Is it possible to force a Shard Leader change?

Hi Guys, I am running a Solr Cloud 4.10, with 4 Solr servers and 5 Zookeeper setup. Solr servers: solr01, solr02, solr03, solr04 I have around 20 collections in Solr cloud, and there are 4 Shards for each Collection. For each Shard, I have 4 Replicas, and sitting on each Solr server, with one

The Query Elevation Component

Hi everyone, I'm reading the docs on the query elevation component and some questions came up: Can I specify a field that the elevate component will look at, such as only looking at the title field? My search handler (using eDisMax) is searching across multiple fields, but if I only want the

Re: Search sort depth limited to 4?

Yes, currently there is the 4 sort field limit. A custom handler could be built that allows for unlimited sorts or you could provide a patch to the export handler. I think though that you'll find that performance drops off quite a bit as the number of sort fields increases. This is because each

Re: SolrCloud create_collection not uploading configs to zookeeper

OK, I can see '/configs' directory in Solr UI and under that I can see configuration fo my 'test' collection. BUt this all seemed to be disjointed information. Doc is definitely not clear. And what that Tree represent anyway where;s information for that. I would ideally put link to that document.

Re: SolrCloud create_collection not uploading configs to zookeeper

Hi Erick, I did read that paragraph. It says: First, if you don't provide the -d or -n options, then the default configuration ($SOLR_HOME/server/solr/configsets/data_driven_schema_configs/conf) is uploaded to ZooKeeper using the same name as the collection. For example, the following command

Re: SolrCloud create_collection not uploading configs to zookeeper

>From the doc you referenced where it outlines the parameters for the create command: -d : The configuration directory. This defaults to data_driven_schema_configs. You should also review the linked section about what configuration directories are all about:

SolrCloud create_collection not uploading configs to zookeeper

Hi, I have 2 nodes solr cluster with 1 node standalone zookeeper. I tried following from following doc to create collection: bin/solr create -c contacts. According to doc should upload config date into /configs/contacts in ZooKeeper. But I can't find any configs directory under zookeeper.

Re: Solr MapReduce Indexer : go-live option throwing exception

Can't really deal with the security issues, but... The resulting indexes created by MRIT are just plain vanilla Solr/Lucene indexes. All the --go-live step does is issue a MERGEINDEXES command from the core where they live to the directory MRIT leaves them in, you might get some joy there, see:

Re: Example of posting to /stream in SolrJ?

I posted this also to another thread, but I'll cross post to this ticket: Take a look at org.apache.solr.client.solrj.io.sql. StatementImpl.constructStream() This uses a SolrStream to connect to the /sql handler. You can use the same approach to send a request to the /stream handler just by

Re: Can't load schema managed-schema: unknown field 'id'

Hmm, this is odd. You should not have had to restart Zookeeper, are you 100% sure you looked in the same place you downloaded to? Of course you would have to reload the collection to get them to "take". BTW, in Solr 5.4+ there's upconfig/downconfig from the bin/solr script, it was put there to

Re: Can't load schema managed-schema: unknown field 'id'

Finally got it to straighten out. So I have two collections, my test collection and my production collection. I "fat fingered" the test collection and both collections were complaining about the missing "id" field. I downloaded the config from both collections and it was showing the id

Re: Can't load schema managed-schema: unknown field 'id'

@John I am using a managed schema with zookeeper/solrcloud. On 07/26/2016 04:21 PM, John Bickerstaff wrote: @Michael - somewhere there should be a "conf" directory for your SOLR instance. For my Dev efforts, I moved it to a different directory and I forget where it was, originally -- but if

Re: Can't load schema managed-schema: unknown field 'id'

ok... I downloaded the config for both of my collections and the downloaded managed-schema file shows "id" as defined? But the online view in the UI shows it as not defined? I've tried re-upping the config and nothing changes. -Mike On 07/26/2016 04:11 PM, John Bickerstaff wrote:

RE: Can't load schema managed-schema: unknown field 'id'

@Michael - there are GUI available for ZooKeeper: http://stackoverflow.com/questions/24551835/available-gui-for-zookeeper I used the Eclipse plugin before and while it is a bit clunky it gets the job done. Alexandre Drouin -Original Message- From: John Bickerstaff

Re: Can't load schema managed-schema: unknown field 'id'

@Michael - somewhere there should be a "conf" directory for your SOLR instance. For my Dev efforts, I moved it to a different directory and I forget where it was, originally -- but if you search for solrconfig.xml or schema.xml, you should find it. It could be on your servers (or on only one of

