You can set solr home by passing -Dsolr.solr.home=/path/to/solr/home when
running your container (the double solr is *not* a typo)
You can also use JNDI to configure it. Look under the "Installation and
Configuration" section in the wiki for container-specific instructions:
http://wiki.apache.org/
I think it is relative to solr home. Just to be safe, you can put absolute
paths to the bin directory in both "exe" and "dir" sections.
On Fri, Aug 8, 2008 at 4:37 AM, Jacob Singh <[EMAIL PROTECTED]> wrote:
> I tried:
>
>
> snapshooter
> bin
> true
> arg1 arg2
> MYVAR
If you specify single,
write.lock will never be created, so makes no sense.
Koji
sundar shankar wrote:
WIll let u know the results in a bit.
What does single do, btw?
Do i need to use this in conjunction with or do i use it separately??
Date: Thu, 7 Aug 2008 12:26:53 -0400> From: [EM
In our implementation we have setup snappuller to check for updated
indexes every 10 mins.
After checking message logs (/var/log/messages) we see that the
snappuller script is opening and closing ssh sessions multiple times
within a very short timeframe, in one case an ssh session is being
opene
I see in the docs:
System property substitution
Solr supports system property substitution, allowing the launching JVM
to specify string substitutions within either of Solr's configuration
files. The syntax ${property[:default value]}. Substitutions are valid
in any element or attribute text. Her
On Thu, Aug 7, 2008 at 8:36 PM, CameronL <[EMAIL PROTECTED]> wrote:
> Does this seem like an issue that might need to be escalated to jira?
Absolutely.
> We
> might be able to provide you some additional info relating to these threads
> locking if you need it.
Yes, some thread dumps or anything
Yonik Seeley wrote:
>
> On Thu, Aug 7, 2008 at 7:32 PM, CameronL <[EMAIL PROTECTED]>
> wrote:
>>
>> What is the current strategy for doing distributed search for Solr? We
>> currently have our index divided over 3 servers:
>> solr1
>> solr2
>> solr3
>>
>> In order to balance the load, our appli
On Thu, Aug 7, 2008 at 7:32 PM, CameronL <[EMAIL PROTECTED]> wrote:
>
> What is the current strategy for doing distributed search for Solr? We
> currently have our index divided over 3 servers:
> solr1
> solr2
> solr3
>
> In order to balance the load, our application calls each with a 'shards'
> p
: I would like to give it a shot. Are there any solr logo success
: criteria/requirements? Any hints or suggestions from community is welcomed.
: Just close your eyes, start dreaming and send my couple of words about what
: you see... I am all ears.
Other people who care more about the logo have
What is the current strategy for doing distributed search for Solr? We
currently have our index divided over 3 servers:
solr1
solr2
solr3
In order to balance the load, our application calls each with a 'shards'
parameter to get the data:
solr1:8080/select?shards=solr1,solr2,solr3...
solr2:8080/s
: is there a way to get key terms from an item? If each item has an id, can I
: pass that ID to a search and get back the key terms like you can with the
: mlt filter.
as long as your id field is searchable, and if you are talking about your
uniqueKey field then it has to be, absolutely.
-Ho
I tried:
snapshooter
bin
true
arg1 arg2
MYVAR=val1
However, it doesn't seem to work. I assumed that it default to the
solrhome, but that doesn't seem to be the case...
Is it defaulting to the jetty root? Or the conf dir?
Thanks,
Jacob
Its not a should, it is recommended that it aint happening as a syncronous
process out your app. From what I remember, When I had docs in the order of
100,000+, my optimize took 70 secs. Commit, took 5-6 I guess. Not sure. it was
some mistake I did long time back.
It really depends on what kin
Thank you so much everyone.
This community is really one of the most helpful I've ever run across on
the interwebs.
So I understand that the document is not live in the index until a
commit is run.
Commits should be run nightly.
Isn't this a problem though to have to wait 24hrs for a new docume
Yes commits are very expensive and optimizes are even expensive.
Coming to your question of numdocs and 0's in update handler section
The numdocs that u see on top are the ones that are committed. The ones u see
below are the ones u have updated, not committed.
update handlers
==com
Okay,
So somehow, when I ran the next batch, it did fire a commit and an
optimize according to the logs I finally got working and the stats page.
