Re: Migrating from cores to collections

I hope, to complete https://issues.apache.org/jira/browse/SOLR-7188 that makes DIH a world class ETL in near future. For those who already faced the bottleneck there is a kind of steroids https://issues.apache.org/jira/browse/SOLR-3585 . On Tue, Dec 1, 2015 at 9:05 PM, Erick Erickson

Help With Phrase Highlighting

Hello everyone, I am having difficulty enabling phrase highlighting and am hoping someone here can offer some help. This is what I have currently: Solr 4.9 solrconfig.xml (partial snip) xml explicit 10

Re: Create Collection Admin Request - unable to specify collection configName

Thanks - the only thing under cloud/tree tab for /configs is mycollection But I am pretty sure my zookeeper config upload is working fine ./server/scripts/cloud-scripts/zkcli.sh -cmd upconfig -confdir ./server/solr/configsets/scbe_public7_config -confname scbe_public7 -z zk.zk.zk.zk:2181

Create Collection Admin Request - unable to specify collection configName

Context: Solr 5.3.1 with ZooKeeper 3.4.6 (SolrCloud) Via the REST APU I am trying to create a collection and tie it to a configuration I have loaded into ZooKeeper Here are the configs loaded into ZooKeeper [zk: localhost:2181(CONNECTED) 5] ls /configs [scbe_public7, mycollection,

RE: Why do documents without the search query term rank highest

WOW! Thanks Chris - I have read your feedback but I will need to go through it a couple more times to get my head around it :) - thanks for taking the time to help - much appreciated! Thanks Stuart PMP, Business Technical Analyst | CRS Consultant | Corporate IT Digital | McDonald's

Re: Create Collection Admin Request - unable to specify collection configName

You shouldn't have to do the linkconfig to see the configs in the ZK tree, and it should appear virtually instantaneously. A couple of possibilities: 1> your browser is caching the view somehow (unlikely). 2> your Solr instance isn't pointing to the same ZK your upconfig command is pointing to.

Re: Create Collection Admin Request - unable to specify collection configName

So I have an ensemble of three Zk nodes running I have tried upconfig to all three ZooKeeper nodes (1 leader, 2 followers) and they seem to work but when I look at cloud/tree it doesn’t make a difference. I tried switching browsers to avoid the potential browser caching issue and still no dice.

RE: Why do documents without the search query term rank highest

: Again, my confusion is why the document 'Home' appears ahead of the : document 'Big Mac' in the ranking when the query term 'big' only appears : once in 'Home' but several times in 'Big Mac'? The key to understanding how documents are scored is in the query structure and the "explain"

Re: Create Collection Admin Request - unable to specify collection configName

Check via the admin UI - cloud/tree tab. Check inside the config directory that you are attempting to use to see if there is a solrconfig.xml and a schema file. I've sometimes uploaded it such that there's a conf directory inside - i.e. one layer too deep. This is gonna be something trivial, I

Re: Create Collection Admin Request - unable to specify collection configName

Try with the conf directory in place with schema.xml & solrconfig.xml inside conf. You should see the files being uploaded ./server/scripts/cloud-scripts/zkcli.sh -cmd upconfig -confdir *./server/solr/configsets/scbe_**public7_config/conf* -confname scbe_public7 -z zk.zk.zk.zk:2181 This is how

Re: Create Collection Admin Request - unable to specify collection configName

bq: I have tried upconfig to all three ZooKeeper nodes (1 leader, 2 followers) and they seem to work but when I look at cloud/tree it doesn’t make a difference. That doesn't answer how you start _Solr_. My question earlier was trying to insure that your Solr instance points at the same zookeeper

Re: Create Collection Admin Request - unable to specify collection configName

Hi, There is no conf directory- all of my files are under the directory name specified $ ls -1 ./server/solr/configsets/scbe_public7_config/ currency.xml lang protwords.txt _rest_managed.json schema.xml solrconfig.xml stopwords.txt synonyms.txt Thanks! -Frank Frank Kelly Principal

