Hmmm. I'm assuming that "nothing in the logs" is node/logs/solr.log, and that you're not finding errors/exceptipons. Just sanity checking here.
My guess: you're picking up the default SOLR_HOME which is in your new installation directory and all your replicas are under the old install directory. There should be some kind of message in the log files indicating that Solr is at least trying to load replicas, something similar to: Using system property solr.solr.home: /Users/Erick/apache/solrVersions/playspace/solr/example/cloud/node1/solr and/or: CorePropertiesLocator Found 3 core definitions underneath /Users/Erick/apache/solrVersions/playspace/solr/example/cloud/node1/solr A bit of background: When Solr starts up, it recursively descends from SOLR_HOME and whenever it finds a "core.properties" file it says "Aha, this must be a core, I'll try to load it". So if SOLR_HOME is doesn't point to an ancestor of your existing replicas, Solr won't find any replicas and everything will stay down. _If_ SOLR_HOME is defined in solr.in.sh, this should just be picked up. Best, Erick On Thu, Feb 14, 2019 at 7:43 PM Zheng Lin Edwin Yeo <edwinye...@gmail.com> wrote: > > Hi, > > Which version of zookeeper are you using? > > Also, if you tried to query the index, did you get any error message? > > Regards, > Edwin > > > On Fri, 15 Feb 2019 at 02:34, Jeff Courtade <courtadej...@gmail.com> wrote: > > > Hi, > > > > I am working n doing a simple point upgrade from solr 7.6 to 7.7 cloud. > > > > 6 servers > > 3 zookeepers > > one simple test collection using the prepackages _default config. > > > > i stop all solr servers leaving the zookeepers up. > > > > change out the binaries and put the solr.in.sh file back in place with > > memory and directory stuff. > > > > The index directory does not move the files dont change > > > > i start up the new binaries and it starts with no errors in the logs but > > all of the indexes are "down" > > > > I have no clue here. nothing in the logs > >