Two different points:
Checking once a minute should be sufficient. Also, when I did this
instead of pulling a file or doing the 'ping' feature, I did a search
of a non-existent fwildcard field "bogus_s:test". The point being to
make sure that the Lucene part could actually talk to its index.
It should not run out of file descriptors from doing this. The code
does a 'new File(healthcheck file name).exists()' and throws away the
descriptor. This should not be a resource leak for file desciptors.
On Sat, Apr 10, 2010 at 12:36 PM, Blargy <zman...@hotmail.com> wrote:
>
> I have my loadbalancer (HAProxy) configured to check Solr for a healthcheck
> file every 2 seconds.
>
> <admin>
> <defaultQuery>solr</defaultQuery>
> <healthcheck type="file">solr/conf/healthcheck.txt</healthcheck>
> </admin>
>
> However it keeps marking my slaves as down and I am seeing this error:
>
> Apr 10, 2010 12:29:20 PM org.apache.solr.core.SolrCore execute
> INFO: [items] webapp=/solr path=/admin/file params={file=healthcheck.txt}
> status=0 QTime=0
> Apr 10, 2010 12:29:20 PM org.apache.solr.common.SolrException log
> SEVERE: java.io.FileNotFoundException:
> /var/solr/home/items/conf/healthcheck.txt (Too many open files)
> at java.io.FileInputStream.open(Native Method)
> at java.io.FileInputStream.<init>(FileInputStream.java:137)
> at java.io.FileReader.<init>(FileReader.java:72)
> at
> org.apache.solr.common.util.ContentStreamBase$FileStream.getReader(ContentStreamBase.java:118)
> at
> org.apache.solr.request.RawResponseWriter.write(RawResponseWriter.java:83)
>
> Obviously solr is keeping too many files open, but how can I solve this
> problem so I can use this file as my healthcheck?
>
> Thanks
>
>
>
> --
> View this message in context:
> http://n3.nabble.com/Healthcheck-Too-many-open-files-tp710631p710631.html
> Sent from the Solr - User mailing list archive at Nabble.com.
>
--
Lance Norskog
goks...@gmail.com
