Java OutOfmemory error during autowarming

During auto-warming of solr search on QueryResultKey, Our Production solrslaves errors throw OutOfMemory error and application need to be bounced. Here is the error logs: SEVERE: Error during auto-warming of key:org.apache.solr.search.queryresult...@9acd5d67

Re: Question about field types and querying

Check if your text field definition in schema.xml has a EdgeNGramFilter applied to it. Cheers Avlesh On Fri, May 29, 2009 at 12:37 AM, ahammad wrote: > > Hello, > > I have a field type of "text" in my collection called "question". > > When I query for the word "customer" for example in the "que

Fwd: java.lang.RuntimeException: after flush: fdx size mismatch

An update:With assertions enabled, I still get fdx size mismatch exceptions in the same manner as before. Without assertions, the fdx size mismatch exception was followed by a chain of LockObtainFailedExceptions. With assertions, there is also a chain of NullPointerExceptions following the LockObt

Fwd: java.lang.RuntimeException: after flush: fdx size mismatch

My apologies for the delay in running this patched Lucene build - I was temporarily pulled onto another piece of work. Here is a sample 'fdx size mismatch' exception using the patch Mike supplied: SEVERE: java.lang.RuntimeException: after flush: fdx size mismatch: 1 docs vs 0 length in bytes of _

Re: Map tika attribute to be the id in Solr Cell

Updating to latest and greatest added that data, thank you for the pointer. Too many copies of Solr 1.4 trunk, and I'd neglected to update. However, the issue with the mapping not working with the ext.metadata.prefix seems to remain: budapest:site epugh$ curl http://localhost:8983/solr/

Re: Recover crashed solr index

hi Mike and Yonik , There is no other exceptions other than missing segments_cje. the disk still have 160G more spaces when the disk crash. current these is only one solr core that accessing this index directory. so it can't be the file lock problem. I am not sure is considered as unusual config

Re: Map tika attribute to be the id in Solr Cell

Grant, I went back and tried to recreate my bug using the example app. And indexing example/site/tutorial.pdf I get the error with this command: budapest:site epugh$ curl http://localhost:8983/solr/update/extract?ext.def.fl=text \&ext.metadata.prefix=metadata_\&ext.map.stream_name=id -F "

Re: Map tika attribute to be the id in Solr Cell

On May 28, 2009, at 8:47 PM, Eric Pugh wrote: Grant, you are quite right! I was too far down in the weeds, and didn't need to be doing all that crazyness. And I don't actually see the metadata fields. I would expect to however! What revision are you running? The following was added

Re: Map tika attribute to be the id in Solr Cell

Grant, you are quite right! I was too far down in the weeds, and didn't need to be doing all that crazyness. However, one other comment, I saw you edited the wiki (thank you!) and the line: + It is highly recommend that you try using the extract only option to see what values actually g

Re: Map tika attribute to be the id in Solr Cell

On May 28, 2009, at 11:29 AM, Eric Pugh wrote: Hi all, I want to use the Tika attribute stream_name as my unique key, which I can do if I specify stream_name and run curl: curl http://localhost:8983/solr/karaoke/update/extract? ext.def.fl=text\&ext.capture=stream_name \&ext.map.stream_

Re: Question about field types and querying

Try using the admin analysis tool (http://:/solr/admin/analysis.jsp) too see what the analysis chain is doing to your query. Enter the field name ("question" in your case) and the Field value (Index) "customize" (since that's what's in the document). For Field value (Query) enter "customer". Check

Re: Recover crashed solr index

On Wed, May 27, 2009 at 5:49 AM, Michael McCandless wrote: > Hmm... so in fact it looks like Solr had done a number of commits > already (especially, given how large your generation is -- the "cje" > in segments_cje means there were a number of commits). > > Were there any other exceptions leading

Re: index version command

Cool. On Thu, May 28, 2009 at 11:10 PM, Otis Gospodnetic < otis_gospodne...@yahoo.com> wrote: > > View the source of that page, it should be all XML, so you can easily parse > it. > > Otis > -- > Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch > > > > - Original Message > > Fr

Re: index version command

View the source of that page, it should be all XML, so you can easily parse it. Otis -- Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch - Original Message > From: nk 11 > To: solr-user@lucene.apache.org > Sent: Thursday, May 28, 2009 2:54:08 PM > Subject: Re: index version

Question about field types and querying

Hello, I have a field type of "text" in my collection called "question". When I query for the word "customer" for example in the "question" field (ie q=question:customer), the first document with the highest score shows up, but does not contain the word customer at all. Instead, it contains the

Re: index version command

Thanks. So I should parse the html result? I was thinking about something more "machine readable". On Thu, May 28, 2009 at 9:01 PM, Otis Gospodnetic < otis_gospodne...@yahoo.com> wrote: > > Hello, > > You can see the version of the index in the Solr Admin page. > > Otis > -- > Sematext -- http://

Re: index version command

Hello, You can see the version of the index in the Solr Admin page. Otis -- Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch - Original Message > From: nk 11 > To: solr-user@lucene.apache.org > Sent: Thursday, May 28, 2009 10:29:10 AM > Subject: index version command > >

Re: job ads ok?