Ahh... I think I get it.
the bin/commit script fires a commit for you, it's not what is run on
commit. And if you want replication to work, you have
Alexander Ramos Jardim wrote:
Oh,
Sorry!
Can you be a little more specific? Do these indexes have different schemas,
or do they represent the same data model?
If they have different schemas, MultiCore support is definitely worth
looking at (again, requires 1.3-ish Solr code, as it was adde
Both the cases are there. As i said i need to index 3 indexes. So 2 indexes
have same schema but other one has different. More specification is like
this --
I have 3 indexes. In which 2 indexes have same data model but the way these
are indexed is different. So i need to fire query from backend o
Oddly, I have 10 docs in the index, as indicated by NumDocs and MaxNumDocs
However, in the Update Handlers section:
commits : 0
autocommits : 0
optimizes : 0
docsPending : 0
All 0s...
Any ideas?
The only idea: you restarted server after 'commit' was issued; you
didn't add any new docs
Okay... so this all makes sense, and I can sorta make it work, however,
the commit is never run by itself when the index is updated. I have to
manually call the commit script from the command line to get it to happen.
Jacob, 'commit' is expensive (you will loose SOLR caches, and etc.)...
'com
I'm using the PHP library provided in JIRA, and I am calling commit()
from it. Unfortunately, I can't see if it is working or not...
Yes, you can't see changes immediately - it requires some time for new
Searcher to open and to warm SOLR caches, several minutes-hours
depending on index si
Hi Sundar,
Oddly, I have 10 docs in the index, as indicated by NumDocs and MaxNumDocs
However, in the Update Handlers section:
commits : 0
autocommits : 0
optimizes : 0
docsPending : 0
All 0s...
Any ideas?
I also tried to follow the instructions at:
http://wiki.apache.org/solr/LoggingInD
Oh,
Sorry!
Can you be a little more specific? Do these indexes have different schemas,
or do they represent the same data model?
2008/8/7 anshuljohri <[EMAIL PROTECTED]>
>
> Thanks zayhen for such a quick response but am not talking about sharding.
> I
> have requirement of indexing 3 indexes.
Thanks zayhen for such a quick response but am not talking about sharding. I
have requirement of indexing 3 indexes. Need to do query on diff indexes
based on input.
-Anshul
zayhen wrote:
>
> 2008/8/7 anshuljohri <[EMAIL PROTECTED]>
>
>>
>> Hi everybody!
>>
>> I need to create multiple indexes
Look at the update handlers section of the Solr stats page. I guess the url is
/admi/stats.jsp. This woudld give u an idea of how many docs are pending commit.
> Date: Thu, 7 Aug 2008 14:53:02 -0700> From: [EMAIL PROTECTED]> To:
> solr-user@lucene.apache.org> Subject: Re: How do I configure comm
Hi folks,
Thanks for the replies. I think I'm getting closer now.
I'm using the PHP library provided in JIRA, and I am calling commit()
from it. Unfortunately, I can't see if it is working or not...
I'm trying to get logging working in jetty for a multi-instance server,
but that's a separate n
2008/8/7 anshuljohri <[EMAIL PROTECTED]>
>
> Hi everybody!
>
> I need to create multiple indexes lets say 3 due to project requirement.
> And
> the query will be fired from backend on different indexes based on input. I
> can't do it in one index with the help of "fq" parameter. As i have already
Or time.
1 1
> Date: Thu, 7 Aug 2008 18:42:30 -0300> From: [EMAIL PROTECTED]> To:
> solr-user@lucene.apache.org> Subject: Re: How do I configure commit to run
> after updates> > You can configure the autocommit feature in solrconfig.xml
> to get commit to>
You can configure the autocommit feature in solrconfig.xml to get commit to
work from time to time or based in the number of documents added to your
index.
2008/8/7 Jacob Singh <[EMAIL PROTECTED]>
> Hi,
>
> I'm using the XML based update interface, and feeding requests to update
> the index via j
Hi everybody!
I need to create multiple indexes lets say 3 due to project requirement. And
the query will be fired from backend on different indexes based on input. I
can't do it in one index with the help of "fq" parameter. As i have already
thought on it but thats of no use.
So i searched a lo
I'm not exactly a long-time user of Solr yet so forgive me if I'm
missing something really obvious here... but, if I'm not mistaken, you
just add to the end of your update XML. This is how I've
done all of my commits, I wasn't even aware there was a CLI method to
this.
