Question

我正在尝试在开发计算机的同一主机上启动3个Zookeeper服务。显然，这不是我将要在生产中执行的操作，而是在测试/开发计算机中探索容错能力和Kafka依赖性。

我已经在我的开发计算机上安装了Kafka 2.5.0，并且能够成功在该主机上设置3个Kafka服务，并且在同一主机上设置了1个Zookeeper服务。使用kafka软件包随附的zookeeper脚本和软件包。

尝试设置3个Zookeeper服务时问题开始了……我做了以下设置3个Zookeeper服务的工作，但是我无法成功启动服务。我有3个配置文件：

config/zookeeper.properties
config/zookeeper1.properties
config/zookeeper2.properties

config / zookeeper.properties的内容为：

dataDir=/tmp/zookeeper
clientPort=2181
maxClientCnxns=0
admin.enableServer=false
initLimit=5
syncLimit=2
server.1=localhost:2888:3888
server.2=localhost:2889:3889
server.3=localhost:2890:3890

在config/zookeeper1.properties上我有clientPort=2182和dataDir=/tmp/zookeeper1

在config/zookeeper2.properties上我有clientPort=2183和dataDir=/tmp/zookeeper2

还创建了文件/ tmp / zookeeper / myid，/ tmp / zookeeper1 / myid，/ tmp / zookeeper2 / myid，并分别输入了ID值1、2、3。

从命令行启动3个动物园管理员时，他们开始正常：

$ sudo bin/zookeeper-server-start.sh config/zookeeper.properties
$ sudo bin/zookeeper-server-start.sh config/zookeeper1.properties
$ sudo bin/zookeeper-server-start.sh config/zookeeper2.properties

我还可以看到领导者和关注者是谁：

$ echo srvr | nc localhost 2181 | grep Mode
Mode: follower
$ echo srvr | nc localhost 2182 | grep Mode
Mode: leader
$ echo srvr | nc localhost 2183 | grep Mode
Mode: follower

但是当我尝试将它们设置为系统服务时，无法正确启动它们...这是我拥有的单位文件：

$ cat /etc/systemd/system/zookeeper.service
[Unit]
Description=zookeeper
After=syslog.target network.target
[Service]
Type=simple
User=kafka
Group=kafka
ExecStart=/opt/kafka/bin/zookeeper-server-start.sh /opt/kafka/config/zookeeper.properties
ExecStop=/opt/kafka/bin/zookeeper-server-stop.sh
[Install]
WantedBy=multi-user.target

$ cat /etc/systemd/system/zookeeper1.service
[Unit]
Description=zookeeper 1
After=syslog.target network.target
[Service]
Type=simple
User=kafka
Group=kafka
ExecStart=/opt/kafka/bin/zookeeper-server-start.sh /opt/kafka/config/zookeeper1.properties
ExecStop=/opt/kafka/bin/zookeeper-server-stop1.sh
[Install]
WantedBy=multi-user.target

$ cat /etc/systemd/system/zookeeper2.service
[Unit]
Description=zookeeper 2
After=syslog.target network.target
[Service]
Type=simple
User=kafka
Group=kafka
ExecStart=/opt/kafka/bin/zookeeper-server-start.sh /opt/kafka/config/zookeeper2.properties
ExecStop=/opt/kafka/bin/zookeeper-server-stop2.sh
[Install]
WantedBy=multi-user.target

尝试以

启动它们之后

$ sudo systemctl daemon-reload
$ sudo systemctl enable  zookeeper
$ sudo systemctl enable  zookeeper1
$ sudo systemctl enable  zookeeper2
$ sudo systemctl start zookeeper
$ sudo systemctl start zookeeper1
$ sudo systemctl start zookeeper2

我看不到它们在运行...