Re: Can't load schema managed-schema: unknown field 'id'

and further on in the file... < uniqueKey>id On Tue, Jul 26, 2016 at 2:17 PM, John Bickerstaff wrote: > I don't see a managed schema file. As far as I understand it, id is set > as a "uniqueKey" in the schema.xml file... > > On Tue, Jul 26, 2016 at 2:11 PM,

Re: Can't load schema managed-schema: unknown field 'id'

I don't see a managed schema file. As far as I understand it, id is set as a "uniqueKey" in the schema.xml file... On Tue, Jul 26, 2016 at 2:11 PM, Michael Joyner wrote: > ok, I think I need to do a manual edit on the managed-schema file but I > get "NoNode" for

Re: Can't load schema managed-schema: unknown field 'id'

@Michael - if you're on Linux and decide to take Alexandre's advice, I can possibly save you some time. I wrestled with getting the data in and out of zookeeper a while ago... sudo /opt/solr/server/scripts/cloud-scripts/zkcli.sh -cmd upconfig -confdir /home/john/conf/ -confname collectionName -z

Re: Can't load schema managed-schema: unknown field 'id'

ok, I think I need to do a manual edit on the managed-schema file but I get "NoNode" for /managed-schema when trying to use the zkcli.sh file? How can I get to this file and edit it? On 07/26/2016 03:05 PM, Alexandre Drouin wrote: Hello, You may have a uniqueKey that points to a field that

Re: Streaming expression - workers is zero somehow?

Take a look at StatementImpl.constructStream() This uses a SolrStream to connect to the /sql handler. You can use the same approach to send a request to the /stream handler just by changing the parameters. Then you can open and read the SolrStream. We don't yet have a load balancing SolrStream.

Example of posting to /stream in SolrJ?

Does anyone have an example of just POST'ing a streaming expression to the /stream handler from SolrJ client code? i.e. I don't want to parse and execute the streaming expression on the client side, rather, I want to post the expression to the server side. Currently, my client code is a big copy

RE: Can't load schema managed-schema: unknown field 'id'

Other than deleting the collection, I think you'll have to edit the manage-schema file manually. Since you are using SolrCloud you will need to use Solr's zkcli (https://cwiki.apache.org/confluence/display/solr/Command+Line+Utilities) utility to download and upload the file from ZooKeeper.

Re: Can't load schema managed-schema: unknown field 'id'

Same error via the UI: Can't load schema managed-schema: unknown field 'id' On 07/26/2016 03:05 PM, Alexandre Drouin wrote: Hello, You may have a uniqueKey that points to a field that do not exists anymore. You can try adding an "id" field using Solr's UI or the schema API since you are

Re: Can't load schema managed-schema: unknown field 'id'

The schema API is failing with the the unknown field "id" error. Where in the UI could I try adding this field back at? On 07/26/2016 03:05 PM, Alexandre Drouin wrote: Hello, You may have a uniqueKey that points to a field that do not exists anymore. You can try adding an "id" field using

RE: Can't load schema managed-schema: unknown field 'id'

Hello, You may have a uniqueKey that points to a field that do not exists anymore. You can try adding an "id" field using Solr's UI or the schema API since you are using the managed-schema. Alexandre Drouin -Original Message- From: Michael Joyner [mailto:mich...@newsrx.com] Sent:

Re: Search sort depth limited to 4?

So I found the limit in the Ref Doc p. 394, under the /export request handler: "Up to four sort fields can be specified per request, with the 'asc' or 'desc' properties" Yikes I'm in trouble. Does anyone know if this can be circumvented? Can I write a custom handler that could handle up to 20?

How to know if a core was reloaded without errors

Hi, I am using the Collection API to reload a core and I was wondering if there is a way to know if the core was reloaded without errors. For my testing, I added a known error (an invalid field in a request handler) in my configuration and I use the url

Search sort depth limited to 4?

Hi, I'm trying to group search results by fields using the streaming API. I don't see a sort limit mentioned in the Solr Ref Doc, but when I use 4 fields I get results and when I use 5 or more I get an exception: java.util.concurrent.ExecutionException: java.io.IOException: JSONTupleStream:

Can't load schema managed-schema: unknown field 'id'

|Help!| | | |What is the best way to recover from: | Can't load schema managed-schema: unknown field 'id' |I was managing the schema on a test collection, fat fingered it, but now I find out the schema ops seem to altering all collections on the core? SolrCloud 5.5.1 ||| | -Mike|||

Solr is locked after killed - hdfs

Hi, I have issue that I saw it's a bug that really didn't resolved: https://issues.apache.org/jira/browse/SOLR-8335 I worked with solr 6.1, on HDFS but I see it's also exists in solr 5.X I started my solr, and because of out of memory exception, it was killed. Now when I start it again, I have

