I think you should read
this https://wiki.apache.org/lucene-java/ScoresAsPercentages
it might help you to make a point.
simon
On Wednesday, November 5, 2014 8:42:59 PM UTC+1, Dustin Boswell wrote:
Is there a way to score documents so that the relevance score has a fixed
range, like from 0
We fixed EdgeNGram tokenizer / filter in the 1.x series but don't ask me
when exactly I think it was lucene 4.4 or so. Those offsets are now correct
while they where broken before.
not sure if this helps you to debug your problem
On Thursday, November 6, 2014 1:31:22 PM UTC+1, Ben George wrote:
Hey,
thanks for raising this. The exception is extremely helpful. This boils
down to a bug / inconsistency in
Lucene where some of the bugfix releases ie. 3.6.1, 4.3.1, 4.5.1 did
specify the minor version in it's version constant
where is shouldn't. I just opened [1] that fixes the issue. It's
FYI 1.3.1 has been release http://t.co/rCL9rAofta
go ahead unless you wanna run on 1.3.1-SNAPSHOT
thanks for raising this again!
simon
On Monday, July 28, 2014 5:17:46 PM UTC+2, Manikandan Pounraj wrote:
Thanks Simon.
We tested with latest build (1.3.1 SNAPSHOT). *All our old data
Peter, thanks so much for raising this. This looks aweful! I think we
should move this into an issue on [1] (please feel free to create one) IMO
we should aim to name the issue in a way to prevent this from happening
altogether. Along the lines we should help you to recover but I don't know
hey,
thanks for raising this, can you gimme more infos ie. which version you are
using and if that happens only on one shard or on all shards in your
system? It could just be what it says, and index corruption maybe due to HW
failure but there could be other reasons
simon
On Friday, June
to be able to eliminate all the elements that are not
['apple', 'banana'].
About the results I'm building an example, and soon as it will be
running I'll send it here, hopefully by the end of the day.
thanks again!
Luca
On Thu, Apr 24, 2014 at 6:42 PM, simonw
simon.w
luca, what do the values in 'importantId` look like? Are they identical to
the values you pass in? The field is not analyzed though. It would also be
helpful if you could show the results that you would expect so we can see
what values they have.
simon
On Thursday, April 24, 2014 5:53:42 PM
hey,
is it possible to look at this index / shard? do you still have it / can
you safe it for further investigations? You can ping me directly at simon
AT elasticsearch DOT com
On Wednesday, April 2, 2014 11:23:38 AM UTC+2, Paweł Chabierski wrote:
Few days ago we found we've got that same
Regarding the corruption - that is a lucene problem that we are working on
to address. The corruption might have happened a while ago already but you
are only happen to see it now. There is no integrity checks in lucene yet
so if there is a corrputed index due to HW failure you will also
I have to thank you for bringing it up! :)
On Wednesday, March 26, 2014 9:40:26 PM UTC+1, Shinsuke Sugaya wrote:
Thank you for the investigation and filing it.
Regards,
shinsuke
On Thursday, March 27, 2014 12:55:56 AM UTC+9, simonw wrote:
I opened an issue for this:
https://github.com
It's actually unrelated to this issue mentioned above. It got broken due
to https://github.com/elasticsearch/elasticsearch/pull/5005
and there is an issue open
here: https://github.com/elasticsearch/elasticsearch/issues/5551 a fix is
right here:
that is actually a bug IMO - Math.abs() can return -1 if it hits
Integer.MIN_VALUE
this code is just broken - can you open an issue!
On Wednesday, March 26, 2014 11:43:20 AM UTC+1, Kevin Wang wrote:
pickIndex() will return the absolute value of the count, so it won't
return a negative
I opened an issue for
this: https://github.com/elasticsearch/elasticsearch/issues/5559
On Wednesday, March 26, 2014 5:53:15 AM UTC+1, Shinsuke Sugaya wrote:
Hi
I encountered the following problem:
Caused by: java.lang.IndexOutOfBoundsException: index (-2) must not be
negative
at
the repeat filter only applies to terms that actually get stemmed. ie if
you have goes it will be stemmed to go but with the repeat filter it
will also emit goes in addition to go
makes sense?
simon
On Thursday, March 13, 2014 12:38:00 AM UTC+1, Nikita Tovstoles wrote:
Could someone please
We have several fixes relate to this in 0.90.12 and 1.0.1 - can you guys
try and upgrade? would be awesome if you could report back!
simon
On Thursday, January 23, 2014 1:47:31 AM UTC+1, JBartak wrote:
We had an interesting problem today while running a 3 data node
ElasticSearch 90.7
do not have log message from the
beginning of this event.
On Thursday, February 6, 2014 3:18:14 PM UTC-5, simonw wrote:
Thanks so much for the infos here! That is really helpful! Just to keep
you updated, I committed a fix for the infinite loop to lucene 5 min ago (
https
started occurring around 4:16am EST and
continued until 4:45am EST. We noticed issues started with a bulk index
request (not sure if that helps at all).
On Tuesday, February 11, 2014 4:21:12 AM UTC-5, simonw wrote:
Jay, first of all good that this prevents the server from going
Hey folks,
we are trying to get to the bottom of this and I'd want to have some more
infos here. Can you provide us with more insight in what you are doing with
ES. What I'd be interested in is:
* are you using delete by query
* are you running parent child queries
* are you seeing any
during normal runtime
of our application
* We have seen the issue in 0.20.2 (JDK6) and 0.90.8 (JDK7u45)
* No not forcefully flushing
On Thursday, February 6, 2014 2:47:38 PM UTC-5, simonw wrote:
Hey folks,
we are trying to get to the bottom of this and I'd want to have some more
infos here
maybe this is interesting in your situation?
http://www.elasticsearch.org/blog/you-complete-me/
simon
On Thursday, January 23, 2014 4:15:05 PM UTC+1, Nikolay Chankov wrote:
Well, it's really strange, that position is not encountered, since with
many results and especially data with lot of
All these version are backwards compatible. You can safely upgrade a lucene
index from 4.3.1 to 4.5.1 or 4.6.0. You can't downgrade so they are not
forward compatible though!
simon
On Monday, January 13, 2014 1:40:47 AM UTC+1, Jason Wee wrote:
Hi, for es version 0.90.2 using lucene 4.3.1
22 matches
Mail list logo