Hi,
I am using solr 4 setup. For the backup purpose once in a day I start one
additional tomcat server with cores having empty data folders and which acts
as a slave server. However it does not replicate data from the master unless
there is a commit on the master. Is there a possibility to pull da
My prev question was
I have updated 250 data to solr.
and some of data have "category" field and some of don't have.
for example.
{
"id":"321",
"name":"anurag",
"category":"30"
},
{
"id":"3",
"name":"john"
}
now i want to search that docs who does not have that field.
what que
Hi Vicky,
May be startup ?
For backups http://master_host:port/solr/replication?command=backup would be
more suitable.
or startup
--- On Thu, 3/14/13, vicky desai wrote:
> From: vicky desai
> Subject: Solr Replication
> To: solr-user@lucene.apache.org
> Date: Thursday, March 14, 2013, 9:
Hi all,
this is not a question. I just wanted to announce that I've written a blog post
on how to set up Maven for packaging and automatic testing of a SOLR index
configuration.
http://blog.it-agenten.com/2013/03/integration-testing-your-solr-index-with-maven/
Feedback or comments appreciated
Informative. Useful.Thanks
On Thu, Mar 14, 2013 at 1:59 PM, Chantal Ackermann <
c.ackerm...@it-agenten.com> wrote:
> Hi all,
>
>
> this is not a question. I just wanted to announce that I've written a blog
> post on how to set up Maven for packaging and automatic testing of a SOLR
> index config
Nice,
Chantal can you indicate there or here what kind of speed for integration tests
you've reached with this, from a bare source to a successfully tested
application?
(e.g. with 100 documents)
thanks in advance
Paul
On 14 mars 2013, at 09:29, Chantal Ackermann wrote:
> Hi all,
>
>
> thi
Hi
I'm getting this error after a few hours of filling solr with documents.
Tomcat is running with -Xms1024m -Xmx4096m.
Total memory of host is 12GB. Softcommits are done every second and hard
commits every minute.
Any idea why this is happening and how to avoid this?
*top*
PID USER P
Hi Paul,
I'm sorry I cannot provide you with any numbers. I also doubt it would be wise
to post any as I think the speed depends highly on what you are doing in your
integration tests.
Say you have several request handlers that you want to test (on different
cores), and some more complex use c
Chantal,
the goal is different: get a general feeling how practical it is to integrate
this in the routine.
If you are able, on your contemporary machine which I assume is not a
supercomputer of some special sort, to run this whole process somewhat useful
for you in about 2 minutes then I'll be
When I shutdown tomcat free -m and top keeps telling me the same values.
Almost no free memory...
Any idea?
On 03/14/2013 10:35 AM, Arkadi Colson wrote:
Hi
I'm getting this error after a few hours of filling solr with
documents. Tomcat is running with -Xms1024m -Xmx4096m.
Total memory of hos
Hi All.
I am monitoring two solr 4.1 solr instances in master-slave setup. On
both nodes I check url /solr/replication?command=details and parse it
to get:
- on master: if replication is enabled -> field replicationEnabled
- on slave: if replication is enabled -> field replicationEnabled
- on slav
Writing "OR -" is simply the same as "-", so the query would match documents
containing category 20 and then remove all documents that had any category
(including 20) specified, giving you nothing.
Try:
http://localhost:8983/search?q=*:*&wt=json&start=0&fq=category:"20"; OR
(*:* -category:[*
In the output of:
/solr/replication?command=details
there is indexVersion mentioned many times:
0
3
22.59 KB
/usr/share/solr/data/index/
1363259880360
4
_1.tvx
_1_nrm.cfs
_1_Lucene41_0.doc
_1_Lucene41_0.tim
_1_Lucene41_0.tip
_1.fnm
_1_nrm.cfe
_1.fdx
_1_Lucene41_0.pos
_1.tvf
_1.fdt
_1_Luce
Hello!
Thanks a lot, Erick! I've attached some stack traces during a normal
'engine' running.
Cheers,
- Luis Cappa
2013/3/13 Erick Erickson
> Stack traces..
>
> First,
> jps -l
>
> that will give you a the process IDs of your running Java processes. Then:
>
> jstack
>
> Usually I pipe the o
Does it only count if you are using SolrCloud? We are using a traditional
Master/Slave setup with Solr 4.1:
1 Master per 14 days:
Documents: ~15mio
Index size: ~150GB (stored fields)
#of masters: +30
Performance: SUCKS big time until caches catches up. Unfortunately that takes
quite some time.