Solr MapReduce Indexer : go-live option throwing exception

Hi, I was trying to use the Mapreduce Indexer tool from cloudera, to index my data in Hive table using Solr. hadoop jar /path/to/lib/solr/contrib/mr/search-mr-*-job.jar org.apache.solr.hadoop.MapReduceIndexerTool -Djute.maxbuffer=--morphline-file /path/to/morphlines.conf --output-dir

Re: solr.NRTCachingDirectoryFactory

And, I might add, you should look through your old logs and see how long it takes to open a searcher. Let's say Shawn's lower bound is what you see, i.e. it takes a minute each to execute all the autowarming in filterCache and queryResultCache... So you're current latency is _at least_ 2 minutes

Re: Streaming expression - workers is zero somehow?

Ok, makes sense now, thanks Joel. We should probably add some earlier error checking vs. letting the code get all the way into the HashQParser. So this raises a separate question that I haven't been able to figure out, namely, do we have an example of just POST'ing the expression to the /stream

Re: Streaming expression - workers is zero somehow?

The difference would be if you are compiling and running the expression in a java class or sending it to the /stream handler to be compiled. If you're compiling it and running it locally you could get this error because the StreamContext would not have the numWorkers variable set. The /stream

Re: Streaming expression - workers is zero somehow?

it's from a unit test, but not sure why that matters? If I wrap the expression in a parallel expression with explicit workers=1, then it works On Thu, Jul 21, 2016 at 11:13 AM, Joel Bernstein wrote: > Are you getting this error from a test case you've setup or from a manual >

Re: solr.NRTCachingDirectoryFactory

On 7/26/16 5:46 AM, Shawn Heisey wrote: On 7/22/2016 10:15 AM, Rallavagu wrote: As Erick indicated, these settings are incompatible with Near Real Time updates. With those settings, every time you commit and create a new searcher, Solr will execute up to 1000 queries

Re: solr extends query

Ah, I see - thanks for explaining that you’re not operating on tokens like that. Given that, I think the best place to implement this is as a QParserPlugin - that gives you the query string and allows you to return a standard Lucene Query. Erik > On Jul 25, 2016, at 9:44 AM, sara

Re: facet.query on text - recommended field type

facet.query is really just a short-cut for numFound using that query standalone. How many facet.query’s are you issuing? And what is the QTime for all those queries when individually made like this: /select?q==0 If one of your queries is “slow” - you mention wildcards and complex phrase

Re: solr 5.5.2 loadOnStartUp does not work

Hello, Thanks for your answer. Yes, it seems a little tricky to me. Best regards, Elisabeth 2016-07-25 18:06 GMT+02:00 Erick Erickson : > "Load" is a little tricky here, it means "load the core and open a > searcher. > The core _descriptor_ which is the internal

facet.query on text - recommended field type

I am experiencing very slow query time when using multiple facet.query on large result sets (like 5 minutes). I don't know how to optimize this, since it is not completely clear to me how the facet.query works. Currently my facet queries use tokenized text fields and contain wildcards or even use

Re: solr.NRTCachingDirectoryFactory

On 7/22/2016 10:15 AM, Rallavagu wrote: > size="5000" > initialSize="5000" > autowarmCount="500"/> > > size="2" > initialSize="2" > autowarmCount="500"/> As Erick

solr-6.1.0 - Using different client and server certificates for authentication doesn't work

Hello. I have setup Solr 6.1.0 to use SSL (on Windows) and to do client authentication based on the client certificate. When I use the same certificate for both the server and the client authentication, everything works OK :

Re: Issue with Block Join Parent Query Parser

How it's parsed? You can check with debugQuery=true On Tue, Jul 26, 2016 at 10:53 AM, Zheng Lin Edwin Yeo wrote: > Hi, > > I'm using Solr 6.1.0 > > Would like to find out, can we use the Block Join Parent Query Parser to > filter the parents when I search for a field in

group.ngroups=true does not work well in distributed searches

Hi , I am using solr 4.7.2 verison. I am using group.ngroups=true in my solr queries and recently shifted to distributed architeture of solr, where i add shards param in my solr queries as follow :

Issue with Block Join Parent Query Parser

Hi, I'm using Solr 6.1.0 Would like to find out, can we use the Block Join Parent Query Parser to filter the parents when I search for a field in the child document? For example, when I just filter by child query like this, I get 8 results. q={!parent

Re: Using log4j.xml in Solr6

After a bit of testing I got it working. Basically all the configuration for log4j by default is under server/resources/log4j.properites. By default log4j should be able to find log4j.xml if you delete log4j.properties. I tried it and that's no the case. I figured out this is due the fact that