You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by "Kevin Risden (JIRA)" <ji...@apache.org> on 2019/03/19 11:31:00 UTC
[jira] [Updated] (SOLR-11194) Suggester BlendedInfixLookupFactory
to support HDFS store for indexwriter
[ https://issues.apache.org/jira/browse/SOLR-11194?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Kevin Risden updated SOLR-11194:
--------------------------------
Component/s: hdfs
Hadoop Integration
> Suggester BlendedInfixLookupFactory to support HDFS store for indexwriter
> -------------------------------------------------------------------------
>
> Key: SOLR-11194
> URL: https://issues.apache.org/jira/browse/SOLR-11194
> Project: Solr
> Issue Type: Improvement
> Security Level: Public(Default Security Level. Issues are Public)
> Components: Hadoop Integration, hdfs, Suggester
> Affects Versions: 6.6
> Reporter: Peter Szantai-Kis
> Priority: Minor
>
> When relative path is used for indexPath in the LookupFactory, the solr cores DataDir is used as a prefix, but still will be created in the working directory of the process. Unfortunately both the BlendedInfixLookupFactory and it's parent AnalyzingInfixLookupFactory uses a store implementation that does not support HDFS. So in case the data dir is on hdfs then the "hdfs:/" prefix will be ignored and because of the missing leading forward slash it will be treated as a relative path. It seems hdfs wasn't really considered as an option when these Suggesters were implemented.
> One possible solution is to use the SolrCores' own DirectoryFactory to lookup the data dir which dynamically will return the proper store in case of HDFS is used
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org