Hello,
I need some advice with my solrcloud cluster and the DIH. I have a cluster
with 3 cloud servers. Every server has an solr instance and a zookeeper
instance. I start it with the -Dzkhost parameter. It works great, i send
updates by an curl(xml) like this:
curl http:/ip:SOLRport/solr/update
Christian,
SSDs will warm up muuuch faster.
Your other questionable require more info / discussion.
Otis
Solr & ElasticSearch Support
http://sematext.com/
On Mar 14, 2013 8:47 AM, "Christian von Wendt-Jensen" <
christian.vonwendt-jen...@infopaq.com> wrote:
> Does it only count if you are using S
On Thu, 2013-03-14 at 13:10 +0100, Arkadi Colson wrote:
> When I shutdown tomcat free -m and top keeps telling me the same values.
> Almost no free memory...
>
> Any idea?
Are you reading top & free right? It is standard behaviour for most
modern operating systems to have very little free memory
On Mar 14, 2013, at 8:10 AM, Rafał Radecki wrote:
> Is this a bug?
Yes, 4.1 had some replication issues just as you seem to describe here. It all
should be fixed in 4.2 which is available now and is a simple upgrade.
- Mark
On Mar 14, 2013, at 9:22 AM, roySolr wrote:
> Hello,
>
> When i run this it goes with 3 doc/s(Really
> slow). When i run solr alone(not solrcloud) it goes 600 docs/sec.
>
> What's the best way to do a full re-index with solrcloud? Does solrcloud
> support DIH?
>
> Thanks
>
SolrCloud suppo
On 03/14/2013 03:11 PM, Toke Eskildsen wrote:
On Thu, 2013-03-14 at 13:10 +0100, Arkadi Colson wrote:
When I shutdown tomcat free -m and top keeps telling me the same values.
Almost no free memory...
Any idea?
Are you reading top & free right? It is standard behaviour for most
modern operatin
Based on what does solr replicate the whole shard again from zero? From
time to time after a restart of tomcat solr copies over the whole shard
to the replicator instead of doing only the changes.
BR,
Arkadi
I'm using solr 3.6.2 to crawl some data using nutch, in my schema I've one
field with all the content extracted from the page, which could possibly
include email addresses, this is the configuration of my schema:
I believe this is the same issue as described, I'm running 4.2 and as you can
see my slave is a couple versions ahead of the master (all three slaves show
the same behavior). This was never the case until I upgraded from 4.0 to
4.2.
Master:
1363272681951
93
1,022.31 MB
Slave:
1363273274085
95
Hi,
Since you have word delimiter filter in your analysis chain, I am not sure if
e-mail addresses are recognised. You can check that on solr admin UI, analysis
page.
If e-mail addresses kept one token, I would use leading wildcard query.
&q=*@gmail.com
There was a similar question recently:
Hi
We have been using Solr 4.0 for a while now and wanted to upgrade to 4.2.
But our application stopped working. When we tried 4.1 it was working as
expected.
Here is a description of the situation.
We deploy a Solr web application under java 7 on a Glassfish 3.1.2.2
server. We added some clas
What calls are you using to get the versions? Or is it the admin UI?
Also can you add any details about your setup - if this is a problem, we need
to duplicate it in one of our unit tests.
Also, is it affecting proper replication in any way that you can tell.
- Mark
On Mar 14, 2013, at 11:12 A
Perhaps as a result of https://issues.apache.org/jira/browse/SOLR-4451 ?
Just a guess.
The root cause looks to be:
> Caused by: java.io.IOException: Keystore was tampered with, or password was
> incorrect
- Mark
On Mar 14, 2013, at 11:24 AM, Uwe Klosa wrote:
> Hi
>
> We have been using Sol
hello everyone,
i know this is a general topic - but would really appreciate info from
others that are doing this now.
- how are others managing this so that users are impacted the least
- how are others handling the scenario where users don't want to migrate
forward.
thx
mark
--
View
Thanks, but nobody has tempered with keystores. I have tested the
application on different machines. Always the same exception is thrown.
Do we have to set some system property to fix this?
/Uwe
On 14 March 2013 16:36, Mark Miller wrote:
> Perhaps as a result of https://issues.apache.org/ji
Hey all,
We're using a Solr 4 core to handle our article data. When someone in our CMS
publishes an article, we have a listener that indexes it straight to solr. We
use the previously instantiated HttpSolrServer, build the solr document, add it
with server.add(doc) .. then do a server.commit(
I found the answer myself. Thanks for the pointer.
Cheers
Uwe
On 14 March 2013 16:48, Uwe Klosa wrote:
> Thanks, but nobody has tempered with keystores. I have tested the
> application on different machines. Always the same exception is thrown.
