Taiwan Hadoop Forum

不好意思，有些問題想請教一下
(參考Cloudera)
用法:
add_peer '<n>', "slave.zookeeper.quorum:zookeeper.clientport.:zookeeper.znode.parent"
範例:
hbase(main):001:0> add_peer '1', "zk.server.com:2181:/hbase"

Q1:
我一直有個疑問→zk.server.com，指的是/etc/hosts 內的名稱?
因為發現參考網路有些人的hbase-site.xml 設定中
<property>
<name>hbase.zookeeper.quorum</name>
<value>zk1.com,zk2.com.zk3.com</value>
</property>

但是在hbase shell 下執行add_peer 卻是
hbase(main):006:0>add_peer '1',"slave-zk1:2181:/hbase"
為何是寫slave-zk1，並非是寫zk1.com:2181:/hbase
(不是因該與/ets/hosts 的內容一樣??)

Q2:
而且對add_peer 有很大的疑問，在Cloudera 作法中雖然是說明"one cluster to another"
但是用法卻像是在同一座叢集的節點去執行add_peer而已(slave.zookeeper.quorum:zookeeper.clientport.:zookeeper.znode.parent)
並沒有真的從ClusterA的節點複製到ClusterB的節點，而且強調(slave.zookeeper.quorum:zookeeper.clientport.:zookeeper.znode.parent
感覺只是單純在ClusterA中的master(10.1.1.1)複製到zk1(10.1.1.2)，還是在同一網域
並非從ClusterA中的master(10.1.1.1)複製到ClusterB的zk1(10.1.2.1)
並沒有真的做到跨出機房備援到另一個機房的Hbase叢集，是否是我說的這個意思??

好像發現了一個問題，在Cloudera 的注意事項上發現
1. You must manage Zookeeper yourself. It must not be managed by HBase, and must be available throughout the deployment.
2. Each host in both clusters must be able to reach every other host, including those in the Zookeeper cluster.
Both clusters should have the same HBase and Hadoop major revision. For example, having 0.90.1 on the master and 0.90.0 on the slave is supported, but 0.90.1 on one cluster and 0.89.20100725 on the other is not.
3. Every table that contains families that are scoped for replication must exist on each cluster and have exactly the same name.
4. HBase version 0.92 or greater is required for multiple slaves, master-master, or cyclic replication. This version ships with CDH4.0.0.

第一點說明了，似乎是不能透過HBase 本身自帶起的Zookeeper，必須自己安裝zookeeper...................

至於第二點的說話不知道是不是我所理解的意思
兩座叢集的節點，之間都必須是可以連線的，所式是指clusterA 的所有節點可以ping 到clusterB 的所有節點
還是指clusterA 的所有節點都可以ssh(免鑰)到clusterB 所有節點??