You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Shawn Heisey (JIRA)" <ji...@apache.org> on 2014/03/28 17:08:16 UTC

[jira] [Commented] (SOLR-5930) Add Threshold counts capability to PingRequestHandler

    [ https://issues.apache.org/jira/browse/SOLR-5930?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13950947#comment-13950947 ] 

Shawn Heisey commented on SOLR-5930:
------------------------------------

I filed an issue once upon a time for a distributed PingRequestHandler.  I ended up not needing it, though.  Here's what I do right now, which works really well:

My distributed index uses what I call a broker core.  It's a core with the same schema and mostly the same config as the other cores, but an empty index.  All of the search handler definitions in solrconfig.xml include the shards parameter.  With this, the client doesn't need to know that the index is distributed -- it just makes a request to what it thinks is a single index.

Because the ping handler just uses one of the local search handlers, it automatically does the right thing, whether it's a broker core or a core with an index.  The load balancer health check is wired to the broker cores on each host, so it will only send queries when the entire index is working.


> Add Threshold counts capability to PingRequestHandler
> -----------------------------------------------------
>
>                 Key: SOLR-5930
>                 URL: https://issues.apache.org/jira/browse/SOLR-5930
>             Project: Solr
>          Issue Type: Improvement
>            Reporter: Kranti Parisa
>            Priority: Minor
>              Labels: pingrequest
>
> PingRequestHandler in solrconfig.xml for doing threshold checks say min no of documents to be presented in the index. we can specify the *:* in ping query but then if that returns 0 documents, we should be able to say it's not healthy.
> extension:
> options to configure  PingRequestHandler for the entire solr instance (in multi-core environment). all the cores should be healthy in order to say the instance is healthy.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

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