[ 
https://issues.apache.org/jira/browse/MAPREDUCE-5329?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13781364#comment-13781364
 ] 

Avner BenHanoch commented on MAPREDUCE-5329:
--------------------------------------------

Hi Siddharth,

All sounds great!  Thank you.

One very last comment, regarding 
*mapreduce.job.shuffle.provider.plugin.classes*, can we change it to be 
*mapreduce.job.shuffle.provider.services* and list the *service names* instead 
of the *service classes*.  This will prevent confusion in the config name of 
hadoop-1 vs. hadoop-2 which is good for me.  Also, this will be good for your 
concern about services that are not running on the specific NM.

In this case the code will simply be something like this:
{code}
// add external shuffle-providers - if any
Collection<String> shuffleProviders = conf.getStringCollection(
    MRJobConfig.SHUFFLE_PROVIDER_SERVICES);

for (final String provider : shuffleProviders) {
  if (provider.equals(ShuffleHandler.MAPREDUCE_SHUFFLE_SERVICEID)) {
    continue; // skip built-in shuffle-provider that was already inserted with 
shuffle secret key
  }
  LOG.info("Adding " + provider + " to serviceData");
  // Please note, the shuffleProvider needs to be able to work with the 
host:port information provided
  // by the AM (i.e. plugins which require custom location / other 
configuration are not supported)
  serviceData.put(provider, ByteBuffer.allocate(0)); // This only serves for 
INIT_APP notifications
}
{code}


> APPLICATION_INIT is never sent to AuxServices other than the builtin 
> ShuffleHandler
> -----------------------------------------------------------------------------------
>
>                 Key: MAPREDUCE-5329
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-5329
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: mr-am
>    Affects Versions: 2.1.0-beta, 2.0.6-alpha
>            Reporter: Avner BenHanoch
>             Fix For: trunk
>
>         Attachments: MAPREDUCE-5329.patch
>
>
> APPLICATION_INIT is never sent to AuxServices other than the built-in 
> ShuffleHandler.  This means that 3rd party ShuffleProvider(s) will not be 
> able to function, because APPLICATION_INIT enables the AuxiliaryService to 
> map jobId->userId. This is needed for properly finding the MOFs of a job per 
> reducers' requests.
> NOTE: The built-in ShuffleHandler does get APPLICATION_INIT events due to 
> hard-coded expression in hadoop code. The current TaskAttemptImpl.java code 
> explicitly call: serviceData.put (ShuffleHandler.MAPREDUCE_SHUFFLE_SERVICEID, 
> ...) and ignores any additional AuxiliaryService. As a result, only the 
> built-in ShuffleHandler will get APPLICATION_INIT events.  Any 3rd party 
> AuxillaryService will never get APPLICATION_INIT events.
> I think a solution can be in one of two ways:
> 1. Change TaskAttemptImpl.java to loop on all Auxiliary Services and register 
> each of them, by calling serviceData.put (…) in loop.
> 2. Change AuxServices.java similar to the fix in: MAPREDUCE-2668  
> "APPLICATION_STOP is never sent to AuxServices".  This means that in case the 
> 'handle' method gets APPLICATION_INIT event it will demultiplex it to all Aux 
> Services regardless of the value in event.getServiceID().
> I prefer the 2nd solution.  I am welcoming any ideas.  I can provide the 
> needed patch for any option that people like.
> See [Pluggable Shuffle in Hadoop 
> documentation|http://hadoop.apache.org/docs/current/hadoop-mapreduce-client/hadoop-mapreduce-client-core/PluggableShuffleAndPluggableSort.html]



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Reply via email to