Definitely. I agree. It's good to stop loading before snapshot. Anyway,
doing index snapshot say every 1 hour and re-indexing documents never than
last 1-1.5 hour should reduce your index recovery time.
On 8 January 2013 07:36, Otis Gospodnetic wrote:
> Hi,
>
> Right, you can continue indexing, b
Hi
I have some experience with practical limits. We have several setup we
have tried to run with high load for long time:
1)
* 20 shards in one collection spread over 5 nodes (4 shards for the
collection per node), no redunancdy (only one replica per shard)
* Indexing 35-50 mio documents per
Thanks *Tomás !! *This was useful.
On Mon, Dec 31, 2012 at 6:03 PM, Tomás Fernández Löbbe <
tomasflo...@gmail.com> wrote:
> If by "cronned commit" you mean "auto-commit": auto-commits are local to
> each node, are not distributed, so there is no something like a
> "cluster-wide" atomicity there.
Hi,
Right, you can continue indexing, but if you need to run
http://master_host:port/solr/replication?command=backup on each node and
if you want a snapshot that represents a specific index state, then you
need to stop indexing (and hard commit). That's what I had in mind. But
if one just want
Please see:
http://lucene.472066.n3.nabble.com/Attention-Solr-4-0-SolrCloud-users-td4024998.html
- Mark
On Jan 7, 2013, at 9:16 PM, davers wrote:
> I have a SolrCloud as seen here: http://d.pr/i/ya86
>
> When I stop solr-shard-1 solr-shard-4 should become the new leader. Instead
> it does not
: "mysearch" requesthandler(see following codes), I found that custom sort
: just effect on the current page when I got multiple page results, but the
: sort is expected when I sets the rows which contains all the results. Does
: anybody know how to solve it or the reason?
I haven't familiarized
Thanks you guys, I got the reason now, there'is something wrong with
compareBottom method in my source,it's not consistent with compare method
--
View this message in context:
http://lucene.472066.n3.nabble.com/custom-solr-sort-tp4031014p4031444.html
Sent from the Solr - User mailing list arch
Hi Upayavira,
The custom sort field is not stored in the index, I want to archieve a
requirement that didfferent search users will get different search results
when they search same keyword by my search engine, the search users have
relationship with the each result document in the solr. But the
I have a SolrCloud as seen here: http://d.pr/i/ya86
When I stop solr-shard-1 solr-shard-4 should become the new leader. Instead
it does not. Here is the output from the logs.
INFO: A cluster state change has occurred - updating...
Jan 07, 2013 6:11:54 PM org.apache.solr.cloud.ShardLeaderElectionC
On Jan 7, 2013, at 4:26 PM, davers wrote:
> KeeperErrorCode = NoNode for /collections/productindex/leaders/shard1
Odd - offhand I don't recall something like this being brought up before. Is
this new for you, or always existed? Solr 4.0?
As far as a key for the colors, there is an open JIRA i
: Subject: When does Solr actually convert textual representation into non-text
: formats (e.g. Date)
The short answer is: any place you want.
At the lowest level, FieldType's are required to support converting
(legal) "String" values into whatever native java object best represents
their t
Hi,
Just thought this possibility: I think dynamic field is solr concept, on lcene
level all fields are the same, but in initial startup, lucene should load all
field information into memory (not field data, but schema).
If we have too many fields (like *_my_fields, * => a1, a2, ...), does this
Dynamic fields resulted in poor response times? How many fields did each
document have? I can't see how a dynamic field should have any
difference from any other field in terms of response time.
Or are you querying across a large number of dynamic fields
concurrently? I can imagine that slowing th
There's no problem with indexing while taking snapshot. The only issue I
found is some problem with index directory:
https://issues.apache.org/jira/browse/SOLR-4170
It looks like Solr always looks in .../data/index/ directory without
reading "index.properties" file (sometimes your index dir name ca
Every time I stop my SolrCloud (3 shards, 1 replica each, total 6 servers)
and then restart it I get the following error:
SEVERE: Error getting leader from zk
org.apache.solr.common.SolrException: Could not get leader props
at
org.apache.solr.cloud.ZkController.getLeaderProps(ZkController
Not to my knowledge. You could do a delete all and then merge the index in with
the core admin API, but that would be a less efficient copy basically, rather
than a straight file move. There is not currently a restore command though.
Also, keep in mind that unless you back up to a network store
Thanks Mark! I will wait for 4.1 then.
Actually I pasted both /admin/cores and /admin/collections to highlight that
the problem was only with SolrJ and both admin/collections and
admin/collections were working as expected.
