Re: job.coordinator.replication.factor configuration

[Yi Pan]

JIRA opened: SAMZA-2460.

Thanks!

-Yi

On Sat, Feb 15, 2020 at 9:42 AM Yi Pan  wrote:

> Hi, Robert,
>
> Thanks for pointing this out. This is an error on the documentation
> side.The meaning of job.coordinator.replication.factor should not change
> post 1.0. I will file a bug against the doc site.
>
> Thanks!
>
> -Yi
>
> On Fri, Feb 14, 2020 at 7:52 AM Robert Wigginton
>  wrote:
>
>> Hello,
>>
>>
>>
>> Could you verify the description of the
>> job.coordinator.replication.factor property?  Documentation is different
>> for releases prior to 1.0.0 and I would like to confirm if the property
>> truly did change so we can set that property correctly.
>>
>>
>>
>> <= 1.0.0 Description
>>
>>
>>
>> job.coordinator.replication.factor
>>
>> If you are using Kafka for coordinator stream, this is the number of
>> Kafka nodes to which you want the coordinator topic replicated for
>> durability.
>>
>> > 1.0.0 Description
>>
>>
>>
>> job.coordinator.replication.factor
>>
>>
>>
>> The frequency at which the input streams’ partition count change should
>> be detected. When the input partition count change is detected, Samza will
>> automatically restart a stateless job or fail a stateful job. A longer time
>> interval is recommended for jobs w/ large number of input system stream
>> partitions, since gathering partition count may incur measurable overhead
>> to the job. You can completely disable partition count monitoring by
>> setting this value to 0 or a negative integer, which will also disable
>> auto-restart/failing behavior of a Samza job on partition count changes.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> [image: Helix Education] 
>>
>>
>>
>> *Robert Wigginton*
>>
>> Sr. Systems Engineer
>>
>> Denver, Colorado
>>
>> C: 303-304-4914
>>
>> www.helixeducation.com
>>
>> Blog  | Twitter
>>  | Facebook
>>  | LinkedIn
>> 
>>
>>
>>
>

Re: job.coordinator.replication.factor configuration

[Yi Pan]

Hi, Robert,

Thanks for pointing this out. This is an error on the documentation
side.The meaning of job.coordinator.replication.factor should not change
post 1.0. I will file a bug against the doc site.

Thanks!

-Yi

On Fri, Feb 14, 2020 at 7:52 AM Robert Wigginton
 wrote:

> Hello,
>
>
>
> Could you verify the description of the job.coordinator.replication.factor
> property?  Documentation is different for releases prior to 1.0.0 and I
> would like to confirm if the property truly did change so we can set that
> property correctly.
>
>
>
> <= 1.0.0 Description
>
>
>
> job.coordinator.replication.factor
>
> If you are using Kafka for coordinator stream, this is the number of Kafka
> nodes to which you want the coordinator topic replicated for durability.
>
> > 1.0.0 Description
>
>
>
> job.coordinator.replication.factor
>
>
>
> The frequency at which the input streams’ partition count change should be
> detected. When the input partition count change is detected, Samza will
> automatically restart a stateless job or fail a stateful job. A longer time
> interval is recommended for jobs w/ large number of input system stream
> partitions, since gathering partition count may incur measurable overhead
> to the job. You can completely disable partition count monitoring by
> setting this value to 0 or a negative integer, which will also disable
> auto-restart/failing behavior of a Samza job on partition count changes.
>
>
>
>
>
>
>
>
>
> [image: Helix Education] 
>
>
>
> *Robert Wigginton*
>
> Sr. Systems Engineer
>
> Denver, Colorado
>
> C: 303-304-4914
>
> www.helixeducation.com
>
> Blog  | Twitter
>  | Facebook
>  | LinkedIn
> 
>
>
>

job.coordinator.replication.factor configuration

[Robert Wigginton]

Hello,

Could you verify the description of the job.coordinator.replication.factor 
property?  Documentation is different for releases prior to 1.0.0 and I would 
like to confirm if the property truly did change so we can set that property 
correctly.

<= 1.0.0 Description

job.coordinator.replication.factor
If you are using Kafka for coordinator stream, this is the number of Kafka 
nodes to which you want the coordinator topic replicated for durability.
> 1.0.0 Description

job.coordinator.replication.factor

The frequency at which the input streams’ partition count change should be 
detected. When the input partition count change is detected, Samza will 
automatically restart a stateless job or fail a stateful job. A longer time 
interval is recommended for jobs w/ large number of input system stream 
partitions, since gathering partition count may incur measurable overhead to 
the job. You can completely disable partition count monitoring by setting this 
value to 0 or a negative integer, which will also disable auto-restart/failing 
behavior of a Samza job on partition count changes.




[Helix Education]

Robert Wigginton
Sr. Systems Engineer
Denver, Colorado

C: 303-304-4914

www.helixeducation.com

Blog | 
Twitter | 
Facebook | 
LinkedIn