Thnx for your input Lewis! Without Kerberos, I may just index to a local core on which to do some heavy processing and then figure out how to get the data securely over to the webserver core, this way I can optimizar the remote core just for webserving .
Greetz from Germány ;) Ralf Kotowski El 20/7/2016 15:28, "Lewis John Mcgibbney" <[email protected]> escribió: Hi RRK, Check out the current configuration options for securing indexing in various plugins. This is currently plugin specific. The Solr configuration can be seen at https://github.com/apache/nutch/blob/master/conf/nutch-default.xml#L1800-L1808 Basically change this property to true, then pass -D solr.auth.username and -D solr.auth.password options to enable basic authentication. Of course this must also be configured on the Solr side. Right now, Kerberos authentication is not available however this might be something which you might want to open a ticket for. Thanks Lewis On Wed, Jul 20, 2016 at 6:11 AM, <[email protected]> wrote: > From: BlackIce <[email protected]> > To: [email protected] > Cc: > Date: Wed, 20 Jul 2016 15:11:22 +0200 > Subject: Indexing to remote Solr server > Hi, > > Up till now we have been running nutch and solr on the same Machine. But > now we have a scenario were we want to have running separate nutch > instances on separaté machines and index to solr over the Internet. > > Since the indexing to Solr will be done over the public Internet it > presents us with new challenges , mostly with security. > > We have been looking into different scenarios, the one we are inclined to > do is securing Solr trough Kerberos! Does nutch Support this? Is nutch > "agnostic" when it comes to Solr authentication? Or does it want a specific > security setup? > > What are others doing? Having nutch index to a local solr instance and then > replicate the core???? > > One of the reasons to do this is to have the "public" server just Busy with > web traffic .... so any input would be greatly apreciated! > >
