SonmiHPC集群节点管理
目前集群系统可进行的节点操作如下：
info： 显示集群节点状态
del： 删除计算节点
[root@sonmi ~]# sonmictl node -h
Manage compute nodes
Usage:
sonmictl node [command]
Available Commands:
del Delete a compute node
info Display the status of cluster nodes
Flags:
-h, --help help for node
Use "sonmictl node [command] --help" for more information about a command.[root@sonmi ~]# sonmictl node -h
Manage compute nodes
Usage:
sonmictl node [command]
Available Commands:
del Delete a compute node
info Display the status of cluster nodes
Flags:
-h, --help help for node
Use "sonmictl node [command] --help" for more information about a command.查看集群节点状态
通过 sonmictl node info 可以查看集群中所有节点的状态，如下所示：
各列含义如下：
- NODENAME：节点名称
- CPU ALLOC/TOTAL：每个节点已分配给用户的CPU核心数/总核心数
- MEMORY FREE/TOTAL：每个节点空闲的内存/总内存
- CPU TEMPERATURE：对应节点的CPU温度
- CPU LOAD：对应节点的CPU负载
- CPU STATE：对应节点的状态，跟slurm中节点的状态对应
删除节点
通过执行sonmictl node del <COMPUTE-NODE>命令就可以从集群中移除对应的节点 COMPUTE-NODE。在示例中通过以下命令从集群中移除compute-0-0节点：
sonmictl node del compute-0-0sonmictl node del compute-0-0
重新注册节点
在将节点从集群中移除之后，如果想要将该节点重新注册到集群，可以登陆到已经删除的节点之上，使用 root 用户执行以下命令，重新注册到集群中：
sonmid-registersonmid-register下面例子中，从集群中移除节点compute-0-0之后，通过SSH登陆到compute-0-0节点，然后重新通过sonmid-register注册回集群中。
[root@sonmi ~]# sonmictl node info
+-------------+-----------------+-------------------+-----------------+----------+-------+
| NODENAME | CPU ALLOC/TOTAL | MEMORY FREE/TOTAL | CPU TEMPERATURE | CPU LOAD | STATE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
| sonmi | 0 / 8 | 2.2GB / 3.6GB | 34.0°C|35.0°C | 0.00 | IDLE |
| compute-0-0 | 0 / 8 | 571MB / 1.7GB | 34.0°C|35.0°C | 2.00 | IDLE |
| compute-0-1 | 0 / 8 | 700MB / 1.7GB | 34.0°C|35.0°C | 2.00 | IDLE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
[root@sonmi ~]# sonmictl node del compute-0-0
Delete compute node: compute-0-0
[root@sonmi ~]# sonmictl node info
+-------------+-----------------+-------------------+-----------------+----------+-------+
| NODENAME | CPU ALLOC/TOTAL | MEMORY FREE/TOTAL | CPU TEMPERATURE | CPU LOAD | STATE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
| sonmi | 0 / 8 | 2.2GB / 3.6GB | 33.0°C|35.0°C | 0.00 | IDLE |
| compute-0-1 | 0 / 8 | 700MB / 1.7GB | 33.0°C|35.0°C | 2.00 | IDLE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
[root@sonmi ~]# ssh 10.1.1.2
╔╗╔╗╔╗──╔╗──────────────╔╗────╔═══╗─────────╔╗─╔╦═══╦═══╗
║║║║║║──║║─────────────╔╝╚╗───║╔═╗║─────────║║─║║╔═╗║╔═╗║
║║║║║╠══╣║╔══╦══╦╗╔╦══╗╚╗╔╬══╗║╚══╦══╦═╗╔╗╔╦╣╚═╝║╚═╝║║─╚╝
║╚╝╚╝║║═╣║║╔═╣╔╗║╚╝║║═╣─║║║╔╗║╚══╗║╔╗║╔╗╣╚╝╠╣╔═╗║╔══╣║─╔╗
╚╗╔╗╔╣║═╣╚╣╚═╣╚╝║║║║║═╣─║╚╣╚╝║║╚═╝║╚╝║║║║║║║║║─║║║──║╚═╝║
─╚╝╚╝╚══╩═╩══╩══╩╩╩╩══╝─╚═╩══╝╚═══╩══╩╝╚╩╩╩╩╩╝─╚╩╝──╚═══╝
Activate the web console with: systemctl enable --now cockpit.socket
Last login: Tue Sep 26 14:37:09 2023 from 10.1.1.1
[root@compute-0-0 ~]# sonmid-register
[root@compute-0-0 ~]# exit
logout
Connection to 10.1.1.2 closed.
