docker device or resource busy

当我们在使用低版本的docker时，删除container出现device or resource busy

1
2

[root@k8s1 ~]# docker rm -f etcd1
Error response from daemon: Driver overlay2 failed to remove root filesystem ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58: remove /var/lib/docker/overlay2/1a90250c2b5b00295a8efce60142aebcf2f30cfa1d5ec7c029e250c5a3a90951/merged: device or resource busy

在这里我们应该如何处理此类问题？

查找mountinfo

[root@k8s1 ~]# grep ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58 /proc/*/mountinfo
/proc/11987/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/12001/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/12002/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/12006/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/12008/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/12009/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/12107/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k
/proc/14625/mountinfo:596 340 0:57 / /var/lib/docker/containers/ca74eaa1b342c64a7f6c558299bedf74a5b7016c194e885bf290281161605d58/shm rw,nosuid,nodev,noexec,relatime shared:111 - tmpfs shm rw,size=65536k

查找进程占用

1
2
3

[root@k8s1 ~]# ps aux|grep 11987
root     11536  0.0  0.0 107408  1696 pts/3    S+   16:48   0:00 grep --color=auto 11987
root     11987  0.0  0.0  18576  7968 ?        Ss   16:06   0:00 /usr/sbin/httpd -DFOREGROUND

停掉对用进程

1	[root@k8s1 ~]# systemctl stop httpd

删除container

1 2	[root@k8s1 ~]# docker rm -f etcd1 etcd1

根因

查看httpd服务的unit

[root@k8s1 ~]# cat /usr/lib/systemd/system/httpd.service 
[Unit]
Description=The Apache HTTP Server
After=network.target remote-fs.target nss-lookup.target
Documentation=man:httpd(8)
Documentation=man:apachectl(8)

[Service]
Type=notify
EnvironmentFile=/etc/sysconfig/httpd
ExecStart=/usr/sbin/httpd $OPTIONS -DFOREGROUND
ExecReload=/usr/sbin/httpd $OPTIONS -k graceful
ExecStop=/bin/kill -WINCH ${MAINPID}
# We want systemd to give httpd some time to finish gracefully, but still want
# it to kill httpd after TimeoutStopSec if something went wrong during the
# graceful stop. Normally, Systemd sends SIGTERM signal right after the
# ExecStop, which would kill httpd. We are sending useless SIGCONT here to give
# httpd time to finish.
KillSignal=SIGCONT
PrivateTmp=true

[Install]
WantedBy=multi-user.target