我正在使用Cassandra作为后端和弹性搜索来部署Titan集群。
以下是不同部分的版本:
Titan DB Cassandra ES Oracle-JDK Ubuntu
0.4.4 1.2.X 0.90.3 1.7 14.04
我已经设置了两个集群,一个Cassandra集群和一个ES集群。
它们都已启动并运行,我已对它们进行了测试。
我的ES群集配置如下:
{
"cluster_name": "Elastic-Cluster",
"nodes": {
"umUbyIWnTvWIXCzlitJZIg": {
"hostname": "titan-cluster01",
"http_address": "inet[/10.7.6.20:9200]",
"name": "Elastic Node01",
"transport_address": "inet[/10.7.6.20:9300]",
"version": "0.90.13"
},
"xY1p4T4-R76gJU8R8DGXSA": {
"hostname": "titan-cluster02",
"http_address": "inet[/10.7.6.22:9200]",
"name": "Elastic Node02",
"transport_address": "inet[/10.7.6.22:9300]",
"version": "0.90.13"
}
},
"ok": true
}
我的文件titan-cassandra-es.properties的内容是:
storage.backend=cassandra
storage.hostname=10.7.6.20,10.7.6.22
storage.port=9160
cache.db-cache = true
cache.db-cache-clean-wait = 20
cache.db-cache-time = 180000
cache.db-cache-size = 0.25
storage.index.search.backend=elasticsearch
storage.index.search.hostname=10.7.6.20,10.7.6.22
storage.index.search.cluster-name="Elastic-Cluster"
storage.index.search.client-only=true
但是,当我尝试使用gremlin创建Titan实例时,我得到了同样的错误。似乎它不识别ES集群节点,因为它警告不存在的节点名称,节点为null。
gremlin>
gremlin> g = TitanFactory.open('/opt/titan/conf/titan-cassandra-es.properties')
WARN org.elasticsearch.client.transport - [Bishop] node null not part of the cluster Cluster ["Elastic-Cluster"], ignoring...
==>titangraph[cassandra:10.7.6.21]
gremlin> WARN org.elasticsearch.client.transport - [Bishop] node null not part of the cluster Cluster ["Elastic-Cluster"], ignoring...
WARN org.elasticsearch.client.transport - [Bishop] node null not part of the cluster Cluster ["Elastic-Cluster"], ignoring...
WARN org.elasticsearch.client.transport - [Bishop] node null not part of the cluster Cluster ["Elastic-Cluster"], ignoring...
WARN org.elasticsearch.client.transport - [Bishop] node null not part of the cluster Cluster ["Elastic-Cluster"], ignoring...
WARN org.elasticsearch.client.transport - [Bishop] node null not part of the cluster Cluster ["Elastic-Cluster"], ignoring...
提前致谢。