Re: Create Collection Admin Request - unable to specify collection configName

Thanks for the suggestion I tried both $ ./server/scripts/cloud-scripts/zkcli.sh -cmd upconfig -confdir ./server/solr/configsets/scbe_public7_config -confname scbe_public7 -z 54.xx.yy.84:2181,54.aa.bb.254:2181,54.cc.dd.124:2181 $ ./server/scripts/cloud-scripts/zkcli.sh -cmd upconfig -confdir

Re: Create Collection Admin Request - unable to specify collection configName

And one more thing I noticed you are specifying your ensemble as zk.zk.zk.zk:2181 . But it should be of the form *zk1:port,zk2:port,zk3:port *. On Tue, Dec 1, 2015 at 8:12 PM, Manikandan Sivanesan wrote: > > Try with the conf directory in place with schema.xml &

Re: Help With Phrase Highlighting

Hi, Did you try hl.mergeContiguous=true ? On Tue, Dec 1, 2015 at 3:36 PM, Teague James wrote: > Hello everyone, > > I am having difficulty enabling phrase highlighting and am hoping someone > here can offer some help. This is what I have currently: > > Solr 4.9 >

Re: Create Collection Admin Request - unable to specify collection configName

Thanks Erick, Yes I start Solr as follows $ ./bin/solr start -cloud -z 54.XX.XX.84:2181,54.XX.XX.124:2181,54.XX.XX.254:2181/solr -h 52.XX.XX.13 And confirmed those are the ZooKeeper nodes in question. Good idea on the downconfig and I confirmed it is working e.g. $

Re: Help With Phrase Highlighting

Hello, Thanks for replying! I tried using it in a query string, but without success. Should I add it to my solrconfig? If so, are there any other hl parameters that are necessary? -Teague > On Dec 1, 2015, at 9:01 PM, Philippe Soares wrote: > > Hi, > Did you try

RE: Why do documents without the search query term rank highest

Thank you! Thanks Stuart PMP, Business Technical Analyst | CRS Consultant | Corporate IT Digital | McDonald's Corporation 2111 McDonald's Drive | Oak Brook, IL 60523 USA Office: +1 630.623.5950 | Cell: 301.633.3298 | stuart.scot...@us.mcd.com -Original Message- From: Upayavira

Re: Why do documents without the search query term rank highest

: I would suggest you ask on a forum related to Adobe CQ. There are many : ways in which CQ could be issuing queries against Solr, and without : insight into that, people here aren't that likely to be able to help you : - unless they happen to also use CQ, which probably amounts to a very : small

Re: Why do documents without the search query term rank highest

I would suggest you ask on a forum related to Adobe CQ. There are many ways in which CQ could be issuing queries against Solr, and without insight into that, people here aren't that likely to be able to help you - unless they happen to also use CQ, which probably amounts to a very small portion of

RE: Why do documents without the search query term rank highest

Thank you for the feedback - I will need some time to put together the response to your suggestions - and you're right, I did get the search URL wrong - just a beginner at this! Thanks Stuart PMP, Business Technical Analyst | CRS Consultant | Corporate IT Digital | McDonald's Corporation 2111

Re: ZooKeeper nodes die taking down Solr Cluster?

Hi Frank, Seems like OOM is reason why Solr failed. What is size of your collection and heap size. You could do thread and heap dump and see what is taking memory. Not sure how threads are named but hope that thread-1103 does not mean there are 1103 threads reading ZK. Regards, Emir On

Re: Is there a way to set zkClientTimeout from command line?

