Does the new mailing list have moderators to deal with spam?
Cheers,
Ivan
Hi all
Recently we've had a few spam emails that have made it through Google's
filters, and there have been a calls for us to change to a
moderate-first-post policy. I am reluctant to adopt this policy for the
following
sure I'll be using discourse to great effect (and it seems to be
being used). So I could be completly wrong about all my points. :)
Doug
On Sunday, May 10, 2015, Ivan Brusic i...@brusic.com wrote:
I should have added something similar to what does expressed in his
last paragraph. My feedback
folks are very smart. Email truly can
become a first class experience and keep some of the great things about
discourse.
Cheers!
Doug
On Sunday, May 10, 2015, Ivan Brusic i...@brusic.com wrote:
I really do not care for the new mailing list.
First of all, I can no longer see real names
I am watching a few select categories with email notifications, but I still
received notifications for other categories, Logstash in my case.
Ivan
On May 4, 2015 6:12 PM, leslie.hawthorn leslie.hawth...@elastic.co
wrote:
Hello everyone,
We took in feedback on moving to a Discourse based forum
I might have found the conflicting setting.
On May 5, 2015 9:43 AM, Ivan Brusic i...@brusic.com wrote:
I am watching a few select categories with email notifications, but I
still received notifications for other categories, Logstash in my case.
Ivan
On May 4, 2015 6:12 PM, leslie.hawthorn
In non big data scenarios, having two servers for a database is simply
done to achieve high availability. Most databases use a master client
scenario, but Elasticsearch does not support such a setup. It really should
because not everyone has tons of data.
Ivan, not affiliated with the OP
On May
The field norm is computed at index time and is stored in a single byte,
which can lead to a loss in precision. This behavior might have changed
with newer versions of Lucene, but probably not.
Ivan
On Apr 30, 2015 6:42 PM, Fil ES lisowski.fili...@gmail.com wrote:
Hello,
I am experiencing an
Although the syntax is straight Lucene (query string query), I suspect that
Github and other sites parse the query term to create a format similar to
the one John mentioned.
Cheers,
Ivan
On May 1, 2015 1:22 AM, Peter Sorensen peter.jens.soren...@gmail.com
wrote:
Sorry for the vague title. If I
Your best option would be to write your own filter. It should be easy since
you have access to the source of the delimiter and length filters. Look at
the existing filter plugins for examples on how to deploy.
Ivan
On Apr 24, 2015 10:39 AM, Nassim nassim.ka...@gmail.com wrote:
Hi,
Is it
What kind of query are you executing? Are you query against a specific
field? A match query against the title field should work.
When using the analyze API, explicit state the field and not the analyzer
for more accurate behavior of what really goes on.
Cheers,
Ivan
On Apr 21, 2015 11:40 AM,
subforums.
Ivan
On Apr 15, 2015 7:21 PM, Leslie Hawthorn leslie.hawth...@elastic.co
wrote:
On Wed, Apr 15, 2015 at 9:02 AM, Ivan Brusic i...@brusic.com wrote:
I should clarify that I have no issues moving to Discourse, as long as
instantaneous email interaction is preserved, just wanted to point out
Which version are you using! The old post filter methods simply named
filter, should have been removed, or at least deprecated.
Cheers,
Ivan
On Apr 13, 2015 1:33 PM, James Green james.mk.gr...@gmail.com wrote:
Indeed. I had used postFilter to add my filters. The documentation for
filters
that impact your thoughts on moving to Discourse?
Folks, please keep the feedback coming!
Cheers,
LH
On Sat, Apr 11, 2015 at 12:09 AM, Ivan Brusic i...@brusic.com wrote:
As one of the oldest and most frequent users (before my sabbatical) of
the mailing list, I just wanted to say that I never had
Isn't the point of BM25 to use variable document length normalization? It
works when used on the entire index/corpus. It is meant to influence the TF
values.
Comparing results between Lucene queries is not advisable. Why did you
switch to BM25? Do you field lengths vary much?
