Incorrect Version number in Solr4.1 with Master/Repeater/Slave implementation

[adityab <ad...@yahoo.com>]

Hi we have just upgraded our Dev lab with Solr 4.1 with no Cloud. so our
implementation is like 
Master - Repeater - Slaves(2). In production we have large cluster so there
will be 8 slaves. 

What observed is 
1. Slave 1 replicating index from master show correct version number 
2. Slave 2 replicating index from Repeater show *incorrect *version number
Though data is correct as Repeater has index from Master. Why ? My guess is
the commit happening on Repeater after replication is increasing the version
number as repeater also a master. 
3. Is there any specific configuration required for Repeaters?

Any suggestions please.
Thanks

*Repeater :*
<http://lucene.472066.n3.nabble.com/file/n4043265/repeater.jpeg> 

*Slave 1 *(Replicated Index from above Repeater)
<http://lucene.472066.n3.nabble.com/file/n4043265/slave_repeater_%282%29.jpg> 

*Slaves 2* (Replicated Index from Master Directly)
<http://lucene.472066.n3.nabble.com/file/n4043265/slave_master2.jpg> 



--
View this message in context: http://lucene.472066.n3.nabble.com/Incorrect-Version-number-in-Solr4-1-with-Master-Repeater-Slave-implementation-tp4043265.html
Sent from the Solr - User mailing list archive at Nabble.com.

Re: Incorrect Version number in Solr4.1 with Master/Repeater/Slave implementation

[Mark Miller <ma...@gmail.com>]

Yes, you can have the same indexes and different versions. The version is basically a timestamp added to the index at commit time. If you just do a commit, you can avoid changing data, but update the version. So if the files look the same, you should be good.

- Mark

On Feb 26, 2013, at 9:25 PM, adityab <ad...@yahoo.com> wrote:

> thanks for the update Mark. looking fwd for this fix. Will try the trunk 4.x
> 
> Is it safe to assume that even with the incorrect version numbers the data
> on slave (replicated from Repeater) is same as whats on master ? At-least
> the files what i see in index dir are same. 
> 
> As per our schedule we are looking for March end to go with 4.1. as one of
> our major issue with So.r 3.5 is index size 40GB+ with 4.1 its 8GB :) 
> 
> 
> 
> --
> View this message in context: http://lucene.472066.n3.nabble.com/Incorrect-Version-number-in-Solr4-1-with-Master-Repeater-Slave-implementation-tp4043265p4043289.html
> Sent from the Solr - User mailing list archive at Nabble.com.

Re: Incorrect Version number in Solr4.1 with Master/Repeater/Slave implementation

[adityab <ad...@yahoo.com>]

thanks for the update Mark. looking fwd for this fix. Will try the trunk 4.x

Is it safe to assume that even with the incorrect version numbers the data
on slave (replicated from Repeater) is same as whats on master ? At-least
the files what i see in index dir are same. 

As per our schedule we are looking for March end to go with 4.1. as one of
our major issue with So.r 3.5 is index size 40GB+ with 4.1 its 8GB :) 



--
View this message in context: http://lucene.472066.n3.nabble.com/Incorrect-Version-number-in-Solr4-1-with-Master-Repeater-Slave-implementation-tp4043265p4043289.html
Sent from the Solr - User mailing list archive at Nabble.com.

Re: Incorrect Version number in Solr4.1 with Master/Repeater/Slave implementation

[Mark Miller <ma...@gmail.com>]

On Feb 26, 2013, at 8:17 PM, adityab <ad...@yahoo.com> wrote:

> the commit happening on Repeater after replication is increasing the version
> number as repeater

Yeah, there was a problem that is fixed in 4.2, and that commit no longer happens. A couple other little bugs around master->slave were also fixed (and new tests added!) I'm looking to drive a release of that in a couple weeks.

If you are able to setup a recent snapshot of 4X in your dev lab, that would be great and we would have time to fix any issue you found before 4.2 is pushed out.

- Mark