpp 
> 64a076ddd29711437d539a06bb0470755828cc87 
> 
> 
> Diff: https://reviews.apache.org/r/70570/diff/3/
> 
> 
> Testing
> ---
> 
> N/A
> 
> 
> Thanks,
> 
> Meng Zhu
> 
>

Re: Review Request 70594: Fixed an issue where /__processes__ never returns a response.

[Chun-Hung Hsiao]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70594/#review215037
---


Ship it!




How about quoting `/__processes__` with backticks?

- Chun-Hung Hsiao


On May 3, 2019, 7:58 p.m., Benjamin Mahler wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70594/
> ---
> 
> (Updated May 3, 2019, 7:58 p.m.)
> 
> 
> Review request for mesos, Alexander Rukletsov and Chun-Hung Hsiao.
> 
> 
> Bugs: MESOS-9766
> https://issues.apache.org/jira/browse/MESOS-9766
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> It's possible for /__processes__ to never return a response if
> a process is terminated after the /__processes__ handler dispatches
> to it, thus leading the Future to be abandoned.
> 
> This patch ensures we ignore those cases, rather than wait forever.
> 
> See MESOS-9766.
> 
> 
> Diffs
> -
> 
>   3rdparty/libprocess/src/process.cpp 
> 124836472313721a5dbfe4b1ca55f0da3cecd66b 
> 
> 
> Diff: https://reviews.apache.org/r/70594/diff/1/
> 
> 
> Testing
> ---
> 
> Added a test in the subsequent patch.
> 
> 
> Thanks,
> 
> Benjamin Mahler
> 
>

Re: Review Request 70595: Adds a regression test for MESOS-9766.

[Chun-Hung Hsiao]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70595/#review215035
---




3rdparty/libprocess/src/tests/process_tests.cpp
Lines 2086 (patched)


Should `/__processes__` be quoted by backticks? Ditto below.



3rdparty/libprocess/src/tests/process_tests.cpp
Lines 2090 (patched)


How about `NoProcessesEndpointHang`?



3rdparty/libprocess/src/tests/process_tests.cpp
Lines 2092-2108 (patched)


I tried this unit test w/o the fix in the following settings:
1. `--gtest_repeat=100`, which resulted in 3 passes and 97 failures.
2. Invoked the test w/o repetition in a for-loop, which resulted in 60 
passes and 40 failures.

So it seems to me that this test passes fairly easily if it only runs once.

Then I replace this snippet with the following code, which just used 1 
actor but failed every time in the above two settings:
```
  UPID pid = spawn(new ProcessBase(), true);

  dispatch(pid, [] { os::sleep(Milliseconds(50)); });

  http::URL url = http::URL(
  "http",
  process::address().ip,
  process::address().port,
  "/__processes__");

  Future response = http::get(url);

  terminate(pid, true);
```



3rdparty/libprocess/src/tests/process_tests.cpp
Lines 2110-2112 (patched)


`AWAIT_EXPECT_RESPONSE_STATUS_EQ(http::OK().status, response);`


- Chun-Hung Hsiao


On May 3, 2019, 7:58 p.m., Benjamin Mahler wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70595/
> ---
> 
> (Updated May 3, 2019, 7:58 p.m.)
> 
> 
> Review request for mesos, Alexander Rukletsov and Chun-Hung Hsiao.
> 
> 
> Bugs: MESOS-9766
> https://issues.apache.org/jira/browse/MESOS-9766
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This test fails on master prior to applying the fix for MESOS-9766.
> It attempts to ensure that processes are terminated after the
> /__processes__ handler dispatches to them.
> 
> 
> Diffs
> -
> 
>   3rdparty/libprocess/src/tests/process_tests.cpp 
> 60f3dd653153c2b2ccf9c3a7eae8c75fd6ff025c 
> 
> 
> Diff: https://reviews.apache.org/r/70595/diff/1/
> 
> 
> Testing
> ---
> 
> Ran in repetition, although it appears to consistently fail on master without 
> repetition needed.
> 
> 
> Thanks,
> 
> Benjamin Mahler
> 
>

Re: Review Request 70580: Added tests for overlapping ranges and sets in task/executor resources.

[Mesos Reviewbot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70580/#review215033
---



Patch looks great!