在systemlog上，我看到以下内容：

May 17 03:56:20 melly-dev2 kafka-server-start.sh: [2020-05-17 03:56:20,039] INFO Opening socket connection to server localhost/127.0.0.1:2183. Will not attempt to authenticate using SASL (unknown error) (org.apache.zookeeper.ClientCnxn)
May 17 03:56:20 melly-dev2 kafka-server-start.sh: [2020-05-17 03:56:20,040] INFO Socket error occurred: localhost/127.0.0.1:2183: Connection refused (org.apache.zookeeper.ClientCnxn)

我在sudo journalctl -u zookeeper.service上看到的鬼怪：

[2020-05-17 06:33:33,096] INFO Notification time out: 6400 (org.apache.zookeeper.server.quorum.FastLeaderElection)
[2020-05-17 06:33:39,497] WARN Cannot open channel to 2 at election address localhost/127.0.0.1:3889 (org.apache.zookeeper.server.quorum.QuorumCnxManager)
java.net.ConnectException: Connection refused (Connection refused)
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:350)
at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:206)
at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:188)
at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
at java.net.Socket.connect(Socket.java:589)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:650)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:707)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectAll(QuorumCnxManager.java:735)
at org.apache.zookeeper.server.quorum.FastLeaderElection.lookForLeader(FastLeaderElection.java:910)
at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:1247)
[2020-05-17 06:33:39,497] WARN Cannot open channel to 3 at election address localhost/127.0.0.1:3890 (org.apache.zookeeper.server.quorum.QuorumCnxManager)
java.net.ConnectException: Connection refused (Connection refused)
at java.net.PlainSocketImpl.socketConnect(Native Method)
at java.net.AbstractPlainSocketImpl.doConnect(AbstractPlainSocketImpl.java:350)
at java.net.AbstractPlainSocketImpl.connectToAddress(AbstractPlainSocketImpl.java:206)
at java.net.AbstractPlainSocketImpl.connect(AbstractPlainSocketImpl.java:188)
at java.net.SocksSocketImpl.connect(SocksSocketImpl.java:392)
at java.net.Socket.connect(Socket.java:589)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:650)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectOne(QuorumCnxManager.java:707)
at org.apache.zookeeper.server.quorum.QuorumCnxManager.connectAll(QuorumCnxManager.java:735)
at org.apache.zookeeper.server.quorum.FastLeaderElection.lookForLeader(FastLeaderElection.java:910)
at org.apache.zookeeper.server.quorum.QuorumPeer.run(QuorumPeer.java:1247)
[2020-05-17 06:33:39,497] INFO Notification time out: 12800 (org.apache.zookeeper.server.quorum.FastLeaderElection)

如何设置/查找Zookeeper日志文件，如何使Zookeeper成功作为服务启动？

Answer 1

[2020-05-17 07:02:35,248] ERROR Unable to access datadir, exiting abnormally (org.apache.zookeeper.server.quorum.QuorumPeerMain) 
org.apache.zookeeper.server.persistence.FileTxnSnapLog$DatadirException: Cannot write to data directory /tmp/zookeeper1/version-2

我怀疑用户melly-dev2无权访问/tmp/zookeeper/下的日志。

此外，请确保将dataDir更改为永久位置（即不在/tmp/下），因为一旦关闭计算机，一切都会丢失。

Answer 2

我程序中缺少的步骤是：

sudo chown -R kafka:kafka /tmp/zookeeper
sudo chown -R kafka:kafka /tmp/zookeeper1
sudo chown -R kafka:kafka /tmp/zookeeper2

sudo chmod -R 777 /tmp/zookeeper 
sudo chmod -R 777 /tmp/zookeeper1
sudo chmod -R 777 /tmp/zookeeper2

问题之一是Kafka随附的默认Zookeeper没有显示写入错误的日志。一旦执行了此命令（基于GiorgosMyrianthous的评论）：

journalctl -u zookeeper.service

我可以清楚地看到错误并解决问题。

无法在Kafka开发箱上启动多个Zookeeper服务

2 个答案: