James's questions

我在 AWS/GCP 上有两个 32 vCPU 实例。我正在尝试设置 cpu 屏蔽，以便系统使用 CPU 0、1，而 cpus 2-31 被屏蔽并且仅由用户空间线程显式使用。

系统信息：

Distributor ID: Ubuntu
Description:    Ubuntu 22.04.1 LTS
Release:    22.04
Codename:   jammy

$ cat /proc/filesystems | grep cpuset
nodev   cpuset

但是，当我尝试运行时cset shield，出现与安装有关的错误：

mount: /cpusets: none already mounted on /run/credentials/systemd-sysusers.service.
cset: **> mount of cpuset filesystem failed, do you have permission?

我深入研究了 cset 代码，失败的调用似乎是一个

$ sudo mount -t cpuset cpuset /cpusets
mount: /cpusets: cpuset already mounted or mount point busy.

/cpusets是一个新创建的文件夹，并且$ cat /proc/mounts | grep cpuset是空的——所以 cpuset 似乎没有安装在其他地方。

也许相关：

$ cat /proc/mounts | grep cgroup
cgroup2 /sys/fs/cgroup cgroup2 rw,nosuid,nodev,noexec,relatime,nsdelegate,memory_recursiveprot 0 0

我的猜测是 AWS/GCP 将 cpuset 用于管理程序或类似的东西。是否可以在 AWS/GCP 上隔离 cpus？我该怎么办？

我最近在我的软件 RAID 阵列中添加了第 5 个驱动器——mdadm 接受了它：

$ lsblk
NAME           MAJ:MIN RM   SIZE RO TYPE  MOUNTPOINT
nvme0n1        259:0    0 894.3G  0 disk
├─nvme0n1p1    259:4    0   512M  0 part
│ └─md0          9:0    0   511M  0 raid1 /boot
└─nvme0n1p2    259:5    0 893.8G  0 part
  └─md1          9:1    0   3.5T  0 raid5
    ├─vg0-swap 253:0    0    32G  0 lvm   [SWAP]
    ├─vg0-tmp  253:1    0    50G  0 lvm   /tmp
    └─vg0-root 253:2    0   2.6T  0 lvm   /
nvme3n1        259:1    0 894.3G  0 disk
├─nvme3n1p1    259:6    0   512M  0 part
│ └─md0          9:0    0   511M  0 raid1 /boot
└─nvme3n1p2    259:7    0 893.8G  0 part
  └─md1          9:1    0   3.5T  0 raid5
    ├─vg0-swap 253:0    0    32G  0 lvm   [SWAP]
    ├─vg0-tmp  253:1    0    50G  0 lvm   /tmp
    └─vg0-root 253:2    0   2.6T  0 lvm   /
nvme2n1        259:2    0 894.3G  0 disk
├─nvme2n1p1    259:8    0   512M  0 part
│ └─md0          9:0    0   511M  0 raid1 /boot
└─nvme2n1p2    259:9    0 893.8G  0 part
  └─md1          9:1    0   3.5T  0 raid5
    ├─vg0-swap 253:0    0    32G  0 lvm   [SWAP]
    ├─vg0-tmp  253:1    0    50G  0 lvm   /tmp
    └─vg0-root 253:2    0   2.6T  0 lvm   /
nvme1n1        259:3    0 894.3G  0 disk
├─nvme1n1p1    259:10   0   512M  0 part
│ └─md0          9:0    0   511M  0 raid1 /boot
└─nvme1n1p2    259:11   0 893.8G  0 part
  └─md1          9:1    0   3.5T  0 raid5
    ├─vg0-swap 253:0    0    32G  0 lvm   [SWAP]
    ├─vg0-tmp  253:1    0    50G  0 lvm   /tmp
    └─vg0-root 253:2    0   2.6T  0 lvm   /
nvme4n1        259:12   0 894.3G  0 disk
├─nvme4n1p1    259:15   0   512M  0 part
│ └─md0          9:0    0   511M  0 raid1 /boot
└─nvme4n1p2    259:16   0 893.8G  0 part
  └─md1          9:1    0   3.5T  0 raid5
    ├─vg0-swap 253:0    0    32G  0 lvm   [SWAP]
    ├─vg0-tmp  253:1    0    50G  0 lvm   /tmp
    └─vg0-root 253:2    0   2.6T  0 lvm   /

$ cat /proc/mdstat
Personalities : [raid1] [raid6] [raid5] [raid4] [linear] [multipath] [raid0] [raid10]
md0 : active raid1 nvme4n1p1[4] nvme1n1p1[2] nvme3n1p1[0] nvme0n1p1[3] nvme2n1p1[1]
      523264 blocks super 1.2 [5/5] [UUUUU]

md1 : active raid5 nvme4n1p2[5] nvme2n1p2[1] nvme1n1p2[2] nvme3n1p2[0] nvme0n1p2[4]
      3748134912 blocks super 1.2 level 5, 512k chunk, algorithm 2 [5/5] [UUUUU]
      bitmap: 3/7 pages [12KB], 65536KB chunk

unused devices: <none>

问题是我的文件系统仍然认为我只连接了 4 个驱动器，并且没有增长到利用额外的驱动器。

我试过了

$ sudo e2fsck -fn /dev/md1
e2fsck 1.45.5 (07-Jan-2020)
Warning!  /dev/md1 is in use.
ext2fs_open2: Bad magic number in super-block
e2fsck: Superblock invalid, trying backup blocks...
e2fsck: Bad magic number in super-block while trying to open /dev/md1

The superblock could not be read or does not describe a valid ext2/ext3/ext4
filesystem.  If the device is valid and it really contains an ext2/ext3/ext4
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>
 or
    e2fsck -b 32768 <device>

/dev/md1 contains a LVM2_member file system

和

$ sudo resize2fs /dev/md1
resize2fs 1.45.5 (07-Jan-2020)
resize2fs: Device or resource busy while trying to open /dev/md1
Couldn't find valid filesystem superblock.

但到目前为止还没有运气：

$ df
Filesystem            1K-blocks       Used Available Use% Mounted on
udev                  131841212          0 131841212   0% /dev
tmpfs                  26374512       2328  26372184   1% /run
/dev/mapper/vg0-root 2681290296 2329377184 215641036  92% /
tmpfs                 131872540          0 131872540   0% /dev/shm
tmpfs                      5120          0      5120   0% /run/lock
tmpfs                 131872540          0 131872540   0% /sys/fs/cgroup
/dev/md0                 498532      86231    386138  19% /boot
/dev/mapper/vg0-tmp    52427196     713248  51713948   2% /tmp
tmpfs                  26374508          0  26374508   0% /run/user/1001
tmpfs                  26374508          0  26374508   0% /run/user/1002

我希望这是足够的信息 - 但如果有用，我很乐意提供更多信息。