[
https://issues.apache.org/jira/browse/SOLR-8374?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15042452#comment-15042452
]
Erick Erickson commented on SOLR-8374:
--------------------------------------
How is the ping request generated? This is probably the result of "df" field in
the configs that's referenced _only_ when there is no field-qualifier in the
incoming query. So I believe that if whoever generates the ping query issues a
query, say id:* rather than just a bare term this isn't a problem.
Pulling _text_ out of the df would affect all sorts of other behavior so I'm
not sure we want to do that. Perhaps define an explicit ping request handler
instead?
> Issue with _text_ field in schema file
> --------------------------------------
>
> Key: SOLR-8374
> URL: https://issues.apache.org/jira/browse/SOLR-8374
> Project: Solr
> Issue Type: Bug
> Components: Hadoop Integration
> Affects Versions: 5.2.1
> Reporter: Romit Singhai
> Priority: Critical
> Labels: patch
>
> In the data_driven_schema_configs, the warning say that _text_ field can be
> removed if not needed. The hadoop indexer fails to index data as ping
> command could not find the collection required for indexing.
> The ping command for collection needs to be fixed (making _text_ optional) as
> _text_ add significantly to index size even if not needed.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
