[
https://issues.apache.org/jira/browse/SOLR-7689?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14588101#comment-14588101
]
Yonik Seeley edited comment on SOLR-7689 at 6/16/15 2:37 PM:
-------------------------------------------------------------
This would definitely be a bug if we actually do use a rewritten query for any
caches. SolrIndexSearcher certainly didn't do that in the past, but it's
possible that a bug was introduced somewhere along the way.
Is this because some other code passes SolrIndexSearcher a rewritten query, or
do you think there's a bug internal to SolrIndexSearcher?
edit: I reviewed SolrIndexSearcher, and nothing jumps out at me as incorrect.
We should not clone the queries in general though (in SolrIndexSearcher) -
that's a lot of overhead and the wrong fix. If there's a bad-egg query
somewhere that is not effectively immutable (i.e. using it in a search can
change it) then that's what should be fixed.
was (Author: [email protected]):
This would definitely be a bug if we actually do use a rewritten query for any
caches. SolrIndexSearcher certainly didn't do that in the past, but it's
possible that a bug was introduced somewhere along the way.
Is this because some other code passes SolrIndexSearcher a rewritten query, or
do you think there's a bug internal to SolrIndexSearcher?
> QueryResultKey saved in cache uses the query after rewrite which causes cache
> misses
> ------------------------------------------------------------------------------------
>
> Key: SOLR-7689
> URL: https://issues.apache.org/jira/browse/SOLR-7689
> Project: Solr
> Issue Type: Bug
> Components: search, SearchComponents - other
> Reporter: Emad Nashed
>
> In SolrIndexSearcher class, the key used to lookup results in
> queryResultCache uses the original query.
> However later in createNormalizedWeight the query gets re-written, and then
> saved in the queryResultCache after it's re-written.
> This causes cache misses for the same query, and un-necessary inserts in the
> queryResultCache.
> I can reproduce this using a re-ranking query that is using a main query as a
> dismax query, the dismax Query could be re-written into a TermQuery, which
> makes sense, but will cause cache misses.
> I tested a quick solution by just using q.clone() when it comes to build
> QueryResultKey, and it works fine, but not sure if that is the best way of
> doing it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]