add "-Dmapred.map.tasks.speculative.execution=false
-Dmapred.reduce.tasks.speculative.execution=false " to PIG_OPTIONS

-D



On Tue, Feb 9, 2010 at 8:38 PM, prasenjit mukherjee
<[email protected]> wrote:
> Any writeup on the rule/conditions  when spec.exec. kicks in ? I
> searched the net and  found only some JIRA references. Looks like its
> based on many many factors etc. Still if I can get hold of some basic
> rules that'll be great.
>
> Thanks,
> Prasen
>
> On Wed, Feb 10, 2010 at 8:40 AM, Dmitriy Ryaboy <[email protected]> wrote:
>> turn off speculative execution.
>> but your map tasks should be idempotent. If they are not, rethink.
>> Speculative execution is a good thing (and so is preemption, it's
>> eviller twin).
>>
>> -D
>>
>> On Tue, Feb 9, 2010 at 6:52 PM, prasenjit mukherjee
>> <[email protected]> wrote:
>>> Any thoughts on this problem ? I am using a DEFINE command ( in PIG )
>>> and hence the actions are not idempotent. Because of which duplicate
>>> execution does have an affect on my results. Any way to overcome that
>>> ?
>>>
>>> On Tue, Feb 9, 2010 at 9:26 PM, prasenjit mukherjee
>>> <[email protected]> wrote:
>>>> But the second attempted job got killed even before the first one was
>>>> completed. How can we explain that.
>>>>
>>>> On Tue, Feb 9, 2010 at 7:38 PM, Eric Sammer <[email protected]> wrote:
>>>>> Prasen:
>>>>>
>>>>> This is most likely speculative execution. Hadoop fires up multiple
>>>>> attempts for the same task and lets them "race" to see which finishes
>>>>> first and then kills the others. This is meant to speed things along.
>>>>>
>>>>> Speculative execution is on by default, but can be disabled. See the
>>>>> configuration reference for mapred-*.xml.
>>>>>
>>>>> On 2/9/10 9:03 AM, prasenjit mukherjee wrote:
>>>>>> Sometimes for the same task I see that a duplicate task gets run on a
>>>>>> different machine and gets killed later. Not always but sometimes. Any
>>>>>> reason why duplicate tasks get run. I thought tasks are duplicated
>>>>>> only if  either the first attempt exits( exceptions etc ) or  exceeds
>>>>>> mapred.task.timeout. In this case none of them happens. As can be seen
>>>>>> from timestamp, the second attempt starts even though the first
>>>>>> attempt is still running ( only for 1 minute ).
>>>>>>
>>>>>> Any explanation ?
>>>>>>
>>>>>> attempt_201002090552_0009_m_000001_0
>>>>>>     /default-rack/ip-10-242-142-193.ec2.internal
>>>>>>     SUCCEEDED
>>>>>>     100.00%
>>>>>>     9-Feb-2010 07:04:37
>>>>>>     9-Feb-2010 07:07:00 (2mins, 23sec)
>>>>>>
>>>>>> attempt_201002090552_0009_m_000001_1
>>>>>>     Task attempt: /default-rack/ip-10-212-147-129.ec2.internal
>>>>>>     Cleanup Attempt: /default-rack/ip-10-212-147-129.ec2.internal
>>>>>>     KILLED
>>>>>>     100.00%
>>>>>>     9-Feb-2010 07:05:34
>>>>>>     9-Feb-2010 07:07:10 (1mins, 36sec)
>>>>>>
>>>>>>  -Prasen
>>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Eric Sammer
>>>>> [email protected]
>>>>> http://esammer.blogspot.com
>>>>>
>>>>
>>>
>>
>

Reply via email to