Cheers,
Ivan
On
You understanding is correct. The former will be translated into a Lucene
phrase query, which uses the term doc positions to find matches.
Both query terms are analyzed, but the latter will simply be a bag-of-words
query, which ignores positions.
Cheers,
Ivan
On Apr 14, 2015 10:38 PM, Dave Reed
In my experience, the client can be older than the server.* The server side
code contains many version checks, so it should know how to handle requests
from older clients. The inverse is much harder to support since clients do
not change their requests based on the server.
* Between minor
As one of the oldest and most frequent users (before my sabbatical) of the
mailing list, I just wanted to say that I never had an issue with it. It
works. As long as I could continue using only email, I am happy.
For realtime communication, there is the IRC channel. If prefer the mailing
list
I noticed the same thing. The link is redirecting for me, but my reader
(AOL Reader) appears not to handle redirects.
Ivan
On Mar 25, 2015 9:11 AM, Magnus Bäck magnus.b...@sonymobile.com wrote:
The not too widely announced move from elasticsearch.(com|org) to
elastic.co the other week seems to
Easiest option, in terms of complexity, would probably be to use a bool
query, where product x and y are matched by an id query with high boosts.
Best option is probably the function score query:
http://www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-function-score-query.html
Off the top of my head, I cannot think of an existing filter that
accomplishes that task.
Creating a custom filter is easy. Simply creating a Lucene filter and
create a plug-in around it. Take a look at existing analysis plug-ins for
inspiration.
The discovery API is not modularized enough to use it outside of the
Elasticsearch context. I would simply use something like Zookeeper, which
is built exactly for situations like yours.
Cheers,
Ivan
On Mar 7, 2015 7:03 PM, Pierre de Soyres pdesoy...@gmail.com wrote:
Hello,
I would like to
Use the field in a filter and not part of the query. Is this field free
text?
Ivan
On Dec 23, 2014 9:12 PM, Roger de Cordova Farias
roger.far...@fontec.inf.br wrote:
Hello
Our documents have metadata indexed with them, but we don't want the
metadata to interfere in the scoring
After a
It used to be 2 concurrent streams. Has the default been upped in recent
versions? I agree, that number is awfully low. If you can disable indexing
during rolling restarts, those numbers can be much higher.
--
Ivan
On Sun, Nov 23, 2014 at 5:48 PM, joergpra...@gmail.com
joergpra...@gmail.com
It all depends on how many fields and how big they are. Retrieving a few
specific fields might be faster in cases, but in general, each field is
another seek in Lucene. Values are not retrieved at the same time. If you
are going to get all the fields, just use the source.
--
Ivan
On Sun, Nov
A term query will not analyze the search terms, so if your countries field
is using the default analyze, there will be no match since the standard
analyzer will lowercase the terms. Either set your field as not_analyzed or
use another query such as match.
--
Ivan
On Sat, Nov 22, 2014 at 4:35
Great work everyone. Feel better about upgrading now.
On Nov 22, 2014 4:42 PM, Boaz Leskes b.les...@gmail.com wrote:
Hi Christian, Daniel,
I believe I found the issue - it has to do with the cloud plugins (both
AWS and GCE) and the way they create the node list for the unicast based
Has an official issue been created? I would like to track the status.
So far, every 1.x.0 release has been buggy. :)
--
Ivan
On Fri, Nov 21, 2014 at 4:06 AM, Mark Walkom markwal...@gmail.com wrote:
It's being looked at, but I don't know much beyond that at the moment
sorry.
On 21 November
Disabling allocation helps, but it does not solve the problem completely.
Just like Nik, one of my complaints (although not my primary one). :)
I found that recovery gets easier when doing a rolling restart. First few
servers always rebalance, the last ones do not.
--
Ivan
On Thu, Nov 20, 2014
node
was chosen to minimise data loss following a data centre failure, however
because of the risk of split brain, the node actually functions more of a
warm DR than any sort of HA...
