Thx guys for your ideas. I'll test and let you know.
Regards,
On Feb 3, 2017 22:31, "Allison, Timothy B." wrote:
> This is a Tika/POI problem. Please download tika-app 1.14 [1] or a
> nightly version of Tika [2] and run
>
> java -jar tika-app.jar
>
> If the problem is fixed, we'll try to upgr
On 2/3/2017 2:56 PM, Shawn Heisey wrote:
> Solr 6.3, not in cloud mode.
> The admin UI has drawn the large segment such that it looks like
> almost the entire segment is deleted docs, which clearly isn't the
> case, as the hovertext shows.
I've got some different servers that are running 5.3.2-S
Solr 6.3, not in cloud mode.
Here is a screenshot of the problematic UI:
https://www.dropbox.com/s/aehcgf20zb0du01/solr-6.3-segments.png?dl=0
The admin UI has drawn the large segment such that it looks like almost
the entire segment is deleted docs, which clearly isn't the case, as the
hovertext
This is a Tika/POI problem. Please download tika-app 1.14 [1] or a nightly
version of Tika [2] and run
java -jar tika-app.jar
If the problem is fixed, we'll try to upgrade dependencies in Solr. If it
isn't fixed, please open a bug on Tika's Jira.
If this is a missing bean issue (sorry, I c
Hi Alexandre,
A boost should works for my needs.
Thanks!
Alexandre Drouin
-Original Message-
From: Alexandre Rafalovitch [mailto:arafa...@gmail.com]
Sent: February 3, 2017 12:14 PM
To: solr-user
Subject: Re: Promote documents if they are present in the search results
Importance: High
On 2/3/2017 10:12 AM, Suresh Pendap wrote:
> Will Solrj client 4.10.3 version work with Solr 6.3 version of the
> server? I was trying to look up the documentation but no where the
> compatibility matrix between server and client is provided. Has some
> one already used this combination?
If it's H
I was cleaning up a replica on a dead node, so active/down was what I needed.
Leader seems like it could be useful.
wunder
Walter Underwood
wun...@wunderwood.org
http://observer.wunderwood.org/ (my blog)
> On Feb 3, 2017, at 10:18 AM, Erick Erickson wrote:
>
> I thought (but can't check now)
Is this a known bug?
Open the stack trace for a log line. When it refreshes, the log trace is
closed. Formatting is too wide, too. This makes the log display unusable. I
have to switch to the old UI to look at logs.
wunder
Walter Underwood
wun...@wunderwood.org
http://observer.wunderwood.org/
The plugin itself is responsible for returning information about
itself via an overridden method IIRC, so you have control over what
version is reported.
As for the other, a slight variant on King's process would be to put
your custom jars in a different directory then used the
directive in solrc
I thought (but can't check now) that there were some filters in one of
the views for "only down nodes", "only recovering nodes" and the like.
Don't remember "only leaders" though
On Fri, Feb 3, 2017 at 10:09 AM, Walter Underwood wrote:
> It would be great if the collection-overview page showed wh
It would be great if the collection-overview page showed which cores are
leaders and active/down without opening the shard info. It is quite a treasure
hunt to find the down shard in a big collection.
As far as I can tell, this is the only place to get the replica name needed for
the DELETEREPL
Could you do a basic boost query with a list of all those document
ids. If they are in results, their relevance is boosted. If they are
not, they are just ignored.
You could have that boost query configured in the solrconfig.xml or
passed in by a client.
Regards,
Alex
http://www.solr-star
Hi,
Will Solrj client 4.10.3 version work with Solr 6.3 version of the server? I
was trying to look up the documentation but no where the compatibility matrix
between server and client is provided.
Has some one already used this combination?
Regards
Suresh
Hello,
I am developing a new feature in my product where an administrator can select a
list of documents for different sections of the application. You can think of
this feature as the classic categorization in an e-commerce site (e.g.: Boots,
Headgear, etc). Since this categorization is diff
Howdy,
In the Solr Wiki I stumbled upon a somewhat vague statement on the
uniqueKey:
> https://wiki.apache.org/solr/SchemaXml#The_Unique_Key_Field
> It shouldn't matter whether you rename this to something else (and
change the value), but occasionally it has in the past. We
recommend that y
Solr _does_ have a query parser that doesn't suffer from this problem --
SimpleQParser chosen as the string "simple".
https://cwiki.apache.org/confluence/display/solr/Other+Parsers#OtherParsers-SimpleQueryParser
In this case, see the "WHITESPACE" operator feature which can be toggled.
Configure to
What we ended up doing was creating separate directories for each version of a
plugin we had written, and in each collection's solrconfig,xml, we add the path
to the specific directory we wanted that collection to use via the " On Feb 3, 2017, at 2:40 AM, Andrea Gazzarini wrote:
>
> Hi Zaccheo,
I have a partial answer. The Javadoc talks about deprecation:
https://lucene.apache.org/core/6_4_0/analyzers-common/org/apache/lucene/analysis/synonym/SynonymFilterFactory.html
So, we are now supposed to use Graph and then Flatten filters in the
configuration. But there was another email earlier o
This kind of information extraction comes from Apache Tika that is
shipped with Solr. However Solr does not ship every possible parser
with its installation. So, I think you are hitting Tika where it
manages to figure out what type of content you have, but does not have
(Apache POI - another O/S pr
I brought down the afflicted servers, waited 5 min, then brought them back up
very slowly. That fixed the problem. The bad shard was assigned a leader.
Great advice as usual.
Erick Erickson wrote
> Yes. If indexing went through you'd lose docs so indexing will fail.
> Querying will fail too unless
I get it: Schema components do not yet support the Blob Store.
thanks
On piatok, 3. februára 2017 10:28:27 CET Michal Hlavac wrote:
> Hi,
>
> it is possible to use BlobStore & Config API with enabled.runtime.lib=true to
> add custom token filters?
> I tried, but it doesn't work.
>
> 1. Uploade
Hi,
I'm using single core Solr 6.4 instance on windows server (windows server
2012 R2 standard),
Java v8, (build 1.8.0_121-b13).
All works more or less ok, except MS Visio vsdx files indexing.
Every time it throws an error (no matters if it tries to index vsdx file or
for example docx with vis
Hi Zaccheo,
I don't think this is possible, this is something related with the
classloader behavior, and even if there's a "priority" rule in the JVM,
I wouldn't rely on that in my application.
That could be good in a dev environment where you can specify the
"order" of the imported libraries (
Hi all,
I developed a custom DocTransformer that is loaded from a .jar in the core
"lib" directory. It works but I have now a problem with versioning:
1. if lib directory contains different versions of the same .jar which one
is loaded? I tried putting both myplugins-1.0.0.jar and myplugins-1.0.1.j
Hi,
it is possible to use BlobStore & Config API with enabled.runtime.lib=true to
add custom token filters?
I tried, but it doesn't work.
1. Uploaded jar lucene-analyzers-morfologik-6.4.0.jar file to blob store
.system with name lucene-analyzers-morfologik-6.4.0
2. Add runtime library {"add-ru
25 matches
Mail list logo
