Hi ,
I wanted to sort the results of the solr search query on the number of
highlights generated per document.
e.g
Doc 1
highlights {
fieldA
FieldB
}
Doc 2
Highlights{
field A
fieldC
fieldC
}
No, I understand that score is calculated depending on many factors like ,
tf , idf boost etc. So wh
And you will have "loadbalancing" since a "random" of the replica behind
the shard will be chosen to handle the query.
On 1/6/13 3:10 AM, Otis Gospodnetic wrote:
Hi Hassan,
Correct. If you have a single shard, then the query will execute the query
on only one node and that is it.
Otis
--
Solr
Erick,
I run 2 solr instances each on serverA and 2 on serverB.
Server A : 8983(leader), 8900(replica)
ServerB : 7574 (leader), 7500 (replica).
Please find the command below in the order in which they are started. Also i
am using single embedded zookeeper 9983 for all servers.
*serverA*
cd /
We made some cache config changes, That is when we noticed incomplete
replicas. We also bootstrap the configuration from script every time server
restarts.
Would cache config changes cause any issue with SolrCloud replication?
mostly when different nodes have different config(cache setting in this
The exception "No live SolrServers" is being thrown when trying to create a new
Collection ( code at end of this mail). On the CloudSolrServer request method,
we have this line
"ClientUtils.appendMap(coll, slices, clusterState.getSlices(coll));" where
"coll" is the new collection I am trying to
Does solr have any built in functionality to read application properties?
As an example, our "application.properties" file has some environment
specific configs (e.g. database urls and credentials) which I would like to
use in data-config.xml. If possible, we would like to avoid maintaining
multip
On 1/6/2013 9:14 AM, Shawn Heisey wrote:
I have now reduced my Eden size from about 4GB to under 256MB (max heap
is still 8GB), but the long pauses are still happening on my Solr 3.5.0
servers, and they are still ParNew collections.
Clarification of my question and my goals:
What I *want* is f
Well, if you indexed only one document, it will only be on one shard
determined by the hash of the , but that shouldn't give
the error you're seeing about live servers, and you should be seeing
the document no matter which server you query.
How exactly are you starting your servers?
On Sun, Jan
On 1/6/2013 9:14 AM, Shawn Heisey wrote:
What other information do I need to provide to track down what's happening?
Some additional info - part of a GC log showing a pause of over 8
seconds (second entry), followed by *much* shorter pauses:
2013-01-06T01:52:02.653-0700: 41837.979: [GC 41837
I filed SOLR-4191 because I was seeing some unusual behavior -
exceptions when accessing /admin/mbeans and /admin/ping. Initially I
thought it was happening while updates/commits were going on, but after
discovering the same thing happening in my production 3.5.0 servers as
well as branch_4x f
Hi Erick,
Yes i did update- i.e added a document using 8983. Ideally this update
should go to all 8983,8900,7500,7574. And i fired query using 7500.
Query is simple
http://serverB:7500/solr/collection1/select?q=*:*
But this give me error as mentioned in my previous post. No live server.
B
Hi,
I would like to store the document content into a single special field (non
indexed, stored only), and create several indexed copy fields (different
analysis applied).
During highlighting, the analysis definitions of the stored field are used,
so that improper or no highlighting is done.
Is the
Changing the schema has nothing to do with replication,
replication only deals with documents that have been
added/updated/deleted. So changing the schema to remove
the stopword filter should be fine as you've indicated
that you're OK with the old data still giving the old behavior.
Best
Erick
O
Actually, it's pretty loose. I recently saw a schema
where the copyfield was interspersed with
definitions in the tag, the code just uses
the DOM to collect all the leaf nodes sometimes.
There's been talk occasionally about creating a
schema/DTD/whatever but the general consensus
has been that p
This is a little confusing, could you post the exact
queries you're using? Because it looks like you're
firing the update to two servers. This is unnecessary,
you should send the update only once, to any
server in the cluster. Ditto with the query...
Best
Erick
On Sat, Jan 5, 2013 at 3:01 AM, ya
Not at this point, the limit is, I think, 100 documents.
I actually spoke imprecisely. Over that limit, an old-style
replication happens which _may_ cause a full index copy,
but usually will only move over the most recent segments
that have changed. If you're optimizing, this will
be the whole inde
It depends whether you use Solr 4.0 features whether there are changes
to the results. Personally I've found them pretty (surprisingly)
compatible from an API perspective.
One of the benefits of SolrCloud is that it can make multiple systems
operate as if they were one, so I can't immediately thin
>
> When I do a query without specifying the field, I would expect to search on
> the alias *var* but it searches on all those fields defined in *qf*.
When you pass qf parameter, edismax query searches all the fields mentioned
in that parameter. df comes into picture when you don't pass qf or you
No, *f.a.qf=a^3* is not a solution as I get the parsing error: "Field
aliases lead to a cycle"
2013/1/6 Juan Miguel Cejuela
> Hi,
>
> I have the following exemplified parameters in my edismax query:
>
> qf=a^3 x^2 y^1
> f.var.qf=x^2 y^1
> df=var
>
> When I do a query without specifying the field
Hi,
I have the following exemplified parameters in my edismax query:
qf=a^3 x^2 y^1
f.var.qf=x^2 y^1
df=var
When I do a query without specifying the field, I would expect to search on
the alias *var* but it searches on all those fields defined in *qf*. Is
this the expected behavior? I don't see
was this finally filled in a jira issue? I'm experiencing the same problem.
With default q.op=AND
"q:xxx zzz" behaves as expected --> "xxx AND zzz"
"q:(xxx AND yyy) zzz" behaves as expected --> "(xxx AND yyy) AND zzz
"q:(xxx OR yyy) zzz" changes the default operator to OR like --> "(xxx OR
yyy)
21 matches
Mail list logo