Regards,
Mat
On Thursday, November 20, 2014 2:32:14 PM UTC+13, Ivan Brusic wrote:
How many shards for each
We have always indexed nested documents, but never fully used them since
issue 3022 is still outstanding. Finally made the move to actually
filtering documents at the nested level.
Tracking metrics with graphite/grafana, I noticed immediately that the
active/current query count is much higher
fields cause the operator
to be honored ?
and Does the operator field within a nested query depend on if the field
in the nested field is actually analyzed or not. ?
Ramdev
On Tuesday, 18 November 2014 14:45:53 UTC-6, Ivan Brusic wrote:
I have never seen the array syntax with the match
Try using Jorg's plugin:
https://github.com/jprante/elasticsearch-plugin-arrayformat
--
Ivan
On Wed, Nov 19, 2014 at 7:15 AM, tch...@360incentives.com wrote:
On Wednesday, 19 November 2014 13:07:13 UTC+2, tch...@360incentives.com
wrote:
Hi Everyone! please advice me how return only
I have never seen the array syntax with the match query, so I am not sure
what the behavior should be. Since your search terms are not analyzed in
your example, a terms query with a minimum match of 100% should work. If
not, perhaps creating a single search term of your existing terms?
--
Ivan
You can set up an index template without creating the index. The template
will only be read when the index is created.
--
Ivan
On Tue, Nov 18, 2014 at 11:10 AM, Kingsley Elmes kingsleyel...@gmail.com
wrote:
Hi,
Does anyone know is it possible to set a field as 'not_analyzed' via the
Rolling upgrades should be supported:
http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/setup-upgrade.html#rolling-upgrades
How else can you perform a rolling upgrade without having a mixed cluster?
--
Ivan
On Thu, Nov 13, 2014 at 1:05 PM, joergpra...@gmail.com
query. I also like the fact that it allows aggregations
to not know about the query.
On Tue, Nov 11, 2014 at 5:27 PM, Ivan Brusic i...@brusic.com wrote:
I suddenly remembered when using facets that I had to apply the same
query filter as a facet filter with the join option disabled. Turns out
The terms are copied to the full name and are not analyzed as specified.
However, two terms are being copied, not one. The term query expects a
single token of Jeremy Smith, while you have two separate non analyzed
tokens.
Cheers,
Ivan
On Nov 12, 2014 10:29 AM, Robert Alkire
There is also an ActionModule
public void onModule(ActionModule module) {
module.registerAction(MyAction.INSTANCE, TransportMyAction.class);
}
It is always easier to follow existing plugins.
Cheers,
Ivan
On Wed, Nov 12, 2014 at 3:50 PM, Pawel pro...@gmail.com wrote:
Hi,
I'm thinking
Interesting in that they use Cassandra for discovery.
http://techblog.netflix.com/2014/11/introducing-raigad-elasticsearch-sidecar.html
--
You received this message because you are subscribed to the Google Groups
elasticsearch group.
To unsubscribe from this group and stop receiving emails
#3022 and related issues can bring about less ambiguous aggregations.
Nested aggregations on pre-filtered nested documents should work as is. If
not, the global scope aggregation should be used.
--
Ivan
On Mon, Nov 10, 2014 at 3:43 PM, Ivan Brusic i...@brusic.com wrote:
Reproducible gist
Quite the opposite: the Elasticsearch team and others have said that
authentication belongs outside of the application. Or at least, security
was not a high priority.
It seems like they are working on security and a release should be
forthcoming:
The status filter cache will indeed contain all entries. And technically,
the cache is per segment, and not across all documents, but this should be
transparent.
Caching is enabled by default for the term filters, but disabled for the
bool filter. You can enable it if you think users will be
To avoid this situation in the future, besides using a service to start
elasticsearch, you can enforce the max nodes setting:
node.max_local_storage_nodes: 1
--
Ivan
On Sun, Nov 9, 2014 at 5:24 PM, Mark Walkom markwal...@gmail.com wrote:
Yellow means unassigned replicas, try removing them
The only solution that I can think of is to execute your query with a post
filter, not a filtered query. In this way, your aggregations by default
will not be filtered. You can then have two histograms, one with the post
filter used as an aggregation filter, and the other one left alone.
