You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by Parker Johnson <Pa...@gap.com> on 2011/04/11 20:42:29 UTC

Will Slaves Pileup Replication Requests?

What is the slave replication behavior if a replication request to pull
indexes takes longer than the replication interval itself?

Anotherwords, if my replication interval is set to be every 30 seconds,
and my indexes are significantly large enough to take longer than 30
seconds to transfer, is the slave smart enough to not send another
replication request if one is already in progress?


-Parker



Re: Will Slaves Pileup Replication Requests?

Posted by Parker Johnson <Pa...@gap.com>.
Thanks Larry.

-Parker

On 4/11/11 12:14 PM, "Green, Larry (CMG - Digital)"
<La...@cmgdigital.com> wrote:

>Yes. It will wait whatever the replication interval is after the most
>recent replication completes before attempting again.
>
>On Apr 11, 2011, at 2:42 PM, Parker Johnson wrote:
>
>> 
>> What is the slave replication behavior if a replication request to pull
>> indexes takes longer than the replication interval itself?
>> 
>> Anotherwords, if my replication interval is set to be every 30 seconds,
>> and my indexes are significantly large enough to take longer than 30
>> seconds to transfer, is the slave smart enough to not send another
>> replication request if one is already in progress?
>> 
>> 
>> -Parker
>> 
>> 
>
>



Re: Will Slaves Pileup Replication Requests?

Posted by "Green, Larry (CMG - Digital)" <La...@cmgdigital.com>.
Yes. It will wait whatever the replication interval is after the most recent replication completes before attempting again.

On Apr 11, 2011, at 2:42 PM, Parker Johnson wrote:

> 
> What is the slave replication behavior if a replication request to pull
> indexes takes longer than the replication interval itself?
> 
> Anotherwords, if my replication interval is set to be every 30 seconds,
> and my indexes are significantly large enough to take longer than 30
> seconds to transfer, is the slave smart enough to not send another
> replication request if one is already in progress?
> 
> 
> -Parker
> 
>