Reviews applied: [70579, 70580]

Passed command: export OS='ubuntu:14.04' BUILDTOOL='autotools' COMPILER='gcc' 
CONFIGURATION='--verbose --disable-libtool-wrappers 
--disable-parallel-test-execution' ENVIRONMENT='GLOG_v=1 MESOS_VERBOSE=1'; 
./support/docker-build.sh

- Mesos Reviewbot


On May 1, 2019, 9:17 p.m., Benjamin Mahler wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70580/
> ---
> 
> (Updated May 1, 2019, 9:17 p.m.)
> 
> 
> Review request for mesos and Greg Mann.
> 
> 
> Bugs: MESOS-9619
> https://issues.apache.org/jira/browse/MESOS-9619
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This is a re-write of https://reviews.apache.org/r/70509/ to use the
> task scheduler.
> 
> 
> Diffs
> -
> 
>   src/tests/master_tests.cpp 964d935771a99efaee63187affe46b551146f310 
> 
> 
> Diff: https://reviews.apache.org/r/70580/diff/1/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Benjamin Mahler
> 
>

Review Request 70596: Launched tasks with more memory in SLRP unit tests.

[Chun-Hung Hsiao]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70596/
---

Review request for mesos, Benjamin Bannier and Jan Schlicht.


Bugs: MESOS-9765
https://issues.apache.org/jira/browse/MESOS-9765


Repository: mesos


Description
---

Raised the task memory to 128MB (which are the value used in most persistent
volume tests) in all SLRP tests that launch tasks to avoid OOM.


Diffs
-

  src/tests/storage_local_resource_provider_tests.cpp 
ecd7aeef1cb3d1a5b4b3419dfd912d41a8c6 


Diff: https://reviews.apache.org/r/70596/diff/1/


Testing
---

sudo make check


Thanks,

Chun-Hung Hsiao

Review Request 70594: Fixed an issue where /__processes__ never returns a response.

[Benjamin Mahler]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70594/
---

Review request for mesos, Alexander Rukletsov and Chun-Hung Hsiao.


Bugs: MESOS-9766
https://issues.apache.org/jira/browse/MESOS-9766


Repository: mesos


Description
---

It's possible for /__processes__ to never return a response if
a process is terminated after the /__processes__ handler dispatches
to it, thus leading the Future to be abandoned.

This patch ensures we ignore those cases, rather than wait forever.

See MESOS-9766.


Diffs
-

  3rdparty/libprocess/src/process.cpp 124836472313721a5dbfe4b1ca55f0da3cecd66b 


Diff: https://reviews.apache.org/r/70594/diff/1/


Testing
---

Added a test in the subsequent patch.


Thanks,

Benjamin Mahler

Review Request 70595: Adds a regression test for MESOS-9766.

[Benjamin Mahler]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70595/
---

Review request for mesos, Alexander Rukletsov and Chun-Hung Hsiao.


Bugs: MESOS-9766
https://issues.apache.org/jira/browse/MESOS-9766


Repository: mesos


Description
---

This test fails on master prior to applying the fix for MESOS-9766.
It attempts to ensure that processes are terminated after the
/__processes__ handler dispatches to them.


Diffs
-

  3rdparty/libprocess/src/tests/process_tests.cpp 
60f3dd653153c2b2ccf9c3a7eae8c75fd6ff025c 


Diff: https://reviews.apache.org/r/70595/diff/1/


Testing
---

Ran in repetition, although it appears to consistently fail on master without 
repetition needed.


Thanks,

Benjamin Mahler

Review Request 70573: Updated bintray upload scripts to remove hard-coded accounts.

[Benno Evers]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70573/
---

Review request for mesos and Till Toenshoff.


Bugs: MESOS-9697
https://issues.apache.org/jira/browse/MESOS-9697


Repository: mesos


Description
---

This updates the Jenkinsfile used in the ASF Jenkins as well
as the associated upload script for bintray to make the
used credential id configurable and to upload the built packages
to the official `apache/mesos` bintray account.


Diffs
-

  support/packaging/Jenkinsfile 7446b2b8b88df6c146f4415b06dda227ca8ce631 
  support/packaging/bintray.sh 4f1795f14b078e8eebe95aaa40cf860d6d19ee97 


Diff: https://reviews.apache.org/r/70573/diff/1/


Testing
---

Not sure how to test jenkins pipelines :/


Thanks,

Benno Evers

Re: Review Request 70497: Optimize the random sorter with random sampling.

[Mesos Reviewbot]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70497/#review215030
---



Patch looks great!