--
Ivan
Just a guess, but I would assume that the count API does not score
documents, which is why it is faster, leading to a setting such as
min_score to be obsolete.
--
Ivan
On Mon, Nov 10, 2014 at 10:55 AM, Roly Vicaria roly...@gmail.com wrote:
Also, I'm trying this on v1.3.2
On Monday,
Off the top of my head, the easiest option would be to use a constant score
query. Wrap the original query and provide a boost to documents that
satisfy your exist filter.
Cheers,
Ivan
On Mon, Nov 10, 2014 at 12:54 PM, Volker s...@klest.de wrote:
I would like to know whether there is an
Reproducible gist: https://gist.github.com/brusic/81e1552ffd49a1f6a7aa
Surely I cannot be the only one to have encountered this issue.
--
Ivan
On Mon, Nov 10, 2014 at 12:53 PM, Ivan Brusic i...@brusic.com wrote:
Is it possible to aggregate only on the nested documents that are returned
Yes, that post explained it a lot better than I wanted to. :) But basically
yes, the exclude portion is only as part of an existing span, but a single
span term is not really a span.
Ultimately, span queries are not very flexible since they do not analyze
terms, which is why I suspect there are
We did some performance testing and found that the performance hit from
using DFS was minor.
--
Ivan
On Wed, Nov 5, 2014 at 8:55 AM, Sofiane Cherchalli sofian...@gmail.com
wrote:
Answering myself:
According to ES blog
is there 2 matches?
exclude : {
span_term : {
field1 : dog
}
}
I though we should exclude match with dog...
Could you please point me to proper information to understand what is
happening?
Thx,
Jade
Le mercredi 22 août 2012 02:01:09 UTC-4, Ivan Brusic a écrit
You would need to create a custom analyzer by basically repeating the
configuration of the snowball analyzer, but adding in the synonym filter.
You can't modify a stock analyzer, unless this has changed (if so, someone
please correct me).
--
Ivan
On Wed, Nov 5, 2014 at 6:43 PM, Iqbal Ahmed
You cannot search/filter on a non-indexed field.
--
Ivan
On Wed, Nov 5, 2014 at 11:45 PM, ramakrishna panguluri
panguluri.ramakris...@gmail.com wrote:
I have 10 fields inserted into elasticsearch out of which 5 fields are
indexed.
Is it possible to search on non indexed field?
Thanks in
An announce list would be awesome, but at least something to this list with
the [ANN] or [ANNOUNCEMENT] prefix like David has been doing.
Elasticsearch 1.4.0 and 1.3.5 were released, but there is no announcement
on the list. Elasticsearch also announced a product called Shield, which
should
:15 PM UTC-5, Ivan Brusic wrote:
Must clauses are queries that must return a document. In the first query,
any document returned MUST have a location of Germany. The valueType should
clause is optional and actually pointless as a filter since it does not
contribute to scoring.
Can you explain
Are you using REST? If so, Jorg wrote a plugin to help with such a task:
https://github.com/jprante/elasticsearch-arrayformat
--
Ivan
On Mon, Nov 3, 2014 at 8:36 AM, Lasse Schou lassesc...@gmail.com wrote:
Hi,
I want to know if it's possible to disable the _index, _type, _id
and _score
Must clauses are queries that must return a document. In the first query,
any document returned MUST have a location of Germany. The valueType should
clause is optional and actually pointless as a filter since it does not
contribute to scoring.
Can you explain what your query should be doing in
Mvel has been removed in recent versions of Elasticsearch due to security
issues. Either change your script to use Groovy (preferred) or install the
mvel plugin.
