lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jens Bo Rasmussen (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (SOLR-3420) Data is not replicated from replica slices to leader slices
Date Fri, 27 Apr 2012 12:35:49 GMT

     [ https://issues.apache.org/jira/browse/SOLR-3420?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Jens Bo Rasmussen updated SOLR-3420:
------------------------------------

    Description: 
Scenario: ensure that data is not lost when leader slices and replica slices switch roles
due to an untimely restart of both nodes
Given a SolrCloud cluster is running with
number of nodes: 2
number of slices per node: 1
number of documents in leader slice1_shard1 on first Solr node: 1
number of documents in replica slice1_shard2 on second Solr node: 1
number of documents in leader slice2_shard1 on first Solr node: 1
number of documents in replica slice2_shard2 on second Solr node: 1
And the second node is stopped
And x new documents are added to the first node
And the first node is stopped
When the second node is started and the slices become leader slices
And the first node is started and the slices become replica slices
Then I would expect that the SolrCloud cluster is running with
number of nodes: 2
number of slices per node: 1
number of documents in replica slice1_shard1 on first Solr node: 1+x
number of documents in leader slice1_shard2 on second Solr node: 1+x
number of documents in replica slice2_shard1 on first Solr node: 1+x
number of documents in leader slice2_shard2 on second Solr node: 1+x

Testresult: data is not replicated
number of servers (Solr nodes): 2
number of slices per node: 1
number of replica per slice: 1
number of documents in replica slice1_shard1 on first Solr node: 1+x
number of documents in leader slice2_shard1 on second Solr node: 1
number of documents in replica slice2_shard2 on first Solr node: 1+x
number of documents in leader slice1_shard2 on second Solr node: 1

==> data is not replicated from replica slices to leader slices

  was:
Scenario: ensure that data is not lost when leader slices and replica slices switch roles
due to an untimely restart of both nodes
Given a SolrCloud cluster is running with
number of nodes: 2
number of slices per node: 1
number of documents in leader slice1_shard1 on first Solr node: 1
number of documents in replica slice1_shard2 on second Solr node: 1
number of documents in leader slice2_shard1 on first Solr node: 1
number of documents in replica slice2_shard2 on second Solr node: 1
And the second node is stopped
And x new documents are added to the first node with leader slices
And the first node with leader slices is stopped
When the second node with replica slices is started and the replica slices become leader slices
And the first node with leader slices is started and the leader slices become replica slices
Then I would expect that the SolrCloud cluster is running with
number of nodes: 2
number of slices per node: 1
number of documents in replica slice1_shard1 on first Solr node: 1+x
number of documents in leader slice1_shard2 on second Solr node: 1+x
number of documents in replica slice2_shard1 on first Solr node: 1+x
number of documents in leader slice2_shard2 on second Solr node: 1+x

Testresult: data is not replicated
number of servers (Solr nodes): 2
number of slices per node: 1
number of replica per slice: 1
number of documents in replica slice1_shard1 on first Solr node: 1+x
number of documents in leader slice1_shard2 on second Solr node: 1
number of documents in replica slice2_shard1 on first Solr node: 1+x
number of documents in leader slice2_shard2 on second Solr node: 1

==> data is not replicated from replica slices to leader slices

    
> Data is not replicated from replica slices to leader slices
> -----------------------------------------------------------
>
>                 Key: SOLR-3420
>                 URL: https://issues.apache.org/jira/browse/SOLR-3420
>             Project: Solr
>          Issue Type: Bug
>          Components: replication (java)
>    Affects Versions: 4.0
>         Environment: Linux 2.6.32-29-server #58-Ubuntu SMP Fri Feb 11 21:06:51 UTC 2011
x86_64 GNU/Linux, Solr snapshot from 29th of February 2012, 2 Solr machines with 1 slice each
and 1 replica
>            Reporter: Jens Bo Rasmussen
>              Labels: replica
>             Fix For: 4.1
>
>   Original Estimate: 24h
>  Remaining Estimate: 24h
>
> Scenario: ensure that data is not lost when leader slices and replica slices switch roles
due to an untimely restart of both nodes
> Given a SolrCloud cluster is running with
> number of nodes: 2
> number of slices per node: 1
> number of documents in leader slice1_shard1 on first Solr node: 1
> number of documents in replica slice1_shard2 on second Solr node: 1
> number of documents in leader slice2_shard1 on first Solr node: 1
> number of documents in replica slice2_shard2 on second Solr node: 1
> And the second node is stopped
> And x new documents are added to the first node
> And the first node is stopped
> When the second node is started and the slices become leader slices
> And the first node is started and the slices become replica slices
> Then I would expect that the SolrCloud cluster is running with
> number of nodes: 2
> number of slices per node: 1
> number of documents in replica slice1_shard1 on first Solr node: 1+x
> number of documents in leader slice1_shard2 on second Solr node: 1+x
> number of documents in replica slice2_shard1 on first Solr node: 1+x
> number of documents in leader slice2_shard2 on second Solr node: 1+x
> Testresult: data is not replicated
> number of servers (Solr nodes): 2
> number of slices per node: 1
> number of replica per slice: 1
> number of documents in replica slice1_shard1 on first Solr node: 1+x
> number of documents in leader slice2_shard1 on second Solr node: 1
> number of documents in replica slice2_shard2 on first Solr node: 1+x
> number of documents in leader slice1_shard2 on second Solr node: 1
> ==> data is not replicated from replica slices to leader slices

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message