Reviews applied: [70419, 70576, 70497]

Passed command: export OS='ubuntu:14.04' BUILDTOOL='autotools' COMPILER='gcc' 
CONFIGURATION='--verbose --disable-libtool-wrappers 
--disable-parallel-test-execution' ENVIRONMENT='GLOG_v=1 MESOS_VERBOSE=1'; 
./support/docker-build.sh

- Mesos Reviewbot


On May 1, 2019, 1:36 a.m., Meng Zhu wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70497/
> ---
> 
> (Updated May 1, 2019, 1:36 a.m.)
> 
> 
> Review request for mesos, Andrei Sekretenko and Benjamin Mahler.
> 
> 
> Bugs: MESOS-9725
> https://issues.apache.org/jira/browse/MESOS-9725
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> This patch optimizes the random sorter by avoiding
> clients shuffling. The sorter now does random sampling
> as the caller asks for the next client.
> 
> The random sampling works in two steps. Clients with the same
> relative weights are grouped together. In the first phase, we
> randomly pick a group of clients. This requires the generation
> of a weighted random number. In the second phase, a client within
> the group is picked. Since all clients in the group have the same
> weight, this can be done in constant time. This two step approach
> minimizes the cost of generating weighted random number which
> could be expensive given the size of the weights.
> 
> 
> Diffs
> -
> 
>   src/master/allocator/sorter/random/sorter.hpp 
> 55e22d7705f163fe47d5aa47416ee0714c5a87c0 
>   src/master/allocator/sorter/random/sorter.cpp 
> 813f5b55d38dd9fa822de53ee944c3f72251a69d 
> 
> 
> Diff: https://reviews.apache.org/r/70497/diff/2/
> 
> 
> Testing
> ---
> 
> make check
> 
> Benchmarking: ran 
> `QuotaParam/BENCHMARK_HierarchicalAllocator_WithQuotaParam.LargeAndSmallQuota/5`
>  with optimized build
> 
> Before:
> 
> Added 3000 agents in 101.884509ms
> Added 3000 frameworks in 19.711779311secs
> Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
> Made 3856 allocations in 16.283607645secs
> Made 0 allocation in 16.31197771secs
> 
> After r/70419, r/70576 and r/70497 :
> 
> Added 3000 agents in 87.147084ms
> Added 3000 frameworks in 19.246494668secs
> Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
> Made 3872 allocations in 12.230518989secs
> Made 0 allocation in 12.012211914secs
> 
> 
> Thanks,
> 
> Meng Zhu
> 
>

Re: Review Request 70593: Fixed flaky test FrameworkPrincipalChangeFails.

[Gastón Kleiman]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70593/#review215027
---



The review message doesn't conform with our commit message style.

I'd propose something like this:

```
Fixed flaky test 'FrameworkPrincipalChangeFails'.

Currently the test is flaky due to the race between `EXPECT_CALL` and the
scheduler API.

This patch moves the EXPECT_CALLs before `v1::scheduler::TestMesos` is
instantiated.

Review: https://reviews.apache.org/r/70593/
```

- Gastón Kleiman


On May 3, 2019, 9:38 a.m., Andrei Sekretenko wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70593/
> ---
> 
> (Updated May 3, 2019, 9:38 a.m.)
> 
> 
> Review request for mesos and Gastón Kleiman.
> 
> 
> Bugs: MESOS-2842
> https://issues.apache.org/jira/browse/MESOS-2842
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Currently the test is flaky due to the race between EXPECT_CALL and the 
> scheduler API. 
> EXPECT_CALLs should be moved before v1::scheduler::TestMesos() - I don't know 
> why we have overlooked this in https://reviews.apache.org/r/70377 .
> 
> 
> Diffs
> -
> 
>   src/tests/http_fault_tolerance_tests.cpp 
> a9e8d0c8962d9c978e0488e16317117d55d214fa 
> 
> 
> Diff: https://reviews.apache.org/r/70593/diff/1/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Andrei Sekretenko
> 
>

Re: Review Request 70593: Fixed flaky test FrameworkPrincipalChangeFails.

[Gastón Kleiman]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70593/#review215022
---


Ship it!




Ship It!

- Gastón Kleiman


On May 3, 2019, 9:38 a.m., Andrei Sekretenko wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/70593/
> ---
> 
> (Updated May 3, 2019, 9:38 a.m.)
> 
> 
> Review request for mesos and Gastón Kleiman.
> 
> 
> Bugs: MESOS-2842
> https://issues.apache.org/jira/browse/MESOS-2842
> 
> 
> Repository: mesos
> 
> 
> Description
> ---
> 
> Currently the test is flaky due to the race between EXPECT_CALL and the 
> scheduler API. 
> EXPECT_CALLs should be moved before v1::scheduler::TestMesos() - I don't know 
> why we have overlooked this in https://reviews.apache.org/r/70377 .
> 
> 
> Diffs
> -
> 
>   src/tests/http_fault_tolerance_tests.cpp 
> a9e8d0c8962d9c978e0488e16317117d55d214fa 
> 
> 
> Diff: https://reviews.apache.org/r/70593/diff/1/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Andrei Sekretenko
> 
>

Review Request 70593: Fixed flaky test FrameworkPrincipalChangeFails.