Cheers,
Ivan
On Oct 29, 2014 2:44 PM, Manuel Sciuto msci...@viajeros.com wrote:
Hello everyone
Do not understand why it does not
The default scoring algorithm is based on TF-IDF.
http://www.elasticsearch.org/guide/en/elasticsearch/guide/current/practical-scoring-function.html
http://www.elasticsearch.org/guide/en/elasticsearch/guide/current/scoring-theory.html
You can enable explain to see how documents are scored:
2014 16:47:56 UTC-3, Ivan Brusic escribió:
The default scoring algorithm is based on TF-IDF.
http://www.elasticsearch.org/guide/en/elasticsearch/guide/
current/practical-scoring-function.html
http://www.elasticsearch.org/guide/en/elasticsearch/guide/
current/scoring-theory.html
You can enable
There is nothing magical about rivers. With some Java code changes, most
rivers can be made to run as standalone Java processes. The only thing the
rivers do is (weakly) guarantee that only one river instance is run per
cluster.
Cheers,
Ivan
On Mon, Oct 27, 2014 at 4:11 AM,
I maintain a mapping on the client side to due the lookups. Thankfully my
taxonomy is static (but somewhat large). There is a PR to do server-side
mappings, but I don't think it would apply to aggregations and is quite old.
An alternative solution would be to create compound values such as
Unless you are moving to new hardware, there is no need to rsync your data.
Both Elasticsaerch 0.90.x and 1.3.x are based on Lucene 4, so the
underlying data is compatible. Of course, you should backup your data
before such an upgrade.
After restarting your new cluster with your old data, I would
There is more to the issue than merely your configuration. What are your
queries? Are you doing a lot of aggregations, especially on on
high-cardinality fields. What kind of hardware are you running now?
Using the API, looks at your field cache usage. The field cache is held
within the Java heap
? And also, is this safe?
On Saturday, October 18, 2014 2:41:50 PM UTC-4, Ivan Brusic wrote:
The number of shards will help you scale out in case you add more nodes
in the future. With your current shard count at 5, you cannot optimally
deploy and distribute a 6+ node cluster. However, your
Is there some reason why your Elasticsearch nodes cannot serve as both
master and data? I believe that dedicated master nodes should only come
into play with large clusters, way beyond the 3 you have. If your master
nodes are tied to your app nodes, then I believe you will have less
resiliency
The structure of your query is odd. Either it is some format that I am not
aware of or the Elasticsearch parser is not doing a good job at determining
it is invalid.
Your two filters should be joined via a bool filter. Sometime like (not
tested):
{
query: {
filtered: {
query: {
The number of shards will help you scale out in case you add more nodes in
the future. With your current shard count at 5, you cannot optimally deploy
and distribute a 6+ node cluster. However, your data is time-based, one per
day. Are queries on historical data important? I would start off with a
Have you tried setting no fields to be returned or the explicit
setNoFields() method?
http://jenkins.elasticsearch.org/job/Elasticsearch%20Master%20Branch%20Javadoc/Elasticsearch_API_Documentation/org/elasticsearch/action/search/SearchRequestBuilder.html#setNoFields()
--
Ivan
On Thu, Oct 16,
They are indeed executed in the defined order. Filters that are more
specific should be placed early on and those that cannot be cached
(geo/timebased) should be placed last.
Cheers,
Ivan
On Thu, Oct 16, 2014 at 5:16 AM, @mromagnoli marce.romagn...@gmail.com
wrote:
Hi everyone,
I have a
You cannot change the similarity on an existing index. There is no
technical measure why it could not occur, it appears to be simply a method
in place to prevent users from creating potentially huge errors. I say that
developers should have the option to shoot themselves in the foot!
Cheers,
The field do not need a custom analyzer, they just need to be simply marked
as non_analyzed.
You can setup a dynamic template that states any new field should be non
analyzed.
http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-root-object-type.html#_dynamic_templates
Perhaps it is easier to talk about the downsides of doc_values.
If you have slow disks, common when using low level VMs with shared disks,
then retrieving your data will be much slower.
