Hi,
I am trying to disable filter cache for some filter queries as they contain
unique ids and cause cache evictions. By adding {!cache=false} the fq is no
longer stored in filter cache, however I have similar conditions in
facet.query and using facet.query={!cache=false}(color:red AND id:XXX)
Hi Solr Users,
I would like for my test disable the cache fonction, so I modified all
information concerning cache in solrconfig.xml
but after restarting my Tomcat cache is always here.
Do you think I forgot something?
Requests are done with QTime=1 or QTime=0
and with this rapidity my
I think you could disable Solr caches by setting their size to 0 (deleting
them won't work, as for example, the FieldValueCache will take default
values, not sure about the other ones). I don't think you'll be able to
disable Lucene's Field Cache.
What's the test that you want to run? Why do you
Hi Bruno,
don't forget the OS disk cache.
On linux you can clear it with this tiny script :
#!/bin/bash
sync echo 3 /proc/sys/vm/drop_caches
Ludovic.
-
Jouve
France.
--
View this message in context:
http://lucene.472066.n3.nabble.com/Disable-cache-tp3995575p3995589.html
Sent from
I'm currently trying to patch aginst trunk, using SOLR-236.patch from
18/12/2009 but getting the following error...
[...@intelcompute solr]$ patch -p0 SOLR-236.patch
patching file src/test/test-files/solr/conf/solrconfig-fieldcollapse.xml
patching file
Sorry, that was when trying to patch the 1.4 branch
attempting to patch the trunk gives...
patching file src/test/test-files/fieldcollapse/testResponse.xml
patching file src/test/org/apache/solr/BaseDistributedSearchTestCase.java
Hunk #2 FAILED at 502.
1 out of 2 hunks FAILED -- saving rejects
Latest SOLR-236.patch is for the trunk, if have updated the latest
patch so it should patch now without conflicts. If I remember
correctly the latest field-collapse-5.patch should work for 1.4, but
it doesn't for the trunk.
2009/12/23 r...@intelcompute.com:
Sorry, that was when trying to patch
Thanks, that latest update to the patch works fine now.
On Wed 23/12/09 13:13 , Martijn v Groningen martijn.is.h...@gmail.com wrote:
Latest SOLR-236.patch is for the trunk, if have updated the latest
patch so it should patch now without conflicts. If I remember
correctly the latest
Still seeing the same error when trying to comment out the fieldCollapsing
block, or even just the fieldCollapseCache block inside it to disable the cache.
fresh trunk and latest patch.
message null java.lang.NullPointerException at
How have you configured field collapsing?
I have field collapsing configured without caching like this:
searchComponent name=collapse
class=org.apache.solr.handler.component.CollapseComponent /
and with caching like this:
searchComponent name=collapse
Here's the sorlconfig, with a few fields removed from the request (which
shouldn't make any difference)
http://www.intelcompute.com/solrconfig.xml
perhaps an old solrconfig to start with? tho it came no later than a 1.3
standard config file.
On Wed 23/12/09 21:33 , Martijn v Groningen
my
solconfig can be seen at http://www.intelcompute.com/solrconfig.xml
[1]
On Tue 22/12/09 10:51 , r...@intelcompute.com wrote:Is
it possible to disable the field collapsing cache? I'm trying to
perform some speed tests, and have managed to comment out the filter,
queryResult, and document caches
Is
it possible to disable the field collapsing cache? I'm trying to
perform some speed tests, and have managed to comment out the filter,
queryResult, and document caches successfully.
on 1.5
...
...
collapse
facet
tvComponent
...
If you take out the fieldCollapsing/fieldCollapseCache element in your
config the fieldcollapse component will not use a cache.
From http://wiki.apache.org/solr/FieldCollapsing#line-63
If the field collapse cache is not configured then the field
collapse logic will not be cached.
That's what I assumed, but I'm getting the following error with it
commented out
MESSAGE null java.lang.NullPointerException at
org.apache.solr.search.fieldcollapse.AbstractDocumentCollapser.createDocumentCollapseResult(AbstractDocumentCollapser.java:276)
at
Which elements did you comment out? It could be the case that you need
to get rid of the entire fieldCollapsing element, not just the
fieldCollapsingCache element.
(Disclaimer: I've not used field collapsing in anger before :)
Toby.
On 22 Dec 2009, at 11:09, r...@intelcompute.com wrote:
I've tried both, the whole fieldCollapsing tag, and just the
fieldCollapseCache tag inside it.
both cause error.
I guess I can just set size, initialSize, and autowarmCount to 0 ??
On Tue 22/12/09 11:17 , Toby Cole wrote:Which elements did you
comment out? It could be the case
Hi Rob,
What patch are you actually using from SOLR-236?
Martijn
2009/12/22 r...@intelcompute.com:
I've tried both, the whole fieldCollapsing tag, and just the
fieldCollapseCache tag inside it.
both cause error.
I guess I can just set size, initialSize, and autowarmCount to 0
Yup, that's the one, with a copy of trunk from last week.
On Tue 22/12/09 12:28 , Martijn v Groningen martijn.is.h...@gmail.com wrote:
Hi Rob,
What patch are you actually using from SOLR-236?
Martijn
2009/12/22 :
I've tried both, the whole fieldCollapsing tag, and just the
actually i got field-collapse-5.patch, what's the diff?
On Tue 22/12/09 12:28 , Martijn v Groningen martijn.is.h...@gmail.com wrote:
Hi Rob,
What patch are you actually using from SOLR-236?
Martijn
2009/12/22 :
I've tried both, the whole fieldCollapsing tag, and just the
On Tue 22/12/09 12:28 , Martijn v Groningen martijn.is.h...@gmail.com wrote:
Hi Rob,
What patch are you actually using from SOLR-236?
Martijn
2009/12/22 :
I've tried both, the whole fieldCollapsing tag, and just the
fieldCollapseCache tag inside it.
both cause error.
In the latest patch some changes where made on the configuration side,
but if you add the CollapseComponent to the conf no field collapse
cache should be enabled. If not let me know.
Martijn
2009/12/22 r...@intelcompute.com:
On Tue 22/12/09 12:28 , Martijn v Groningen
To avoid this possible bug, you could change the cache to only have a
few entries.
On Tue, Dec 22, 2009 at 6:34 AM, Martijn v Groningen
martijn.is.h...@gmail.com wrote:
In the latest patch some changes where made on the configuration side,
but if you add the CollapseComponent to the conf no
23 matches
Mail list logo