Sorry for the confusion.
Regards
Jay
-Original Message-
From: M
Is it possible to restore an index (previously backed up) using the same
kind of http reste like request ? Something like
...solr/replication?command=restore ?
On Mon, Jan 7, 2013 at 2:12 PM, Mark Miller wrote:
> You should be able to continue indexing fine - it will just keep a point
> in time
> http://127.0.0.1:7500/solr/admin/cores?
Why did you paste that as the example then :) ?
4.0 has problems using the collections api with the CloudSolrServer. You will
be able to do it for 4.1, but for 4.0 you have to use an HttpSolrServer and
pick a node to talk to. For 4.0, CloudSolrServer is
You should be able to continue indexing fine - it will just keep a point in
time snapshot around until the copy is done. So you can trigger a backup at
anytime to create a backup for that specific time, and keep indexing away, and
the next night do the same thing. You will always have backed up
: My question is, why do i need two redundant fields to sort a multivalued field
: ('date_max' and 'date_min' for 'date')
: For me it's just a waste of space, poisoning the fieldcache.
how does two fields "poion the fieldcache" ? ... if there was a function
that could find the "min" or "max" val
Hi,
There may be a better way, but stopping indexing and then
using http://master_host:port/solr/replication?command=backup on each node
may do the backup trick. I'd love to see how/if others do it.
Otis
--
Solr & ElasticSearch Support
http://sematext.com/
On Mon, Jan 7, 2013 at 10:33 AM, L
Hello FF,
Something like SPM for Solr will help you understand what's making Solr
slow - CPU maxed? Disk IO? Swapping? Caches too small? ...
There are no general rules/recipes, but once you see what is going on we
can provide guidance.
Yes, you can have 1 or more replicas of a shard.
Otis
--
So
Can you run the SolrJ client from another machine (so you go over the
network) and put Wireshark in between? It will tell you if something is
actually trying to connect of if the problem is even earlier.
Otherwise, if you are on U*ix style machines look into dtrace/truss to see
the activity. On Wi
Right Mark,
I am accessing the Collections API using Solrj. This is where I am stuck. If
I just use the Collections API using http thru the browser, the behavior is
as expected. Is there an example of using the Collections API using SolrJ?
My code looks like
String[] urls =
{"http://127.0.0.1:898
Hi Rafat,
The following are scripts started in the same order (external zk, 1 instance
running at localhost:2181). I also tried with the embedded zk with the same
result
#Start of Server 1
export SOLR_HOME=/home/apache-solr-4.0.0
cd shard1A
java \
-Djetty.port=8983 \
-Djetty.home=$SOLR_HOME/e
On Jan 7, 2013, at 12:33 PM, Jay Parashar wrote:
> With my setup (4 servers running at localhost 8983, 8900, 7574 and 7500) when
> I manually do a
> http://127.0.0.1:7500/solr/admin/cores?action=CREATE&name=myColl1&instanceDir=default&dataDir=myColl1Data&collection=myColl1&numShards=2
> it cre
Thanks guys. Yeah, separate rolling collections seem like the better way to
go.
-Scott
On Sat, Dec 29, 2012 at 1:30 AM, Otis Gospodnetic <
otis.gospodne...@gmail.com> wrote:
> https://issues.apache.org/jira/browse/SOLR-4237
Hello!
Can you share the command you use to start all four Solr servers ?
--
Regards,
Rafał Kuć
Sematext :: http://sematext.com/ :: Solr - Lucene - Nutch - ElasticSearch
> Any clue to why this is happening will be greatly appreciated. This has
> become a blocker for me.
> I can use the HTTPS
Any clue to why this is happening will be greatly appreciated. This has become
a blocker for me.
I can use the HTTPSolrServer to create a core/make requests etc, but then it
behaves like Solr 3.6
http://host:port/solr/admin/cores and not
http://host:port/solr/admin/collections
With my setup (4
Hi Robi,
thank you for the contribution. It's exiting to read, that your index
isn't contaminated by the number of fields. I can't exclude other
mistakes, but my first experience with extensive use of dynamic fields
have been very poor response times.
Even though I found an other solution, I
Hi Uwe,
We have hundreds of dynamic fields but since most of our docs only use some of
them it doesn't seem to be a performance drag. They can be viewed as a sparse
matrix of fields in your indexed docs. Then if you make the
sortinfo_for_groupx an int then that could be used in a function que
Hello,
Using a SOLR Cloud architecture, what is the best procedure to backup and
restore SOLR index and configuration ?