[root@sonmi ~]# sonmictl node info
+-------------+-----------------+-------------------+-----------------+----------+-------+
| NODENAME | CPU ALLOC/TOTAL | MEMORY FREE/TOTAL | CPU TEMPERATURE | CPU LOAD | STATE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
| sonmi | 0 / 8 | 2.2GB / 3.6GB | 34.0°C|36.0°C | 0.00 | IDLE |
| compute-0-0 | 0 / 8 | 573MB / 1.7GB | 34.0°C|36.0°C | 2.00 | IDLE |
| compute-0-1 | 0 / 8 | 700MB / 1.7GB | 34.0°C|36.0°C | 2.00 | IDLE |
+-------------+-----------------+-------------------+-----------------+----------+-------+[root@sonmi ~]# sonmictl node info
+-------------+-----------------+-------------------+-----------------+----------+-------+
| NODENAME | CPU ALLOC/TOTAL | MEMORY FREE/TOTAL | CPU TEMPERATURE | CPU LOAD | STATE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
| sonmi | 0 / 8 | 2.2GB / 3.6GB | 34.0°C|35.0°C | 0.00 | IDLE |
| compute-0-0 | 0 / 8 | 571MB / 1.7GB | 34.0°C|35.0°C | 2.00 | IDLE |
| compute-0-1 | 0 / 8 | 700MB / 1.7GB | 34.0°C|35.0°C | 2.00 | IDLE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
[root@sonmi ~]# sonmictl node del compute-0-0
Delete compute node: compute-0-0
[root@sonmi ~]# sonmictl node info
+-------------+-----------------+-------------------+-----------------+----------+-------+
| NODENAME | CPU ALLOC/TOTAL | MEMORY FREE/TOTAL | CPU TEMPERATURE | CPU LOAD | STATE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
| sonmi | 0 / 8 | 2.2GB / 3.6GB | 33.0°C|35.0°C | 0.00 | IDLE |
| compute-0-1 | 0 / 8 | 700MB / 1.7GB | 33.0°C|35.0°C | 2.00 | IDLE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
[root@sonmi ~]# ssh 10.1.1.2
╔╗╔╗╔╗──╔╗──────────────╔╗────╔═══╗─────────╔╗─╔╦═══╦═══╗
║║║║║║──║║─────────────╔╝╚╗───║╔═╗║─────────║║─║║╔═╗║╔═╗║
║║║║║╠══╣║╔══╦══╦╗╔╦══╗╚╗╔╬══╗║╚══╦══╦═╗╔╗╔╦╣╚═╝║╚═╝║║─╚╝
║╚╝╚╝║║═╣║║╔═╣╔╗║╚╝║║═╣─║║║╔╗║╚══╗║╔╗║╔╗╣╚╝╠╣╔═╗║╔══╣║─╔╗
╚╗╔╗╔╣║═╣╚╣╚═╣╚╝║║║║║═╣─║╚╣╚╝║║╚═╝║╚╝║║║║║║║║║─║║║──║╚═╝║
─╚╝╚╝╚══╩═╩══╩══╩╩╩╩══╝─╚═╩══╝╚═══╩══╩╝╚╩╩╩╩╩╝─╚╩╝──╚═══╝
Activate the web console with: systemctl enable --now cockpit.socket
Last login: Tue Sep 26 14:37:09 2023 from 10.1.1.1
[root@compute-0-0 ~]# sonmid-register
[root@compute-0-0 ~]# exit
logout
Connection to 10.1.1.2 closed.
[root@sonmi ~]# sonmictl node info
+-------------+-----------------+-------------------+-----------------+----------+-------+
| NODENAME | CPU ALLOC/TOTAL | MEMORY FREE/TOTAL | CPU TEMPERATURE | CPU LOAD | STATE |
+-------------+-----------------+-------------------+-----------------+----------+-------+
| sonmi | 0 / 8 | 2.2GB / 3.6GB | 34.0°C|36.0°C | 0.00 | IDLE |
| compute-0-0 | 0 / 8 | 573MB / 1.7GB | 34.0°C|36.0°C | 2.00 | IDLE |
| compute-0-1 | 0 / 8 | 700MB / 1.7GB | 34.0°C|36.0°C | 2.00 | IDLE |
+-------------+-----------------+-------------------+-----------------+----------+-------+