Wonderful - thanks for the clarification -Frank On 12/1/15, 12:48 PM, "Chris Hostetter" wrote: > >Custmizations are what solr.in.sh is designed for. > >This is covered in some depth in the ref guide... >

Fwd: Indexing rich data (msword and pdf) in apache solr-5.3.1

I start working in solr 5x by extract solr in D://solr and run solr server with : D:\solr\solr-5.3.1\bin>solr start ; Then I create a core in standalone mode : D:\solr\solr-5.3.1\bin>solr create -c mycore I need indexing from system files (word and pdf) and the schema API don’t have a field

Fwd: index rich data with solarium php solr Client

I get this Error Invalid character encoding detected after position 79 of query string / form data (while parsing as UTF-8) this is my function to index rich data from

Re: Maximum in Multivalued field

Solr will easily handle 1,000. But note a couple of things: 1> There's no _requirement_ that you use multiValued fields. If it's a text field and you use minimal analysis, you can just shove them all into a single value. Confusing I know. MultiValued means that this is accepted (xml format of a

RE: Why do documents without the search query term rank highest

Hi I think this is what you requested - the query 'big' as entered into the Solr dashboard with Debug applied - I have also attached the Schema and Config files Again, my confusion is why the document 'Home' appears ahead of the document 'Big Mac' in the ranking when the query term 'big' only

Re: Is there a way to set zkClientTimeout from command line?

Custmizations are what solr.in.sh is designed for. This is covered in some depth in the ref guide... https://cwiki.apache.org/confluence/display/solr/Taking+Solr+to+Production Alternatively, for quick dev purposes, you can override it as an env variable when execing the script just like any

Re: Migrating from cores to collections

Yes, DIH works with SolrCloud. I don't particularly like it as it doesn't parallelize well, i.e. all the action happens on one Solr server. Admittedly it does send the docs to the correct shards etc. But often the bottleneck becomes acquiring the data, and there DIH will be the bottleneck since

Re: optimize cache-hit-ratio of filter- and query-result-cache

1.1> Absolutely. The filterCache is simply a map. The key is the fq clause, so fq=field1:(1 OR 2 OR 3) is different than fq=field1:(3 OR 2 OR 1). 2.1> not sure. But don't get very hung up on queryResultCache. It's useful pretty much for paging and the hit ration is often very low as it only gets

Re: Create Collection Admin Request - unable to specify collection configName

Why are you adding the /solr to the -z parameter when you start Solr? You never specify it for upconfig/downconfig and the like. Of course I'm almost always in dev and have only a single ZK running. And when I just tried adding /solr the results were a completely useless admin screen that hung

Re: Create Collection Admin Request - unable to specify collection configName

If I'm carefully looking at your commands, the IP are different. For upconfig "54.XX.XX.124:2181" For downconfig, you used 54.XX.XX.84:2181 Are these zookeeper on an ensemble setup? Anyways, to avoid confusion, what I'd recommend is, just spin one only zookeeper first, start solr by passing

Re: Help With Phrase Highlighting

Hi Teague, I couldn't understand the part of "document size" in your question, but if you'd like Solr to return snippet My search phrase instead of My search phrase you should use FastVectorHighlighter. In case use of FVH, your highlight field (hl.fl=text) need to be indexed with options

Re: Create Collection Admin Request - unable to specify collection configName

Adding /solr to the zk string 'namespaces' the data within a sor directory inside zookeeper, which is a useful feature. It allows you to share zk between multiple applications. However, you must use the same at startup and with zkcli. So either remove the /solr or add it to the zkcli lines also.

Re: Block Joins

Rick, fl=score,[child parentFilter=type_s:book childFilter=*{!score=avg}*that], This childFilter value doesn't make sense. What do yo want to achieve? On Tue, Dec 1, 2015 at 7:28 PM, Rick Leir wrote: > Hi all, > Scoring is confusing me. Is the following correct? > >

RE: Synonyms in Search Results and More Accurate Matches

Hello - it looks like you have synonyms enabled at query time, which is fine, but also means TF*IDF stats are different for tbrush and toothbrush, causing this order to be the way it is. There is no solution available in Solr right now that would boost user-entered terms over expanded synonyms

Re: optimize cache-hit-ratio of filter- and query-result-cache

Thanks. The statements on http://wiki.apache.org/solr/SolrCaching#showItems are not explicitly enough for my question.