Thanks,
Guillaume
Hi,
like I just wrote in my reply to the similar suggestion form Jack.
I'm not looking for a way to preprocess my data.
My question is, why do i need two redundant fields to sort a multivalued
field ('date_max' and 'date_min' for 'date')
For me it's just a waste of space, poisoning the fieldca
If the Multiple-to-one mapping would be stable (e.g. independent of a
query), why not implement it as a custom update.chain processor with a copy
to a separate field? There is already a couple of implementations
under FieldValueMutatingUpdateProcessor (first, last, max, min).
Regards,
Alex.
Pe
Am 31.08.2012 13:35, schrieb Erick Erickson:
... what would the correct behavior
be for "sorting on a multivalued field"
Hi Erick,
in generally you are right, the question of multivalued fields is which
value the reference is. But there are thousands of cases where this
question is implicit
Hi Jack,
thank you for the hint.
Since I have already a solrj client to do the preprocessing, mapping to
sort fields isn't my problem. I will try to explain better in my reply
to Erick.
Uwe
(Sorry late reaction)
Am 30.08.2012 16:04, schrieb Jack Krupansky:
You can also use a "Field Mutatin
This should be clarified some. In the "client" API, SolrServer is represents a
connection to a single server backend/endpoint and should be re-used where possible.
The approach being discussed is to have one "client" connection (represented by SolrServer class) per solr core, all residing in a "
This is the exact approach we use in our multithreaded env. One server per
core. I think this is the recommended approach.
-Original Message-
From: Parvin Gasimzade [mailto:parvin.gasimz...@gmail.com]
Sent: Monday, January 07, 2013 7:00 AM
To: solr-user@lucene.apache.org
Subject: Re: Max
if you are inside solr, as it seems to be the case, you can do this
QParserPlugin qplug =
req.getCore().getQueryPlugin(LuceneQParserPlugin.NAME);
QParser parser = qplug.createParser("PATIENT_GENDER:Male OR
STUDY_DIVISION:\"Cancer Center\"", null, req.getParams(), req);
Query q = parser.parse();
I know that but my question is different. Let me ask it in this way.
I have a solr with base url localhost:8998/solr and two solr core
as localhost:8998/solr/core1 and localhost:8998/solr/core2.
I have one baseSolr instance initialized as :
SolrServer server = new HttpSolrServer( url );
I have a
Can you explain why you want to implement a different sort first? There
may be other ways of achieving the same thing.
Upayavira
On Sun, Jan 6, 2013, at 01:32 AM, andy wrote:
> Hi,
>
> Maybe this is an old thread or maybe it's different with previous one.
>
> I want to custom solr sort and pas
Am 06.01.2013 02:32, schrieb andy:
I want to custom solr sort and pass solr param from client to solr server,
Hi Andy,
not a answer of your question, but maybe an other approach to solve your
initial question. Instead of writing a new SearchComponent I decided to
(miss)use the function >htt
This might help:
https://wiki.apache.org/solr/Solrj#HttpSolrServer
Note that the associated SolrRequest takes the path, I presume relative to
the base URL you initialized the HttpSolrServer with.
Best
Erick
On Mon, Jan 7, 2013 at 7:02 AM, Parvin Gasimzade wrote:
> Thank you for your responses
This is all quite strange, lots of people are using SolrCloud,
some with very large clusters, so I'm guessing it's something
in your setup that isn't obvious.
How certain are you that your network between the two
machines is reliable? And have you tried with a nightly build?
I'm grasping at straw
Thank you for your responses. I have one more question related to Solr
multi-core.
By using SolrJ I create new core for each application. When user wants to
add data or make query on his application, I create new HttpSolrServer for
this core. In this scenario there will be many running HttpSolrServ
Hi,
I found a own hack. It's based on free interpretation of the function
strdist().
Have:
- one multivalued field 'part_of'
- one unique field 'groupsort'
Index each item:
For each group membership:
add groupid to 'part_of'
concat groupid and sortstring to new string
ad
Is there a way to get Lucene's query from Solr query?. I have a requirement
to search for terms in multiple heterogeneous indices. Presently, I am using
the following approach
try {
Directory directory1 = FSDirectory.open(new
File("E:\\database\\patient\\ind
You should use language detection processor factory, like below:
content
language
en
*true
content,fullname*
true
en,fr,de,es,ru,it
0.7
Once you have defined fields like content_en, content_fr etc., they will b
49 matches
Mail list logo