>
> Do we have to set some system property to fix
Hi Arkadi,
If the update delta between the shard leader and replica >100 docs, then
Solr punts and replicas the entire index. Last I heard, the 100 was
hard-coded in 4.0 so is not configurable. This makes sense because the
replica shouldn't be out-of-sync with the leader unless it has been offline
Hi
After doing a query to Solr to get the uniqueIds (string of 20 characters)
of 700 documents in a collection, I'm getting an out of memory error using
Solr 4.2. I tried to increase the JVM-Memory 1G (from 3G to 4G) however this
didn't change anything.
This was working on 3.5.
I've moved from
ids request to shard with star query are slow
I have a distributed solr environment and I am investigating all the
request where the shard took significant amount of time. One common pattern
i saw was all the ids request with q=*:* and ids= took around
2-3sec. i picked some shard request q=xyz an
On Thursday, March 14, 2013 at 4:57 PM, Uwe Klosa wrote:
> I found the answer myself. Thanks for the pointer.
Would you mind sharing you answer, Uwe?
On Thu, Mar 14, 2013 at 12:07 PM, raulgrande83 wrote:
> JVM: IBM J9 VM(1.6.0.2.4)
I don't recommend using this JVM.
On 3/14/2013 9:24 AM, Uwe Klosa wrote:
This exception occurs in this part
new ConcurrentUpdateSolrServer("http://solr.diva-portal.org:8080/search";,
5, 50)
Side comment, unrelated to your question:
If you're already aware that ConcurrentUpdateSolrServer has no built-in
error handling and you
On Mar 14, 2013, at 1:27 PM, Shawn Heisey wrote:
> I have been told that it is possible to override the handleError method to
> fix this
I'd say mitigate more than fix. I think the real fix requires some dev work.
- Mark
On 3/14/2013 3:35 AM, Arkadi Colson wrote:
Hi
I'm getting this error after a few hours of filling solr with documents.
Tomcat is running with -Xms1024m -Xmx4096m.
Total memory of host is 12GB. Softcommits are done every second and hard
commits every minute.
Any idea why this is happening and how
Hi everyone,
Is there an official definition of the "Current" flag under Core >
Home > Statistics?
What would it mean if a shard leader is not "Current"?
Thanks,
Michael Della Bitta
Appinions
18 East 41st Street, 2nd Floor
New York, NY 10017-627
Hi,
I think that in solr 4.2 the new feature to proxy a request if the
collection is not in the requested node has a bug.
If I do a query with the parameter rows=0 and the node doesn't have the
collection. If the parameter is rows=4 or superior then the search works as
expected
the curl return
I'll add a test with rows = 0 and see how easy it is to replicate.
Looks to me like you should file a JIRA issue in any case.
- Mark
On Mar 14, 2013, at 2:04 PM, yriveiro wrote:
> Hi,
>
> I think that in solr 4.2 the new feature to proxy a request if the
> collection is not in the requested
The log of the UI
null:org.apache.solr.common.SolrException: Error trying to proxy request for
url: http://192.168.20.47:8983/solr/ST-3A856BBCA3_12/select
I will open the issue in Jira.
Thanks
-
Best regards
--
View this message in context:
http://lucene.472066.n3.nabble.com/Solr-4-2-me
: It looks strange to me that if there is no document yet (foundVersion < 0)
: then the only case when document will be imported is when input version is
: negative. Guess I need to test specific cases using SolrJ or smth. to be sure.
you're assuming that if foundVersion < 0 that means no documen
Sorry for the duplicated mail :-(, any advice on a configuration for searching
emails in a field that does not have only email addresses, so the email
addresses are contained in larger textual messages?
- Mensaje original -
De: "Ahmet Arslan"
Para: solr-user@lucene.apache.org
Enviados:
I've been looking all over for a clear answer to this question and can't seem
to find one. It seems like a very basic concept to me though so maybe I'm
using the wrong terminology. I want to be able to search across multiple
collections (as it is now called in SolrCloud world, previously called
Co
Yes, with SolrCloud, it's just the collection param (as long as the schemas are
compatible for this):
http://wiki.apache.org/solr/SolrCloud#Distributed_Requests
- Mark
On Mar 14, 2013, at 2:55 PM, kfdroid wrote:
> I've been looking all over for a clear answer to this question and can't seem
>
Hey Michael
I was a bit confused because you mentioned SolrCloud in the subject. We're
talking about http://host:port/solr/#/collection1 (f.e.) right? And there, the
left-upper Box "Statistics" ?