I would worry abo
Hi,
I'm using the XML based update interface, and feeding requests to update
the index via jetty. It all works great, however now I'm trying to get
replication running, and here's what I understand:
1. An index update comes in.
2. Solr runs the commit script
3. a post-commit event is specified i
2008/8/6 Henrib <[EMAIL PROTECTED]>
>
> My bad, sorry, I read too fast and skipped the fist necessary which is to
> get
> solr.home through JNDI which you can't have nor set I presume... (which
> from
> there would allow reading a multicore.properties, etc...)
The most irritating thing is that he
WIll let u know the results in a bit.
What does single do, btw?
Do i need to use this in conjunction with or do i use it
separately??
> Date: Thu, 7 Aug 2008 12:26:53 -0400> From: [EMAIL PROTECTED]> To:
> solr-user@lucene.apache.org> Subject: Re: Unlock on startup> > On Thu, Aug 7,
> 2008
On Thu, Aug 7, 2008 at 12:17 PM, sundar shankar
<[EMAIL PROTECTED]> wrote:
> I had the war created from JUly 8th nightly. Do u want me to take the latest
> and try it out??
Yes, please.
-Yonik
I had the war created from JUly 8th nightly. Do u want me to take the latest
and try it out??
> Date: Wed, 6 Aug 2008 20:46:59 -0400> From: [EMAIL PROTECTED]> To:
> solr-user@lucene.apache.org> Subject: Re: Unlock on startup> > What date did
> you get 1.3-dev?> I committed a patch for this on J
Nope I dont see any error logs when my Jboss starts up. I havent added solr
classes to my log4j.xml though. Should I add them and try again?
What does single do, btw? Do i need to use this in
conjunction with or do i use it separately??
> Date: Wed, 6 Aug 2008 16:57:23 -0700> From: [EMAIL
Try Bojan Smid's latest patch from SOLR-236.
Otis
--
Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch
- Original Message
> From: Vaijanath N. Rao <[EMAIL PROTECTED]>
> To: solr-user@lucene.apache.org
> Cc: [EMAIL PROTECTED]
> Sent: Thursday, August 7, 2008 9:03:44 AM
> Subjec
http://www.tokenizer.org/www.vanns.com/price.htm?q=
- you can manage via robots.txt if you want to stop Tokenizer...
Quoting Fuad Efendi <[EMAIL PROTECTED]>:
Your shopping site rocks, I like it: http://www.vanns.com
From SEO perspective, you have two links with anchor texts (see
http://www.va
Your shopping site rocks, I like it: http://www.vanns.com
From SEO perspective, you have two links with anchor texts (see
http://www.vanns.com/shop/servlet/category/desc/45/1/gps/):
Garmin -> http://www.vanns.com/shop/servlet/brand/52/1/garmin/
nuvi 200 -> http://www.vanns.com/shop/servlet/ite
Hi All,
I am using the Solr (1.3) nightly build. I need to know where should I
make the changes in either config or the way I query the Solr to achive
a) collapse the documents on a field
I reffered to
http://wiki.apache.org/solr/FieldCollapsing?highlight=%28collapse%29,
but still did not ge
Why not add a feature where in the top of search results where you have the
most sold product that is related with your search? Or a feature like Amazon
that points which other related itens the users bought with this product.
I would like to see a price range facet too. A very important feature f
Thanks for the feedback and corrections - the definition was wrong
indeed. So, I have settled on this definition:
On Thu, 7 Aug 2008 00:50:59 -0700 (PDT)
matt connolly <[EMAIL PROTECTED]> wrote:
> Where do I file a bug report?
https://issues.apache.org/jira
thanks!
B
_
{Beto|Norberto|Numard} Meijome
Contrary to popular belief, Unix is user friendly. It just happens to be very
sele
On Wed, 6 Aug 2008 21:35:47 -0700 (PDT)
Otis Gospodnetic <[EMAIL PROTECTED]> wrote:
>
>
>
>
> 2 Tokenizers?
i wondered about that too, but didn't have the time to test...
B
_
{Beto|Norberto|Numard} Meijome
"Always listen to experts. They'll t
I found a bug in the HTML Standard Strip filter where it doesn't place word
boundaries at html tags that should be ends of blocks.
I've just discovered that if I index some text like this:
titlesome text
it is stripped and indexed as "titlesome" and "text". Putting a space or
newline between th
46 matches
Mail list logo
