Some good food for thought here…. I hadn’t really dug quite so much into the specifics of the flow.
Do we think that having a generic “create” is making things simpler, or being more verbose and having create_core and create_collection suffice? I guess I am wondering if there are good arguments for keeping create? Mostly because when you run “bin/solr create -h”, we give the help output for the “create_collection” command… One thought…. What if, when you run bin/solr create -h, we actually detect if you are in solrcloud mode and then delegate to either create_collection or create_core to decide the help output? > On Jul 9, 2023, at 4:28 PM, Shawn Heisey <apa...@elyograg.org> wrote: > > On 7/8/23 15:03, Ishan Chattopadhyaya wrote: >> I'd rather we remove all three and encourage users to issue API commands >> via curl. >> I'm very much in favour of the scripts being used for only essential tasks, >> but not for things where the API can be used. With the APIs being the >> primary means to achieve tasks, users develop more familiarity when >> starting out. > > If configs are already uploaded to ZK, creating collections works with just > the API in cloud mode. The bin/solr option will do the config upload for you > before creating the collection. > > Out of the box, standalone mode can't create a core completely with the API. > If standalone mode is configured with configsets, then it can. If we do keep > the bin/solr create option for standalone mode, I would like for it to be > able to detect what user Solr is running as and fail if the create command is > running under a different user. > > There has been discussion about making future versions of Solr default to > cloud mode. The API limitation for creating cores in standalone mode is one > thing in favor of that change. > > I have been trying to think of ways we can automate the deployment of a > SolrCloud install, especially a high availability cluster that has three or > more nodes. We have a solr service installer, I think we should also provide > a ZK service installer. I don't think it would be a good plan to have that > whole idea use the embedded ZK, because with the embedded ZK, stopping or > restarting Solr also takes down one of the ZK nodes. > > Even though I personally don't think of Windows as a viable platform for > running Solr, others do, so I think there should be service installers for > Windows. Picking a service wrapper is something we could bikeshed forever > on. Another thing to decide is whether to support 32-bit Windows or just > 64-bit. I think it should just be 64-bit. > > Thanks, > Shawn > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@solr.apache.org > For additional commands, e-mail: dev-h...@solr.apache.org > _______________________ Eric Pugh | Founder & CEO | OpenSource Connections, LLC | 434.466.1467 | http://www.opensourceconnections.com <http://www.opensourceconnections.com/> | My Free/Busy <http://tinyurl.com/eric-cal> Co-Author: Apache Solr Enterprise Search Server, 3rd Ed <https://www.packtpub.com/big-data-and-business-intelligence/apache-solr-enterprise-search-server-third-edition-raw> This e-mail and all contents, including attachments, is considered to be Company Confidential unless explicitly stated otherwise, regardless of whether attachments are marked as such.
