lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Avanish Raju <>
Subject Re: Core property name ignored when creating collection using API
Date Mon, 02 Feb 2015 14:31:34 GMT
I had a discussion with @search_mb about this on IRC, and he explained how
my collection query would still work with "user6", though we couldn't
resolve why the solr Core Admin page doesn't show the collection name as "

Detailed chat log follows:

[05:43] (yaraju) Hi everyone! Was wondering if anyone is around who can
help me learn if I'm missing something obvious on this thread....
[05:43] (yaraju)
[05:46] (search_mb) @yaraju - you can have many shards/replicas on one node
for one core. So you have your collecetion named user6, but solr create
dirs for the purpose of multishards on the same machine so it sufixes your
name with <shard_id>_<replica_id>
[05:46] (search_mb) i don't know if it is customizable
[05:47] (yaraju) @search_mb thanks! I was able to set the instance dir and
[05:47] (yaraju) And data dir*
[05:47] (yaraju) However i can't set the collection name.
[05:48] (yaraju) The ip address mention in the thread is public, so you
should be able to take a look, if that helps
[05:53] (yaraju) @search_mb my main concern is, the collection name should
not sound like a shard/replica specific name. After all, a collection is a
logical unit over multiple shards/replicas, right? So under the solr admin
UI, when browsing the collections, why does it not show up as "user6"?
[05:56] (yaraju) Instead it shows up as
[05:58] (search_mb) Becase as an admin you wouldn't have possiblity to
distinguished on which shard/replica you are looking at
[05:58] (search_mb) @yaraju but you can make request ot user6 collection
like this*%3A*&wt=json&indent=true
[05:58] (search_mb) @yaraju and it will distribute your search over all
[06:00] (yaraju) @search_mb aah, i see. So i shouldn't really worry about
that name in admin UI. I was so assuming solr Admin UI would know that
Config through Zookeeper somehow
[06:01] (yaraju) Thanks so much for clarifying
[06:01] (search_mb) no problem :)
[06:02] (search_mb) @yaraju look at this one
[06:04] (yaraju) Thanks! Distributed requests across multiple nodes is the
main reason I'm trying to make collections.
[06:06] (yaraju) Btw why is it that collection1 shows up correctly on solr
admin? And not a shard/replica?
[06:08] (yaraju) Sorry... Meant to say, either way does it not matter how
it shows in solr admin?
[06:09] (search_mb) maybe because it is in "standalone" mode
[06:09] (search_mb) no shards/no replicas
[06:11] (yaraju) Oh. I believe it's 1 shard and 2 replicas
[06:11] (yaraju)
[06:14] (search_mb) @yaraju honestly I don't know :) I assume some
operations order take place here - collection1 was created first as a
standalone and then someone add shard and replicas But I'm not sure in this
[06:15] ** search_mb has quit (Quit: sry must go)

[06:16] *(*yaraju*)* Hmm, i see. Since Collection1 was part of the
quickstart example which i haven't changed, i can't imagine how that

[06:16] *(*yaraju*)* But thanks a lot!!

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message