Re: Migrating from cores to collections

I've never used DIH in earnest. I'm not sure if/how it works with SolrCloud. There is a ticket somewhere to make a 'standalone' DIH that sits outside of Solr and pushes to it, which would be a much better idea, I think, and would work better with SolrCloud. Others here (perhaps in a separate

Re: Synonyms in Search Results and More Accurate Matches

One technique that works well is to use copyField to end up with two indexed fields, one with synonyms, one without. Then you can qf=title^5 title_with_synonyms^1 with edismax and weight the “exacter” field higher than one with synonyms. — Erik Hatcher, Senior Solutions Architect

Re: Synonyms in Search Results and More Accurate Matches

Index-time synonym expansion maximizes recall (not missing any documents, especially partial matches), but minimizes precision and relevancy - you are unable to select or boost exact matches. Ditto for ngrams. As Erik indicates, using edismax with separate fields for precision (exact matches) and

Re: SolrCloud Shard + Replica on Multiple servers with SolrCloud

Answers inline On Tue, Dec 1, 2015, at 06:03 AM, Adrian Liew wrote: > Hi all, > > Will really like to seek anyone's opinion on my query below. Desperate to > know if this is possible or if someone is keen to share their thought > experience. > > Best regards, > Adrian > > > -Original

Re: ZooKeeper nodes die taking down Solr Cluster?

Thanks Emir - responses inline below >Can you please confirm that Solr nodes are aware of entire ZK ensemble? Can you explain how I could find that out - I looked into the logs and the Admin UI and didn¹t see A way to examine if the Solr nodes saw the entire ensemble >Can you give more info how

Is there a way to set zkClientTimeout from command line?

I am executing "/bin/solr" for SolrCloud and noticed that there is a Bash parameter that is being inspected but is never being set except by default if [ -z "$ZK_CLIENT_TIMEOUT" ]; then ZK_CLIENT_TIMEOUT="15000" #I would like this to be settable from command line fi

Re: ZooKeeper nodes die taking down Solr Cluster?

Hi Frank, Can you please confirm that Solr nodes are aware of entire ZK ensemble? Can you give more info how it is deployed - ZK on separate servers? What is load on Solr when it happens? Do you see any errors in Solr logs? Thanks, Emir -- Monitoring * Alerting * Anomaly Detection *

Re: Synonyms in Search Results and More Accurate Matches

I do not have synonyms enabled at query time. Below is my fieldtype On Tue, Dec 1, 2015 at 4:18 AM, Markus Jelsma wrote: > Hello - it looks like you have synonyms enabled at query time, which is > fine, but also means TF*IDF stats are different for

Re: Block Joins

On Mon, Nov 30, 2015 at 11:47 PM, wrote: > > I'd happy that it helps. You need to thank Varun who made [child]. There is > no performance impact for sure, it's a compromise or either nobody > considered it as an issue. You can raise a new JIRA and

Maximum in Multivalued field

We are considering using a multivalued field that can contain up to 1000 unique values. This field will not be searched on but just used in facet and filter. What is the maximum number of values that a multivalued field can contain? Is there another more efficient way of doing this? Thanks

Re: Block Joins

Hi all, Scoring is confusing me. Is the following correct? $ curl http://localhost:8983/solr/dorsetdata/query -d ' q={!parent which="content_type:parentDocument" *score=max*} type_s:page AND that& wt=json=true& fl=score,[child parentFilter=type_s:book childFilter=*{!score=avg}*that

Why do documents without the search query term rank highest

Hi All, I hope this is the way to ask a question - please guide me if there is a different protocol I have a question about results ranking for Solr V4.2 in combination with the CMS tool Adobe CQ (V5.6). Despite trying different ways to configure the ranking of documents I am confused why