Also, you cannot filter on doc_values fields, so it depends on your other
use cases.
The amount field seems
What type of query are you using? Perhaps the query you are using is not
using the same analyzer at search time.
--
Ivan
On Tue, Oct 7, 2014 at 6:06 AM, Lee Gee lee...@gmail.com wrote:
My users have issues with apostrophes: I need to index and search aaa's
as it is, and without the
There is no voting or other gamification, just a plain ol' mailing list.
Many of us respond as just another way to contribute to open-source.
--
Ivan
On Mon, Oct 6, 2014 at 3:37 AM, ahmed jamal maaz jamalahmedm...@gmail.com
wrote:
Hi all,
These are very good advises. I really appreciate it.
The code is difficult to debug due to the distributed nature of
Elasticsearch. Requests get serialized and are sent via a binary protocol,
so you cannot focus on specific classes. Dependency injection adds in
additional complexity. You cannot simply determine the construction time
values in the
I have not looked at that video, but most comparisons are with Solr and not
SolrCloud.
--
Ivan
On Oct 3, 2014 12:37 PM, Gaurav gupta gupta.gaurav0...@gmail.com wrote:
Kevin,
I found the recent comparision from the search experts @
Your indices should be fine as is. Lucene is guaranteed to be able to read
data from 1 major revision prior. Elasticsearch 0.20 is Lucene 3 and the
latest Elasticsearch is Lucene 4. Because of various bugs at the Lucene
level, you should run an optimize (normally discouraged) to upgrade the
Seems like a version mismatch. What versions of elasticsearch/logstash are
you using? Are you using the 'elasticsearch' output in logstash or
'elasticsearch_http'? Try using the latter.
--
Ivan
On Mon, Sep 29, 2014 at 1:24 PM, larrychu...@gmail.com wrote:
I get this in the logs when starting
In general, newer client libraries should not be used with older clusters.
Most of the version checking happens on the server side and the older code
does not know about the newer client.
--
Ivan
On Fri, Sep 26, 2014 at 9:54 AM, David Pilato da...@pilato.fr wrote:
I have no idea. Could be an
Otis, from what I understand, the default size for the cache is unbounded,
so cache eviction should not occur due to inconsistent range checks in the
default case.
--
Ivan
On Mon, Sep 22, 2014 at 9:27 PM, Otis Gospodnetic
otis.gospodne...@gmail.com wrote:
Hi,
It sounds like every single ES
The query string query is working because the ampersand is also being
stripped from the query.
Your best bet is to use the pattern tokenizer and explicitly define which
characters to split the input text on.
You basically want to create your own aggregation, which are basically
collectors at the Lucene level. Look at existing plugins which provide
custom aggregation.
Basically, elasticsearch uses a scatter-gather/map-reduce model for
distributed collections.
--
Ivan
On Sep 18, 2014 12:56 AM, tim
I have yet to switch over to groovy, so I can't comment on where your
current script is wrong (it looks good to me as well). However, you can use
the standard function score, which are easier to understand and do not
rely on scripting (technically better performance).
You cannot join documents in Lucene/Elasticsearch (at least not like a
RDBMS). You would need to either denormalize your data, join on the client
side or execute 2+ queries.
--
Ivan
On Fri, Sep 12, 2014 at 12:45 AM, matej.zerov...@gmail.com wrote:
Hello!
Can anyone shine some light on my
I would strongly prefer to maintain control of the indexing side and not in
Elasticsearch. In fact, the Elasticsearch team has talked about deprecating
river plugins. I do not have any numbers, but I would suspect that the
majority of users do not use a river plugin. And yes, the correct term is
new to this so I find some of the information hard to
understand. So sorry if I am asking stupid questions.
On 12 Sep 2014, at 18:26, Ivan Brusic i...@brusic.com wrote:
I would strongly prefer to maintain control of the indexing side and not
in Elasticsearch. In fact, the Elasticsearch team has
I think this release might be their biggest one since 1.0. Lots of big
changes including a change in the consensus algorithm. It might take time,
but that is only a guess.
