Original link: Zookeeper of the actual combat cluster model
The previous article introduces the installation and application of Zookeeper's stand-alone cluster mode, and today it is said that how to install the configuration in the true multi-machine cluster mode.
Here, unzip the zookeeper installation package to the/OPT directory, and on the three machines directory path is the same, the assumption is/opt/zookeeper.
1. First edit the conf/zoo.cfg files in each zookeeper directory, at which point the contents of these three zookeeper profiles are the same, as follows
$ cat/opt/zookeeper/conf/zoo.cfg
ticktime=2000
datadir=/opt/zookeeper/data
clientport=2181
initlimit=10
synclimit=5
server.1=fanbinx1:2888:3888
server.2=fanbinx2:2888:3888
server.3= fanbinx3:2888:3888
There are a few points to be aware of
* DataDir: The datadir of the three zookeeper instances are the same and are assigned to the data directory in the Zookeeper instance directory.
* ClientPort: Three zookeeper instance ports are the same.
* Server.: Three zookeeper instances have the same server list.
2. Create the data directory and instance ID files
Run the following command on each of the three machines to produce a different instance ID file
First machine
Mkdir/opt/zookeeper/data
echo 1 >/opt/zookeeper/data/myid
Second Machine
Mkdir/opt/zookeeper/data
echo 2 >/opt/zookeeper/data/myid
Third machine
Mkdir/opt/zookeeper/data
echo 3 >/opt/zookeeper/data/myid
Note that you need to create a myID file in each zookeeper DataDir directory, which records the instance IDs of each zookeeper.
3. Start Zookeeper Service
Enter each zookeeper's Bin directory, and then run "./zkserver.sh start" to start a zookeeper service.
4. Client Connection
Casually into a zookeeper bin directory, and then run
./zkcli.sh-server
To connect the zookeeper service separately.
Create a Znode node on one of the client
Create/mykey myvalue
Then view the newly created Zonde node on the other client
Get/mykey
5. View Zookeeper Status
After starting zookeeper, because zookeeper will have a set of leader election algorithm, so if you want to know that zookeeper is leader can run in each zookeeper Bin directory./zkserver.sh Status command to view the.
If it's leader,
$./zkserver.sh status
JMX enabled by default
Using config:/opt/zookeeper/bin/. /conf/zoo.cfg
Mode:leader
If it's not leader,
$./zkserver.sh status
JMX enabled by default
Using config:/opt/zookeeper/bin/. /conf/zoo.cfg
Mode:follower
At this point, you can stop that phase of leader, and then look at the other two zookeeper instances, at which point the remaining two zookeeper instances will elect a leader.