[Andrei Sekretenko]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70593/
---

Review request for mesos and Gastón Kleiman.


Bugs: MESOS-2842
https://issues.apache.org/jira/browse/MESOS-2842


Repository: mesos


Description
---

Currently the test is flaky due to the race between EXPECT_CALL and the 
scheduler API. 
EXPECT_CALLs should be moved before v1::scheduler::TestMesos() - I don't know 
why we have overlooked this in https://reviews.apache.org/r/70377 .


Diffs
-

  src/tests/http_fault_tolerance_tests.cpp 
a9e8d0c8962d9c978e0488e16317117d55d214fa 


Diff: https://reviews.apache.org/r/70593/diff/1/


Testing
---


Thanks,

Andrei Sekretenko

Review Request 70591: Introduced struct RoleInfo to track reservations and framework IDs in one map.

[Andrei Sekretenko]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70591/
---

Review request for mesos, Benjamin Mahler and Meng Zhu.


Bugs: MESOS-9701
https://issues.apache.org/jira/browse/MESOS-9701


Repository: mesos


Description
---

This patch introduces `struct RoleInfo` which contains the framework IDs and 
reservations tied to the role and replaces `roles` and 
`reservationScalarQuantities` hashmaps with `hashmap`.

I personally do not like the name `RoleInfo` and would appreciate any better 
naming ideas. 
However, naming this entity `Role` would have been even worse: it would imply 
using `const Role& role` throughout the code, which would require changing all 
the places which use `std::string role` as a key...


Diffs
-

  src/master/allocator/mesos/hierarchical.hpp 
c2058baca5159da4cdcab77afd5de3c0d5ae6c48 
  src/master/allocator/mesos/hierarchical.cpp 
64a076ddd29711437d539a06bb0470755828cc87 


Diff: https://reviews.apache.org/r/70591/diff/1/


Testing
---

make check

Benchmarking: 5 runs of 
`BENCHMARK_HierarchicalAllocator_WithQuotaParam.LargeAndSmallQuota/5` with the 
optimized build.

Performance impact in this benchmark seems to be negligible.

BEFORE:
Added 3000 agents in 51.553929ms
Added 3000 frameworks in 15.174748344secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.400805171secs
Made 0 allocation in 12.5850238secs

Added 3000 agents in 55.739336ms
Added 3000 frameworks in 14.730404769secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.563439682secs
Made 0 allocation in 13.063555055secs

Added 3000 agents in 54.414733ms
Added 3000 frameworks in 15.10136842secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 13.501664915secs
Made 0 allocation in 12.89034382secs

Added 3000 agents in 52.58252ms
Added 3000 frameworks in 14.048350298secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.299952145secs
Made 0 allocation in 11.888248811secs

Added 3000 agents in 52.821439ms
Added 3000 frameworks in 15.344450583secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.63425secs
Made 0 allocation in 12.427171541secs


AFTER:

Added 3000 agents in 69.716648ms
Added 3000 frameworks in 15.249001979secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.860494226secs
Made 0 allocation in 12.228866329secs

Added 3000 agents in 52.639388ms
Added 3000 frameworks in 15.207895482secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.504777266secs
Made 0 allocation in 12.70388062secs

Added 3000 agents in 56.865794ms
Added 3000 frameworks in 15.284003915secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 13.86859815secs
Made 0 allocation in 12.538958231secs

Added 3000 agents in 56.028013ms
Added 3000 frameworks in 13.892577869secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.341724418secs
Made 0 allocation in 12.23022189secs

Added 3000 agents in 52.368219ms
Added 3000 frameworks in 13.978581104secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks, with random sorter
Made 3856 allocations in 12.701682501secs
Made 0 allocation in 12.141360313secs


Thanks,

Andrei Sekretenko

Review Request 70589: Logged when `/__processes__` returns.

[Alexander Rukletsov]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/70589/
---

Review request for mesos, Benjamin Mahler and Till Toenshoff.


Repository: mesos


Description
---

Adds a log entry when a response with generated by `/__processes__`
is about to be returned to the client.


Diffs
-

  3rdparty/libprocess/src/process.cpp 124836472313721a5dbfe4b1ca55f0da3cecd66b 


Diff: https://reviews.apache.org/r/70589/diff/1/


Testing
---

built & run on Mac OS 10.13.6

```
I0503 13:13:00.885133 59219968 process.cpp:3599] Handling HTTP event for 
process '__processes__' with path: '/__processes__'
I0503 13:13:00.886037 59219968 process.cpp:3412] HTTP GET for /__processes__ 
from 192.168.178.47:59548: '200 OK' after 1.06214ms
```


Thanks,

Alexander Rukletsov