You might send it to j...@apache.org too. --tim On Thu, May 28, 2009 at 1:44 PM, Jodi Showers wrote: > I'll take that as a positive. > > Homestars is looking for a contract developer to aid us with our present > solr (localsolr) install. > > The problems relate to boosting the resulting sort ord

Re: job ads ok?

I'll take that as a positive. Homestars is looking for a contract developer to aid us with our present solr (localsolr) install. The problems relate to boosting the resulting sort order. In addition to this specific challenge, there are features we'd like to add to the current localsolr a

Re: job ads ok?

No, that would be a violation of netiquette. Please just send them directly to me. On Thu, May 28, 2009 at 10:06 AM, Jodi Showers wrote: > Greetings, > > Is is ok to post solr job ads here? > > thanks. > Jodi >

job ads ok?

Greetings, Is is ok to post solr job ads here? thanks. Jodi

Map tika attribute to be the id in Solr Cell

Hi all, I want to use the Tika attribute stream_name as my unique key, which I can do if I specify stream_name and run curl: curl http://localhost:8983/solr/karaoke/update/extract?ext.def.fl=text \&ext.capture=stream_name\&ext.map.stream_name=stream_name -F "fi...@angeleyes.kar " However

Regarding More Like This Handler

Hi, I am using Solr.In that i am using Standered Request handler. *-* < requestHandler name="*standard*" class="*solr.SearchHandler*" default="*true *"> *-* *-* * * *

Re: Solrits - Sort Buttons under the searchform

On May 28, 2009, at 10:35 AM, Jörg Agatz wrote: In the First, i tryed to print the Respons on the Display.. i tryed with this Code: SORT HIER wird SORTIERT you search for $(response.responseHeader.params.q) But it dosent Work, hi only Print: SORT HIER wird SORTIERT you search

Re: Solrits - Sort Buttons under the searchform

In the First, i tryed to print the Respons on the Display.. i tryed with this Code: SORT HIER wird SORTIERT you search for $(response.responseHeader.params.q) But it dosent Work, hi only Print: SORT HIER wird SORTIERT you search for $(response.responseHeader.params.q)

Re: Solrits - Sort Buttons under the searchform

On May 28, 2009, at 9:52 AM, Jörg Agatz wrote: I Tryed to create a Sort button unter the SearchForm... i would like to sort : "score" and "date" so i musst use the Link: * http://192.168.105.54:8983/solr/itas?q=SEARCHWORD&sort=date asc* and *http://192.168.105.54:8983/solr/itas?q=SEARCHWORD&s

Solrits - Sort Buttons under the searchform

Hollo.. I Tryed to create a Sort button unter the SearchForm... i would like to sort : "score" and "date" so i musst use the Link: * http://192.168.105.54:8983/solr/itas?q=SEARCHWORD&sort=date asc* and *http://192.168.105.54:8983/solr/itas?q=SEARCHWORD&sort=date desc * and the Same with score,

Re: lock issue

Are you using a shared network disk? If you are, that is going to be very slow, much slower than using Solr over HTTP. --wunder\ On 5/28/09 2:58 AM, "Shalin Shekhar Mangar" wrote: > On Thu, May 28, 2009 at 3:14 PM, Ashish P wrote: > >> >> Hi Shalin, >> I am trying to do the same point again.

Re: lock issue

Andrey Klochkov wrote: Even if you point multiple embedded solr servers to the same index, you should write with only one. Once you do a commit on the writer, you'd need But it's what lucene index locking is for, isn't it? Locking should handle that issue. Solr doesn't use index locking i

Re: Problems getting up and running.

2009/5/28 Koji Sekiguchi > How does your look like in solrconfig.xml? > > > ${solr.data.dir:./solr/data} It looks just like this. Which made me notice that this variable isn't being defined in the tomcat init.d script, so I added it. Now it kind of works. I think. Although... /var/solr/dat

Re: Problems getting up and running.

Hello, In the solrconfig.xml file, there is a property: ${solr.data.dir:./solr/data} Try setting something else in here and see what happens...I'm not sure how solr works with Ubuntu, but it's worth a shot... Tim Haughton wrote: > > OK, I spoke too soon. > > When you tried it on

Re: Problems getting up and running.

