Glad you have it worked out. Somehow I missed your last post stating that it is already installed, so my previous comment is OBE. ... View more

Are you also using Ambari 2.4.x? I believe Ambari Infra is shipped wit Ambari 2.4.x, not HDP 2.5. It's sort of like Ambari Metrics is part of Ambari but not directly part of HDP, even though it works with HDP. Assuming you have Ambari 2.4.x, do you don't see "Ambari Infra" when you select Add Service (it does not have the name Solr in it in Ambari)? ... View more

@Sami Ahmad Which version of Ambari are you on? Ambari Infra is only in Ambari 2.4.0 and later. ... View more

A.so, to install the right version of Solr, I'd recommend this page: https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.5.3/bk_security/content/installing_external_solr.html ... View more

@Bilal Arshad Oh, I think I see the problem. Zookeeper already has some data under this configuration. Notice that it says, "re-using existing configuration directory vertex_index". That is probably from your earlier failed attempt when it failed on the jar file (binary files are not allowed). You have three options (maybe more), in this preferred order: Delete the data using Solr's zkcli.sh command --- ($SOLR_DIR/server/scripts/cloud-scripts/zkcli -zkhost localhost:9983 -cmd delete /config/vertext_index). Now you can create the collection -- I have not used this command in a while, but I think that is correct syntax. If this does not work, I'd just go for #2 since you don't have anything data in Solr to worry about. Delete ALL Zookeeper data --- Shutdown Solr. Delete ALL zookeeper data (find the directory Delete the zoo_data directory. Restart solr and create the collection. Leave the bad files in Zookeeper and use a different configName --- Add -n vertext_index2 to your command. This creates a new directory for configs. By default the configName is the same as your collection. ... View more

@Bilal Arshad - It looks like your ..../solr-5.5.1/conf directory does not contain all of the config files. Make sure it contains solrconfig.xml along with schema.xml and the other config files for Solr. If that is not the case, then I suppose there could be a file/directory permission issue. You should have all of these files in your conf directory: ├── currency.xml ├── lang │ └── stopwords_en.txt ├── protwords.txt ├── schema.xml ├── solrconfig.xml ├── stopwords.txt └── synonyms.txt ... View more

@Bilal Arshad - This should work. Here is what I did. Build: export MAVEN_OPTS="-Xmx1536m -XX:MaxPermSize=512m" && mvn clean install I did not care about packaging above (mvn clean package -Pdist,external-hbase-solr ) but it should still work the same. Now find the config files: $ find . -name schema.xml ./distro/src/conf/solr/schema.xml ./distro/target/conf/solr/schema.xml $ ls ./distro/src/conf/solr/ currency.xml lang protwords.txt schema.xml solrconfig.xml stopwords.txt synonyms.txt Now create the collection: $ solr create -c testAtlas -d /home/me/atlas/distro/target/conf/solr -shards 1 -replicationFactor 1 -p 8984 Connecting to ZooKeeper at localhost:9984 Uploading /home/me/atlas/distro/target/conf/solr for config testAtlas to ZooKeeper at localhost:9984 Creating new collection 'testAtlas' using command: http://HOSTNAME:8984/solr/admin/collections?action=CREATE&name=testAtlas&numShards=1&replicationFactor=1&maxShardsPerNode=1&collection.configName=testAtlas { "responseHeader":{ "status":0, "QTime":2691}, "success":{"":{ "responseHeader":{ "status":0, "QTime":2505}, "core":"testAtlas_shard1_replica1"}}} ... View more

I'm glad it worked! You might want to post something on HCC about your Hive issue. However, the way to figure out what is wrong is to look carefully at the logs. Of course, you can see Some logs in Ambari from the restart, but also, go look in Ambari to figure out which sub-component is not running (or all). Then go look on that host at /var/log for the component's logs. For example, /var/log/hive/ or /var/log/hive-catalog. Do an ls -ltr on the directory and find at all of the most recent files (both ending in .log and .out). Look carefully near the bottom of those files for errors and clues. Good luck. ... View more