--
Ivan
On Wed, Sep 10, 2014 at 2:57 AM, joergpra...@gmail.com
joergpra...@gmail.com wrote:
I use the Github issue
filter. It seems strange to
me to use the FilteredQuery when the query string is empty, but perhaps
that would be the most straight forward way of doing this.
thank you,
emanuel
Dne pondělí, 8. září 2014 17:21:21 UTC+2 Ivan Brusic napsal(a):
Which filter was ignored? I am assuming you meant
Hopefully Mike McCandless will get some of the new Lucene features into
Elasticsearch:
http://blog.mikemccandless.com/2014/08/a-new-proximity-query-for-lucene-using.html
I suspect it will come soon.
--
Ivan
On Mon, Sep 8, 2014 at 2:11 PM, Nikolas Everett nik9...@gmail.com wrote:
On Mon,
Malini, I would suggest starting a new thread instead of adding to an old
one.
I find the Java API for the boost functions to be confusing, or at least,
not as clean as the rest of the Java API. I wonder if the Elasticsearch
team would accept a PR. Jörg's example above could be used as a skeleton
Which filter was ignored? I am assuming you meant the post filter (which
might be still called filter at the Java API), which in this case the
filter is bypassed by design. Post filters allow you to filter the
documents returned, but leave the aggregations as is. Sounds like you are
looking for
On Wed, Sep 3, 2014 at 11:47 AM, joergpra...@gmail.com
joergpra...@gmail.com wrote:
ES scales best over multiple machines horizontally, not vertically. More
RAM does not automatically mean better performance at linear scale at a
certain point - it depends on the JVM if it can keep up.
Can you simply boost the non analyzed field? If the scores are still too
similar, try using a dis_max query with the non analyzed query getting a
higher boost:
http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/query-dsl-dis-max-query.html
--
Ivan
On Wed, Sep 3, 2014 at 7:16
There is a plugin which can help:
https://github.com/jprante/elasticsearch-index-termlist
--
Ivan
On Wed, Sep 3, 2014 at 11:47 AM, David Pilato da...@pilato.fr wrote:
I don't think you can as far as I remember the same thread about it some
months ago.
--
David ;-)
Twitter : @dadoonet /
I have something similar which uses search analytics to determine relevant
filters. No plugin or framework since everything works on the client side
during the creation of the query. The process is far from ideal and is
currently very conservative, providing only a slight boost. It does not
work
The actual limitation in Java is compressed pointers:
http://docs.oracle.com/javase/7/docs/technotes/guides/vm/performance-enhancements-7.html#compressedOop
Thankfully Elasticsearch can run multiple nodes on the same server. Just
pay attention to the direct size (off heap memory), mlockall
Working on it is my full time job so review would be quick
Nik
On Sep 2, 2014 6:51 PM, Ivan Brusic i...@brusic.com wrote:
For those that are not regulars on the mailing list, I am a fairly
active member that has used Elasticsearch for years.
I am leaving my full-time job to focus on other (techie
Is giving.assignee a sub-object or a nested document? Can you provide your
mapping? Use the mapping API for exact results (
http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/indices-get-mapping.html
)
Perhaps enabling explain would provide some hints,
--
Ivan
On Wed, Sep
ago. Would love to
work on something in that regard.
Cheers,
Ivan
On Wed, Sep 3, 2014 at 2:16 PM, Itamar Syn-Hershko ita...@code972.com
wrote:
On Thu, Sep 4, 2014 at 12:10 AM, Ivan Brusic i...@brusic.com wrote:
Thanks Jörg.
The incentives for an open-source project is to pad my resume
The _id field is one of the few reserved field names in Elasticsearch:
http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/mapping-id-field.html
You can set it to whatever you want, as long as it is not an object or
(empty) array, as in your case. I have no idea what the proper
1 - 100 of 513 matches
Mail list logo