Tim Haughton wrote: OK, I spoke too soon. When you tried it on your Mac, did it create the index in the right place? Yes, Solr created the index directory in $CATALINA_HOME/bin/solr/data/index as I expected, because I started tomcat from bin directory. (and I didn't specify -Dsolr.solr.home

Re: lock issue

> > Even if you point multiple embedded solr servers to the same index, you > should write with only one. Once you do a commit on the writer, you'd need But it's what lucene index locking is for, isn't it? Locking should handle that issue. Solr doesn't use index locking in a proper way? -- Andr

Re: Problems getting up and running.

OK, I spoke too soon. When you tried it on your Mac, did it create the index in the right place? Mine is still trying to create it under the webapps directory. Cheers, Tim 2009/5/28 Tim Haughton > 2009/5/28 Koji Sekiguchi > >> >> Ok. >> I've just tried it (the way you quoted above) on my Mac

Re: Problems getting up and running.

2009/5/28 Koji Sekiguchi > > Ok. > I've just tried it (the way you quoted above) on my Mac and worked fine... > Do you see any errors on Tomcat log when starting? > Sussed it. As you would imagine it was the stupidest of things. And probably the *one* thing left out of my description. My solr.xm

Re: Problems getting up and running.

Tim Haughton wrote: Hi Koji, quoting from that page: "/some/path/solr.war" is the absolute path to where ever you want to keep the Solr war using the appropriate syntax for your Operating System. In Tomcat 5.5 and later,* the war file must be stored outside of the webapps directory* for this to

Re: Problems getting up and running.

Hi Koji, quoting from that page: "/some/path/solr.war" is the absolute path to where ever you want to keep the Solr war using the appropriate syntax for your Operating System. In Tomcat 5.5 and later,* the war file must be stored outside of the webapps directory* for this to work. Otherwise, this

Re: Problems getting up and running.

If you have webapps directory under /usr/local/tomcat, place solr.war to there. http://wiki.apache.org/solr/SolrTomcat Koji Tim Haughton wrote: Hi, I'm all at sea with this. I'm trying to get Ubuntu 9.04 (x64 Server), Tomcat 6.0.18 and Solr 1.3.0 to play together. So far, no dice. Here's whe

Re: Lucene Query to Solr query

Yeah, it'd be nice if Query could be serialized across the wire and could be deserialized in a new QParser on the Solr side. An alternative to that is to use Lucene contrib's XML query parser. Still a work in progress, but the basics are here: http://issues.apache.org/jira/browse/SOLR-839

Problems getting up and running.

Hi, I'm all at sea with this. I'm trying to get Ubuntu 9.04 (x64 Server), Tomcat 6.0.18 and Solr 1.3.0 to play together. So far, no dice. Here's where I'm at: I installed Java: sudo aptitude install sun-java6-jdk I downloaded the latest Tomcat (6.0.18) and stuck it in /usr/local/tomcat I downlo

Re: term vectors

The main problem with putting TEI specifically into an XQuery engine is that TEI is such a mess! At least the stuff I've seen. A query to find a word in a "div" requires doing //div1 || //div2 || //div3 ... queries (and that's the prettiest example). Having to know the messed up structur

Re: lock issue

On Thu, May 28, 2009 at 3:14 PM, Ashish P wrote: > > Hi Shalin, > I am trying to do the same point again. You suggested to use HTTP for > replication. > But for performance reasons I want to use embedded. What kind of performance are you looking for? We use http with very large indexes and very

Re: lock issue

Hi Shalin, I am trying to do the same point again. You suggested to use HTTP for replication. But for performance reasons I want to use embedded. And because replication is not available with embedded approach, I am trying this i.e. from two different web applications (same network) using embedded

Re: lock issue

On Thu, May 28, 2009 at 2:54 PM, Ashish P wrote: > > Hi, > I am committing to same index from two different embedded servers. > My locktype is simple and writelocktimeout is commitLockTimeout is 10. > > I read in a post "Update from multiple JVMs" where Hoss said this case is > supported but

lock issue

Hi, I am committing to same index from two different embedded servers. My locktype is simple and writelocktimeout is commitLockTimeout is 10. I read in a post "Update from multiple JVMs" where Hoss said this case is supported but I am getting following error. I tried single lock also but agai

Re: solr search

Hi I want to search two words say "delhi india". My default search field is title field . The two words must be searched from the place table only. Also I must get the same result set for both "delhi india" and "india delhi". How can I do? Any idea? Thanks in advance dharhsana wrote: > > hi

Re: Lucene Query to Solr query

That was exactly my point, Reza. As long as the query objects being used are "simple", there is no harm in using toString. But Erik's waring is valid as well. Putting it in SolrJ would not make sense as the behavior of toString is not consistent. Query objects like SpanQuery don't even have a toSt

Re: Lucene Query to Solr query

I found out that toString() works well as long as you use string values and do not use Lucene utility classes like NumberTools.longToString(someLongValue) Maybe nice future to have a Lucene query wrapper in solr package! Reza On May 26, 2009, at 2:18 PM, Reza Safari wrote: That is not what