[Bug] 执行start-all.sh后无法启动集群问题

[pengjianbo]

Search before asking

• I searched in the issues and found nothing similar.

Version

2.0.10，3C3D集群

Describe the bug and provide the minimal reproduce step

发现start-all.sh没办法启动集群，排查是ssh到目标主机后找不到java环境变量。我已确认配置了环境变量，在/etc/profile,/etc/bashrc,~/.bashrc,~/.profile都加上了环境变量，都执行了source xxx，服务器也尝试重启还是提示找不到java。使用ssh测试结果如下：

最后通过在start-all.sh文件中的bash命令后加上-lc解决了该问题，希望能在官方脚本上加上。

What did you expect to see?

正常启动集群

What did you see instead?

root@n1:/opt/iotdb# ./sbin/cluster/start-all.sh
The system starts the ConfigNode And DataNode of n1
The system starts the ConfigNode And DataNode of n2
The system starts the ConfigNode And DataNode of n3
Cluster startup complete ...

实际上没启动

Anything else?

No response

Are you willing to submit a PR?

• I'm willing to submit a PR!