「Bug」CentOS7 报 OOM，systemd 内存占用率达到 57.8%

top 命令输出（输入 M 进程按内存排序，输入 m 可视化内存占用）：

top - 13:24:35 up 22 days, 23:01,  1 user,  load average: 15.64, 18.52, 12.97
Tasks: 358 total,   5 running, 353 sleeping,   0 stopped,   0 zombie
%Cpu(s): 10.3 us, 21.7 sy,  0.0 ni, 61.5 id,  1.5 wa,  0.0 hi,  5.1 si,  0.0 st
KiB Mem : 98.5/16247608 [||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||  ]
KiB Swap:  0.0/0        [                                                                                                    ]

   PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND                                                                                                                 
     1 root      20   0 9605472   9.0g    348 S   0.0 57.8 798:26.87 systemd                                                                                                                 
 82720 polkitd   20   0 4978420 276312      0 S   2.3  1.7  29:45.70 mysqld                                                                                                                  
 57250 root      20   0 6290796 193804  11516 S  43.9  1.2   4573:23 kubelet                                                                                                                 
 97595 root      20   0  256856 166472     44 R  30.8  1.0 235:49.03 ruby                                                                                                                    
 57071 root      20   0 1796564 110960    564 S  15.1  0.7   2129:00 dockerd                                                                                                                 
 82641 root      20   0   26.1g  85152      0 S   2.0  0.5  37:26.06 dotnet                                                                                                                  
 82103 root      20   0   26.1g  81612    764 S   3.3  0.5  75:23.47 dotnet                                                                                                                  
 81469 root      20   0   26.1g  78808    852 S   2.6  0.5  37:30.94 dotnet                                                                                                                  
 85327 root      20   0   26.1g  78404      0 S   0.3  0.5  15:03.34 dotnet                                                                                                                  
 15729 polkitd   20   0 1853084  78152   1876 S   2.6  0.5   8:14.68 mongod                                                                                                                  
 89447 root      20   0   26.1g  77728      0 S   3.6  0.5  36:48.23 dotnet    

systemctl status 发现状态为 degraded

systemctl --failed 输出：

  UNIT               LOAD   ACTIVE SUB    DESCRIPTION
● chronyd.service    loaded failed failed NTP client/server
● irqbalance.service loaded failed failed irqbalance daemon
● postfix.service    loaded failed failed Postfix Mail Transport Agent
● tuned.service      loaded failed failed Dynamic System Tuning Daemon
● vgauthd.service    loaded failed failed VGAuth Service for open-vm-tools
● vmtoolsd.service   loaded failed failed Service for virtual machines hosted on VMware

LOAD   = Reflects whether the unit definition was properly loaded.
ACTIVE = The high-level unit activation state, i.e. generalization of SUB.
SUB    = The low-level unit activation state, values depend on unit type.

6 loaded units listed. Pass --all to see loaded but inactive units, too.
To show all installed unit files use 'systemctl list-unit-files'.

感觉上很像 systemd 发生了内存泄漏，但是 journalctl 的日志看不明白，ps 1(1 即 systemd pid) 搞一圈也找不到问题点。。。

网上搜一圈有说是 CentOS7 的 bug。给出的解决方案，我总结如下：

1. 重启虚拟机，回收内存。
   1. 实测网上给出的回收内存命令：systemctl daemon-reexec 没有任何效果。。
   2. 只能关闭电源，实测 reboot/poweroff/init 0 命令均无法关机，报错:

Failed to start poweroff.target: Connection timed out
See system logs and 'systemctl status poweroff.target' for details.

Broadcast message from root@xxx-xxx on pts/0 (Tue 2020-11-24 13:45:11 CST):

The system is going down for power-off NOW!

1. 升级 systemd 到新版本，新版本应该修复了这个 Bug: yum update systemd
   • 或者升级整个系统 yum update

升级后应该就没毛病了.