[
https://issues.apache.org/jira/browse/SOLR-9448?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mikhail Khludnev updated SOLR-9448:
-----------------------------------
Description:
If subquery collection has a different unique key field name (let's say
{{foo_id}} is different to {{id}} in primary collection), you've got
{{NullPoniterException}} from {{QueryComponent.mergeIds()}}. To accommodate the
difference between uniqueKey field names between collections, add the following
parameters {{foo.fl=id:foo_id&foo.distrib.singlePass=true}}. The former one
renames uniqueKey field, the later switches to single pass search. There is no
way to rename field in default processing.
In a rare case when a collection under subquery has no unqueKey at all, it
leads to something like
was:
h2.UPD
The sentences below seems not really actual. The more objective synopsis is
described in the first comment. It seems like fl=foo:\[subquery]&collection=bar
can be fixed just by declaring fields in schema.
h3. Old description
straightforward \[subquery] implementation executes requests on a caller
collection, but just hitting another one with
{{caller/select?q=..&collection=callee}}. The problem is that for
{{GET_FIELDS}} it uses uniqKey from a caller collection but not a callee one.
Another observation, at that case both single sharded collections are
collocated at the same instance. Then, subquery can't be parsed if it queries a
field which are absent in caller schema. All of this seems pretty strange like
hitting an edge case.
h2. workaround
Perhaps you can collocate secondary index and call it {{fromIndex=callee}}.
Or you can name uniqKey the same, keeping the different app semantic.
> [subquery] calls another collection fails with "undefined field" or NPE from
> mergeIds
> -------------------------------------------------------------------------------------
>
> Key: SOLR-9448
> URL: https://issues.apache.org/jira/browse/SOLR-9448
> Project: Solr
> Issue Type: Bug
> Security Level: Public(Default Security Level. Issues are Public)
> Components: SolrCloud
> Affects Versions: 6.1, 6.2
> Reporter: Mikhail Khludnev
> Assignee: Mikhail Khludnev
> Attachments: SOLR-9448.patch, SOLR-9448.patch, SOLR-9448.patch,
> SOLR-9448.patch
>
>
> If subquery collection has a different unique key field name (let's say
> {{foo_id}} is different to {{id}} in primary collection), you've got
> {{NullPoniterException}} from {{QueryComponent.mergeIds()}}. To accommodate
> the difference between uniqueKey field names between collections, add the
> following parameters {{foo.fl=id:foo_id&foo.distrib.singlePass=true}}. The
> former one renames uniqueKey field, the later switches to single pass search.
> There is no way to rename field in default processing.
> In a rare case when a collection under subquery has no unqueKey at all, it
> leads to something like
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
