Mark, Thank you for your valuable feedback. The newbie's views are always appreciated.
Admin Admin UI command is designed for creating a collection based on the configuration you already have. Obviously, it makes that point somewhat less than obvious. To create a new collection with configuration files all in place, you can bootstrap it from a configset. Which is basically what you did when you run "solr -e", except "-e" also populates the files and does other tricks. So, if you go back to the command line and run "solr" you will see a bunch of options. The one you are looking for is "solr create_core" which will tell you all the parameters as well as the available configurations to bootstrap from. I hope this helps. Regards, Alex. ---- Solr Analyzers, Tokenizers, Filters, URPs and even a newsletter: http://www.solr-start.com/ On 26 September 2015 at 21:03, Mark Fenbers <mark.fenb...@noaa.gov> wrote: > Greetings, > > Being a Solr newbie, I've run the examples in the "Solr Quick Start" > document and got a feel for Solr's capabilities. Now I want to move on and > work with my own data and my own Solr server without using the example setup > (i.e., "solr -e" options). This is where the documentation dries up. > Unless I merely haven't found it. > > So, I've been able to launch Solr, and it is running. Then, I used the > browser-based web pages ("admin UI", I think it is called) and created my > new "core". But it says I have to create the directory and subdirs first, > which I did (why can't it do that for me?). Then it complains about a > missing solrconfig.xml, which I copied in from one of a number of places it > is provided in the distribution. Then, it complained about the schema.xml, > and on and on. I think it complained about 6 times before I resolved them > all! So what's the point of the Admin UI creating a new core for you if you > have to do so much manual setup anyway? Why can a simple config.bash script > take care of this administrivia? Moreover, when it was all done complaining, > I was able to /dataimport and index, but the searching doesn't work, so I > have to troubleshoot... (and I've done this with the help of another thread, > which still isn't resolved). Had I had more clear instructions to work > from, I might have it working long ago without bugging this user-group. > > My point is that this process is way too clunky for a mature Apache project > like Solr/Lucene. So clunky, in fact, that I am highly suspicious > (convinced perhaps) that I simply am missing something (or several things), > like a document/tutorial that explains how to move on from the "solr -e" > examples and setup Solr to work in my own environment. Can someone please > point me to the document(s)/tutorial(s) that I am missing? > > Mark
