lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erick Erickson <erickerick...@gmail.com>
Subject Re: How to start solr in solr cloud mode using external zookeeper ?
Date Tue, 03 Mar 2015 23:48:23 GMT
Have you seen this page?:
https://cwiki.apache.org/confluence/display/solr/Solr+Start+Script+Reference

This is really "the new way"....

Best,
Erick

On Tue, Mar 3, 2015 at 7:18 AM, Aman Tandon <amantandon.10@gmail.com> wrote:
> Thanks Shawn, also thanks for sharing info about chroot.
>
> I am trying to implement the solr cloud with solr-5.0.0.
>
> I also checked the documentations https://wiki.apache.org/solr/SolrCloud,
> the method shown there is using start.jar. But after few update start.jar
> (jetty) will not work. So I want to go through the way which will work as
> it is even after upgrade.
>
> So how could i start it from bin directory with all these parameters of
> external zookeeper or any other best way which you can suggest.
>
> With Regards
> Aman Tandon
>
> On Tue, Mar 3, 2015 at 8:09 PM, Shawn Heisey <apache@elyograg.org> wrote:
>
>> On 3/3/2015 4:21 AM, Aman Tandon wrote:
>> > I am new to solr-cloud, i have connected the zookeepers located on 3
>> remote
>> > servers. All the configs are uploaded and linked successfully.
>> >
>> > Now i am stuck to how to start solr in cloud mode using these external
>> > zookeeper which are remotely located.
>> >
>> > Zookeeper is installed at 3 servers and using the 2181 as client port. ON
>> > all three server, solr server along with external zookeeper is present.
>> >
>> > solrcloud1.com (solr + zookeper is present)
>> > solrcloud2.com
>> > solrcloud3.com
>> >
>> > Now i have to start the solr by telling the solr to use the external
>> > zookeeper. So how should I do that.
>>
>> You simply tell Solr about all your zookeeper servers on startup, using
>> the zkHost property.  Here's the format of that property:
>>
>> server1:port,server2:port,server3:port/solr1
>>
>> The /solr1 part (the ZK chroot) is optional, but I recommend it ... it
>> can be just about any text you like, starting with a forward slash.
>> What this does is put all of SolrCloud's information inside a path in
>> zookeeper, sort of like a filesystem.  With no chroot, that information
>> is placed at the "root" of zookeeper.  If you want to use a zookeeper
>> ensemble for multiple applications, you're going to need a chroot.  Even
>> when multiple applications are not required, I recommend it to keep the
>> zookeeper root clean.
>>
>> You can see some examples of zkHost values in the javadoc for SolrJ:
>>
>>
>> http://lucene.apache.org/solr/5_0_0/solr-solrj/org/apache/solr/client/solrj/impl/CloudSolrClient.html#CloudSolrClient%28java.lang.String%29
>>
>> Thanks,
>> Shawn
>>
>>

Mime
View raw message