You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Bill Bell (JIRA)" <ji...@apache.org> on 2010/08/13 05:49:16 UTC
[jira] Created: (SOLR-2044) Replication ping.jsp page for use with
Amazon EC2 and Multiple Cores
Replication ping.jsp page for use with Amazon EC2 and Multiple Cores
--------------------------------------------------------------------
Key: SOLR-2044
URL: https://issues.apache.org/jira/browse/SOLR-2044
Project: Solr
Issue Type: Bug
Components: replication (java)
Affects Versions: 1.4.1
Reporter: Bill Bell
Fix For: 1.4.2
Attachments: ping.jsp
When hosting SOLR at Amazon there is an issue with load balancing and replication. This is especially useful when you are using multiple slaves behind
the load balancer and a MASTER SOLR box. The Load Balancer at Amazon has no way of knowing that each SLAVE has data from the MASTER replicated. As soon as the instance starts,
Amazon starts sending requests to the Slaves. You need to have the Slave running in order to get the files from the Master.
The solution is to add a ping.jsp page that Amazon hits on the Slave Tomcat/Jetty instances. The ping.jsp checks all cores to make sure they all have data, otherwise it returns a non-200 values which signals Amazon
to not send requests to the Slaves.
Put this into your solr.war file as /admin/replication/ping.jsp
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] Updated: (SOLR-2044) Replication ping.jsp page for use with
Amazon EC2 and Multiple Cores
Posted by "Bill Bell (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-2044?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Bill Bell updated SOLR-2044:
----------------------------
Attachment: ping.jsp
1.4.x file to ad to solr.war
> Replication ping.jsp page for use with Amazon EC2 and Multiple Cores
> --------------------------------------------------------------------
>
> Key: SOLR-2044
> URL: https://issues.apache.org/jira/browse/SOLR-2044
> Project: Solr
> Issue Type: Bug
> Components: replication (java)
> Affects Versions: 1.4.1
> Reporter: Bill Bell
> Fix For: 1.4.2
>
> Attachments: ping.jsp
>
>
> When hosting SOLR at Amazon there is an issue with load balancing and replication. This is especially useful when you are using multiple slaves behind
> the load balancer and a MASTER SOLR box. The Load Balancer at Amazon has no way of knowing that each SLAVE has data from the MASTER replicated. As soon as the instance starts,
> Amazon starts sending requests to the Slaves. You need to have the Slave running in order to get the files from the Master.
> The solution is to add a ping.jsp page that Amazon hits on the Slave Tomcat/Jetty instances. The ping.jsp checks all cores to make sure they all have data, otherwise it returns a non-200 values which signals Amazon
> to not send requests to the Slaves.
> Put this into your solr.war file as /admin/replication/ping.jsp
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] Commented: (SOLR-2044) Replication ping.jsp page for use
with Amazon EC2 and Multiple Cores
Posted by "Peter Wolanin (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-2044?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12919440#action_12919440 ]
Peter Wolanin commented on SOLR-2044:
-------------------------------------
There is already a "health check" functionality - seems like this might be combined? Or instead you might have an external script which does this same check and sets or removes the health check file?
> Replication ping.jsp page for use with Amazon EC2 and Multiple Cores
> --------------------------------------------------------------------
>
> Key: SOLR-2044
> URL: https://issues.apache.org/jira/browse/SOLR-2044
> Project: Solr
> Issue Type: Bug
> Components: replication (java)
> Affects Versions: 1.4.1
> Reporter: Bill Bell
> Fix For: 1.4.2
>
> Attachments: ping.jsp
>
>
> When hosting SOLR at Amazon there is an issue with load balancing and replication. This is especially useful when you are using multiple slaves behind
> the load balancer and a MASTER SOLR box. The Load Balancer at Amazon has no way of knowing that each SLAVE has data from the MASTER replicated. As soon as the instance starts,
> Amazon starts sending requests to the Slaves. You need to have the Slave running in order to get the files from the Master.
> The solution is to add a ping.jsp page that Amazon hits on the Slave Tomcat/Jetty instances. The ping.jsp checks all cores to make sure they all have data, otherwise it returns a non-200 values which signals Amazon
> to not send requests to the Slaves.
> Put this into your solr.war file as /admin/replication/ping.jsp
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
[jira] Commented: (SOLR-2044) Replication ping.jsp page for use
with Amazon EC2 and Multiple Cores
Posted by "Bill Bell (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/SOLR-2044?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12919442#action_12919442 ]
Bill Bell commented on SOLR-2044:
---------------------------------
True it could be combined with a parameter like "?replication=true" to indicate that the ping returns true if replication has completed.
> Replication ping.jsp page for use with Amazon EC2 and Multiple Cores
> --------------------------------------------------------------------
>
> Key: SOLR-2044
> URL: https://issues.apache.org/jira/browse/SOLR-2044
> Project: Solr
> Issue Type: Bug
> Components: replication (java)
> Affects Versions: 1.4.1
> Reporter: Bill Bell
> Fix For: 1.4.2
>
> Attachments: ping.jsp
>
>
> When hosting SOLR at Amazon there is an issue with load balancing and replication. This is especially useful when you are using multiple slaves behind
> the load balancer and a MASTER SOLR box. The Load Balancer at Amazon has no way of knowing that each SLAVE has data from the MASTER replicated. As soon as the instance starts,
> Amazon starts sending requests to the Slaves. You need to have the Slave running in order to get the files from the Master.
> The solution is to add a ping.jsp page that Amazon hits on the Slave Tomcat/Jetty instances. The ping.jsp checks all cores to make sure they all have data, otherwise it returns a non-200 values which signals Amazon
> to not send requests to the Slaves.
> Put this into your solr.war file as /admin/replication/ping.jsp
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org