Mailing-List: contact mesos-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: mesos-dev@incubator.apache.org
Date: Tue, 27 Mar 2012 05:39:34 +0000 (UTC)
From: "jiraposter@reviews.apache.org (Commented) (JIRA)" <jira@apache.org>
To: mesos-dev@incubator.apache.org
Message-ID: 
 <706186313.21869.1332826774733.JavaMail.tomcat@hel.zones.apache.org>
In-Reply-To: 
 <875053756.33641.1321409092750.JavaMail.tomcat@hel.zones.apache.org>
Subject: [jira] [Commented] (MESOS-86) Expose master url to the scheduler
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/MESOS-86?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13239198#comment-13239198 ] 

jiraposter@reviews.apache.org commented on MESOS-86:
----------------------------------------------------


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/3083/
-----------------------------------------------------------

(Updated 2012-03-27 05:39:13.559518)


Review request for mesos, Benjamin Hindman and John Sirois.


Changes
-------

swapped the arguments of frameworkmessage and sendframeworkmessage to be consistent.

added an optional SlaveID field to SlaveInfo


Summary
-------

    New Scheduler callbacks for master info (ip and port)
    
    --> registered() callback now includes maste info 
    --> new re-registered() callback is called when a new master is elected
    --> new masterLost() callback is called when mesos core doesnt detect an
    active master (e.g: during election)
   --> new executorLost() callback

New Executor callbacks
 --> reregistered() callback
--> slaveLost() callback


This addresses bug MESOS-86.
    https://issues.apache.org/jira/browse/MESOS-86


Diffs (updated)
-----

  include/mesos/executor.hpp ef7bdf9 
  include/mesos/mesos.proto 23aad17 
  include/mesos/scheduler.hpp 5bcf065 
  src/common/values.cpp 09d162f 
  src/examples/java/TestExceptionFramework.java d15b132 
  src/examples/java/TestFramework.java f86646b 
  src/examples/java/TestMultipleExecutorsFramework.java cdbcc48 
  src/examples/long_lived_executor.cpp 4b4abad 
  src/examples/long_lived_framework.cpp 2775f5a 
  src/examples/no_executor_framework.cpp fb31b25 
  src/examples/python/test_framework.py 0d71831 
  src/examples/test_executor.cpp ba54004 
  src/examples/test_framework.cpp 622f6ac 
  src/exec/exec.cpp 060707ad 
  src/java/jni/convert.cpp 5c2bf21 
  src/java/jni/org_apache_mesos_MesosExecutorDriver.cpp 579ebbe 
  src/java/jni/org_apache_mesos_MesosSchedulerDriver.cpp 547d54b 
  src/java/src/org/apache/mesos/MesosSchedulerDriver.java 129a108 
  src/java/src/org/apache/mesos/Scheduler.java b01468b 
  src/java/src/org/apache/mesos/SchedulerDriver.java 0638ee3 
  src/launcher/executor.cpp da4203c 
  src/master/master.cpp 6110c17 
  src/messages/messages.proto 7f9cffe 
  src/python/native/mesos_scheduler_driver_impl.cpp 9e088ef 
  src/python/native/proxy_executor.hpp 32df90b 
  src/python/native/proxy_executor.cpp d453406 
  src/python/native/proxy_scheduler.hpp d0e5dfd 
  src/python/native/proxy_scheduler.cpp 9717f44 
  src/python/src/mesos.py 25d762b 
  src/sched/sched.cpp 9440381 
  src/tests/exception_tests.cpp c14bacc 
  src/tests/fault_tolerance_tests.cpp 9e7d181 
  src/tests/master_tests.cpp 57db468 
  src/tests/resource_offers_tests.cpp a405888 
  src/tests/utils.hpp 9241567 

Diff: https://reviews.apache.org/r/3083/diff


Testing
-------

make check


Thanks,

Vinod


> Expose master url to the scheduler
> ----------------------------------
>
>                 Key: MESOS-86
>                 URL: https://issues.apache.org/jira/browse/MESOS-86
>             Project: Mesos
>          Issue Type: New Feature
>            Reporter: Vinod Kone
>            Assignee: Vinod Kone
>
> Introduce new / augment old api calls (callbacks) to expose the master url.
> Currently there is no way for the scheduler to know what the leading master is. It would be useful for scheduler's to know this information (e.g: pointing to the webui of the leading master). Here is the plan:
> 1) Augment registered() callback to included this info
> 2) Implement new callback re-registered() that gets called whenever a master re-election happens. This also contains the new master url.
> 3) Also implement a new callback masterLost()  that gets called whenever the current master loses election.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira