以前文章說了一下JanusGraph使用Hbase+ES的配置(傳送門),今天對Hbase+Solr進行一下配置,已測試經過。node
先是以前遇到過的異常,NoNodeException: KeeperErrorCode = NoNode for /hbase/hbaseid bash
這按照 JanusGraph問題筆記之NoNodeException(hbase)文末解法配置,KO。# 不用改的
storage.backend=hbase
index.search.backend=solr
cache.db-cache = true
cache.db-cache-clean-wait = 20
cache.db-cache-time = 180000
cache.db-cache-size = 0.5
# 酌情改爲本身集羣的
storage.hostname=manager.big.com,master.big.com,worker.big.com
index.search.solr.zookeeper-url=zk1:2181/solr,zk2:2181/solr,zk3:2181/solr
index.search.solr.http-urls=http://manager.big.com:8983/solr
index.search.solr.configset=graph_core #注意,要預先在Solr中建立!
storage.hbase.table=janusgraph_solr
# 踩過坑的
index.search.solr.mode=cloud
storage.hbase.ext.zookeeper.znode.parent=/hbase-unsecure #注意,這是小坎坷說的問題要改的配置
gremlin.graph=org.janusgraph.core.JanusGraphFactory
複製代碼
以上配置,使用以下測試:oop
./bin/gremlin
gremlin> graph = JanusGraphFactory.open('path_to/janusgraph-0.4.0-hadoop2/conf/janusgraph-hbase-solr.properties')
複製代碼
打印出==>standardjanusgraph[hbase:[……]]
之類的就能夠了。post