If so, the Output comes from /solr/collection1/admin/luke (
http://svn.apache.org/viewvc/lucene/de
I'm assuming from that link I would use the following:
/Query all shards of multiple compatible collections, explicitly specified:/
http://localhost:8983/solr/collection1/select?collection=collection1_NY,collection1_NJ,collection1_CT
where collection1_NY, NJ and CT could be books, movies, music i
Stefan,
Thanks a lot! Makes sense. So I don't have to worry about my leader
thinking it's out of date, then.
Michael Della Bitta
Appinions
18 East 41st Street, 2nd Floor
New York, NY 10017-6271
www.appinions.com
Where Influence Isn’t a Game
On
Perhaps the wording of "Current" is a bit too generic in that context? I'd like
to change that description if that clarifies things .. but not sure which one
is a better fit?
On Thursday, March 14, 2013 at 8:26 PM, Michael Della Bitta wrote:
> Stefan,
>
> Thanks a lot! Makes sense. So I don
Something like 'Reader is Current' might be better. Personally, I don't even
know if it's worth showing.
- Mark
On Mar 14, 2013, at 3:40 PM, Stefan Matheis wrote:
> Perhaps the wording of "Current" is a bit too generic in that context? I'd
> like to change that description if that clarifies t
Hi, I am new with Solr and I am extracting metadata from binary files through
URLs stored in my database. I would like to know what fields are available
for indexing from PDFs (the ones that would be initiated as in column=””).
For example how would I extract something like file size, format or f
Take a look at Solr Cell:
http://wiki.apache.org/solr/ExtractingRequestHandler
Include a dynamicField with a "*" pattern and you will see the wide variety
of metadata that is available for PDF and other rich document formats.
-- Jack Krupansky
-Original Message-
From: Luis
Sent: Th
Sure. copyField it into a new indexed non-stored field with the following
type definition:
Content of filter_email.txt is (including <> signs):
You will have the emails only left as tokens. Can't display them easily,
but can certainly search.
Regards,
Hi Scott,
Not sure why IW would be closed, but:
* consider not (hard) committing after each doc, but just periodically,
every N minutes
* soft committing instead
* using 4.2
Otis
--
Solr & ElasticSearch Support
http://sematext.com/
On Thu, Mar 14, 2013 at 11:55 AM, Danzig, Scott wrote:
> He
Wow! That's great. And it's a lot of work, especially getting it all
keyboard-complete. Thank you.
On 03/14/2013 01:29 AM, Chantal Ackermann wrote:
Hi all,
this is not a question. I just wanted to announce that I've written a blog post
on how to set up Maven for packaging and automatic testi
3docs/s is lower, I test with 4 node is more 1000docs/s and 4k/doc with
solrcloud. Every leader has a replica.
I am tuning to improve to 3000docs/s. 3docs/s is too slow.
3x!
--
View this message in context:
http://lucene.472066.n3.nabble.com/Advice-solrCloud-DIH-tp4047339p4047559.html
Sent fr
give u to test embeded solr:
import java.io.File;
import java.io.IOException;
import java.net.MalformedURLException;
import java.util.ArrayList;
import java.util.Collection;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.core.SimpleAnalyzer;
import org.apache.lucene
H, could you raise a JIRA and assign it to me? Please be sure and
emphasize that it's embedded because I'm pretty sure this is fine for the
regular case.
But I have to admit that the embedded case completely slipped under the
radar.
Even better if you could make a test case, but that might no
Hi!
Take a look on
http://wiki.apache.org/solr/SchemaXml#Common_field_options
parameter "*omitTermFreqAndPositions"*
or you can use a custom similarity class that overrides the term freq and
return one for only that field.
http://wiki.apache.org/solr/SchemaXml#Similarity
B
On my solr 4 setup a query returns a higher "NumFound" value during a
*:* query than the "Num Docs" value reported on the statistics page of
collection1. Why is that? My data is split across 3 data import handlers
where each handler has the same type of data but the ids are guaranteed
to be dif
Hi,
I have a multi core setup and there is continuous updation going on in each
core. Hence I dont prefer a bckup as it would either cause a downtime or if
during a backup there is a write activity my backup will be corrupted. Can
you please suggest if there is a cleaner way to handle this
--
V
I wrote a simple test to reproduce a very similar stack trace to the above
issue, where only some line numbers differences.
Any ideas as to why the following happens? Any help would be very
appreciated.
* The test case:
> @Test
> public void documentCommitAndRollbackTest() throws Exc
65 matches
Mail list logo
