使用 systemd 部署 OceanBase 数据库安装后,启动失败

【 使用环境 】测试环境
【 OB】
【 使用版本 】4.3.3
【问题描述】systemctl start oceanbase 启动数据库失败
10月 25 17:36:07 bogon bash[49363]: oceanbase service started at 2024-10-25 17:36:07
10月 25 17:36:10 bogon bash[49363]: change etc owner to root
10月 25 17:36:10 bogon bash[49363]: daemon process with PID 2968 is running.
10月 25 17:36:10 bogon bash[49363]: The agent service is exist
10月 25 17:36:10 bogon bash[49363]: The observer has been installed before
10月 25 17:36:10 bogon bash[49363]: observer PID file not found.
10月 25 17:36:10 bogon systemd[1]: oceanbase.service: Control process exited, code=exited, status=1/FAILURE
░░ Subject: Unit process exited
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ An ExecStart= process belonging to unit oceanbase.service has exited.
░░
░░ The process’ exit code is ‘exited’ and its exit status is 1.
10月 25 17:36:10 bogon systemd[1]: oceanbase.service: Failed with result ‘exit-code’.
░░ Subject: Unit failed
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ The unit oceanbase.service has entered the ‘failed’ state with result ‘exit-code’.
10月 25 17:36:10 bogon systemd[1]: oceanbase.service: Unit process 2968 (obshell) remains running after unit stopped.
10月 25 17:36:10 bogon systemd[1]: oceanbase.service: Unit process 2989 (obshell) remains running after unit stopped.
10月 25 17:36:10 bogon systemd[1]: Failed to start oceanbase.
░░ Subject: oceanbase.service 单元已失败
░░ Defined-By: systemd
░░ Support: https://lists.freedesktop.org/mailman/listinfo/systemd-devel
░░
░░ oceanbase.service 单元已失败。
░░
░░ 结果为“failed”

state: FAILED
operator: RUN
start_time: 2024-10-24T17:24:58.194944174+08:00
end_time: 2024-10-24T17:25:16.31021836+08:00
subtasks:
id: 23232252033028866
task_id: 6
name: Start observer
state: FAILED
operator: RUN
start_time: 2024-10-24T17:24:58.211394878+08:00
end_time: 2024-10-24T17:25:15.70046807+08:00
execute_times: 1
execute_agent:
ip: 192.168.64.129
port: 2886
task_logs:
- start observer
- check if first start
- generate start cmd
- ‘start cmd: export LD_LIBRARY_PATH=’’/home/admin/oceanbase/lib’’; /home/admin/oceanbase/bin/observer -I
192.168.64.129 -c 1 -d /home/admin/oceanbase/store -p 2881 -P 2882 -n ob -z
zone1 -o ‘‘log_disk_size=13G,datafile_next=2G,enable_syslog_wf=false,cpu_count=16,datafile_size=2G,enable_syslog_recycle=true,memory_limit=6G,__min_full_resource_pool_memory=1073741824,datafile_maxsize=20G,max_syslog_file_count=4,system_memory=1G’’’
- observer health check, retry [1/300]
- observer health check, retry [2/300]
- observer health check, retry [3/300]
- observer health check, retry [4/300]
- observer health check, retry [5/300]
- observer health check, retry [6/300]
- observer health check, retry [7/300]
- observer health check, retry [8/300]
- ‘ERROR: observer health check failed: check observer process exist: false, ,’

【备注】基于 LLM 和开源文档 RAG 的论坛小助手已开放测试,在发帖时输入 [@论坛小助手] 即可召唤小助手,欢迎试用!

推荐使用 obdiag 工具进行日志采集,并且将采集到的日志提供给我们以便确认问题所在。
日志采集场景: 集群基础信息
日志采集命令: obdiag gather scene run --scene=observer.base

另外,为了确定具体的问题,在此需要您提供一些信息:

  1. 您是否确认 PID 文件路径配置正确?
  2. 您能否提供完整的错误日志内容?
  3. 您是否有尝试手动创建 PID 文件或检查相关权限设置?
  4. 您是否在其他节点上也遇到了相同的问题?
  5. 您是否可以提供 systemd 服务文件的内容?

附上敏捷诊断工具 obdiag 使用帮助链接

请问该问题解决了么, health check failed 需要提供一下observer日志方便查看一下启动报错原因