[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <5b5da2d3-37b3-60ee-5a94-149094cbe9dc@huawei.com>
Date: Fri, 26 Apr 2024 12:55:04 +0800
From: Lu Jialin <lujialin4@...wei.com>
To: Tejun Heo <tj@...nel.org>, Zefan Li <lizefan.x@...edance.com>, Johannes
Weiner <hannes@...xchg.org>
CC: <cgroups@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: [QUESTION]cgroup mount return -EBUSY
Hello
I encountered a problem when I try to unmount the subsystem right after
all it's subcgroups are removed, the cgroup_root would remain. Mounting
this subsystem to another cgroup_root would return -EBUSY.
The problem could be reproduced with the following script.
test.sh:
mkdir /tmp/test1
mount -t cgroup -o pids pids /tmp/test1
mkdir /tmp/test1/test
rmdir /tmp/test1/test
umount /tmp/test1
mkdir /tmp/test
mount -t cgroup -o pids,cpu none /tmp/test
test.sh should return this.
mount: mounting none on /tmp/test failed: Device or resource busy.
It seems that when unmounting /tmp/test1, the original cgroup_root for
this PID is not released.
/test # cat /proc/cgroups
#subsys_name hierarchy num_cgroups enabled
cpuset 0 1 1
cpu 0 1 1
cpuacct 0 1 1
blkio 0 1 1
devices 0 1 1
freezer 0 1 1
net_cls 0 1 1
perf_event 0 1 1
net_prio 0 1 1
hugetlb 0 1 1
pids 1 1 1
rdma 0 1 1
misc 0 1 1
debug 0 1 1
Powered by blists - more mailing lists