Mailing-List: contact mesos-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: mesos-dev@incubator.apache.org
Content-Type: multipart/alternative;
	boundary="===============2180678955089637090=="
MIME-Version: 1.0
Subject: Re: Review Request: Non-disruptive Slave Restart with Recovery!
From: "Vinod Kone" <vinodkone@gmail.com>
To: "Benjamin Hindman" <benh@berkeley.edu>,
	"John Sirois" <john.sirois@gmail.com>
Date: Tue, 03 Apr 2012 20:58:18 -0000
Message-ID: <20120403205818.4079.20135@reviews.apache.org>
Cc: "Charles Reiss" <woggling@gmail.com>,
	"mesos" <mesos-dev@incubator.apache.org>,
	"Vinod Kone" <vinodkone@gmail.com>
In-Reply-To: <20120323011812.382.77230@reviews.apache.org>
References: <20120323011812.382.77230@reviews.apache.org>

--===============2180678955089637090==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/4462/
-----------------------------------------------------------

(Updated 2012-04-03 20:58:17.982746)


Review request for mesos, Benjamin Hindman and John Sirois.


Changes
-------

merged with trunk


Summary
-------

Sorry for the huge  CL!

Slave restarts now supports recovery!
--> Non-disruptive restart means running tasks are not lost
--> Re-connects with live executors
--> Checkpoints and reliably sends status updates
--> Ability to kill executors if the slave upgrade is incompatible with run=
ning executors


This addresses bug mesos-110.
    https://issues.apache.org/jira/browse/mesos-110


Diffs (updated)
-----

  src/Makefile.am d5edaa2 =

  src/common/hashset.hpp 1feb610 =

  src/common/utils.hpp 1d81e21 =

  src/exec/exec.cpp e8db407 =

  src/launcher/launcher.cpp a141b9a =

  src/local/local.hpp 55f9eaf =

  src/local/local.cpp affe432 =

  src/master/master.cpp 4dc9ee0 =

  src/messages/messages.proto 87e1548 =

  src/sched/sched.cpp dcadb10 =

  src/slave/constants.hpp f0c8679 =

  src/slave/isolation_module.hpp c896908 =

  src/slave/lxc_isolation_module.hpp b7beefe =

  src/slave/lxc_isolation_module.cpp 66a2a89 =

  src/slave/main.cpp 85cba25 =

  src/slave/process_based_isolation_module.hpp f6f9554 =

  src/slave/process_based_isolation_module.cpp 2b37d42 =

  src/slave/slave.hpp 279bc7b =

  src/slave/slave.cpp 3358ec4 =

  src/tests/fault_tolerance_tests.cpp 6772daf =

  src/tests/slave_restart_tests.cpp PRE-CREATION =

  src/tests/utils.hpp e81ec82 =


Diff: https://reviews.apache.org/r/4462/diff


Testing
-------

make check.

Note that only the new test in tests/slave_restart_tests.cpp  engages in re=
covery!

Recovery is disabled for old tests (though they still checkpoint relevant i=
nfo!)


Thanks,

Vinod


--===============2180678955089637090==--