Would it be acceptable to change a central "slave config"? Because it's possible to have the replication process distribute solrconfig.xml files to the slaves that are different from the master.
That way, your master has it's own solrconfig.xml, and a solrconfig_slave.xml in the conf directory. At replication, the solrconfig_slave.xml is what's sent to the slave as solrconfig.xml, presumably this file has the whole master thing removed. See: http://wiki.apache.org/solr/SolrReplication, the "replicating solrconfig.xml" section.... which is another way of saying that I have no clue how to do what you asked, but this solution seems like it might do. Best Erick On Thu, Oct 27, 2011 at 5:34 PM, Michael Dodd <md...@vocus.com> wrote: > I've been given the project of setting up a CentOS-based solr replication > slave for a project here at work. I think it's configured correctly, and > replication seems to be happening correctly. > > I've got some CentOS experience, but I'm having to get up to speed on Solr in > a short period of time. The guy who was working on this piece of the project > is no longer available and I'm not sure he knew what he was doing anyway. > > The main problem I'm having is that the project lead wants to make sure the > slaves have master disabled for all cores without changing the solrconfig.xml > for every core. > > This link seems to apply to what I'm trying to do: > > http://wiki.apache.org/solr/SolrReplication#enable.2BAC8-disable_master.2BAC8-slave_in_a_node > > My first question is this: > In the tomcat/solr implementation I'm using, I can't tell where/how to pass > system parameters (-Denable.master=false) to solr. I've found how to pass > this sort of thing into tomcat, but it doesn't seem like this is the same > thing. > > Next question: > The link references setting these properties in a solrcore.properties file. > I've created the file and landed it next to the applicable solrconfig.xml but > it doesn't seem to apply it's settings. > > Both the master and the slave node are on solr v3.3 > The master is running on windows server 2008 r2, and was set up well before > my involvement > The slave is running CentOS 6.0 > > Thanks for reading. I'm happy to provide more info as needed. >