问题描述
我正在运行MapR集群,并希望使用 Druid 进行一些时间序列分析.MapR对Zookeeper使用非标准端口(端口5181代替了常规端口2181).
I'm running a MapR cluster and want to do some timeseries analysis with Druid. MapR uses a non-standard port for Zookeeper (port 5181 instead of the conventional port 2181).
当我启动Druid协调器服务时,它尝试在传统的Zookeeper端口上连接并失败:
When I start the Druid coordinator service, it attempts to connect on the conventional Zookeeper port and fails:
2015-03-03T17:46:49,614 INFO [main-SendThread(localhost:2181)] org.apache.zookeeper.ClientCnxn - Opening socket connection to server localhost/0:0:0:0:0:0:0:1:2181.
2015-03-03T17:46:49,617 WARN [main-SendThread(localhost:2181)] org.apache.zookeeper.ClientCnxn - Session 0x0 for server null, unexpected error, closing socket connection and attempting reconnect
java.net.ConnectException: Connection refused
德鲁伊文档显示,可以通过设置Zookeeper主机.
属性.它没有指定用于编辑Zookeeper端口的属性. config/_common/common.runtime.properties
中的druid.zk.service.host
The Druid documentation shows that the Zookeeper host can be set via the druid.zk.service.host
property in config/_common/common.runtime.properties
. It does not specify a property to edit the Zookeeper port.
Druid的Zookeeper端口是否可配置?或者,如果将Druid与MapR一起使用,是否有必要运行其他Zookeeper服务?
Is the Zookeeper port for Druid configurable? Or is it necessary to run an additional Zookeeper service if using Druid with MapR?
推荐答案
解决方案是将端口添加到 config/_common/common.runtime.properties
中的主机名:
The solution was to add the port to the hostnames in config/_common/common.runtime.properties
:
druid.zk.service.host=zkNode1:5181,zkNode2:5181,zkNode3:5181
这篇关于配置Druid以在端口5181上连接到Zookeeper的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持!