mount mount experiment

The classmates who developed it today called and asked a strange question: Can multiple nas be mounted in the same folder?
My first reaction was no, but I didn't think about why.

Now let's do an experiment to verify:

Can the same directory mount multiple partitions/hard disks/logical volumes/nas at the same time?
Can the same partition/hard disk/logical volume/nas be mounted to multiple directories

Environmental preparation

Experimental Baseline
Operating System: CentOS Linux release 8.4.2105
The virtual machine creates 3 new hard disks as experimental mounts
View hard drive

[root@C8-196 ~]# lsblk
NAME        MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sda           8:0    0  200G  0 disk 
├─sda1        8:1    0    1G  0 part /boot
└─sda2        8:2    0  199G  0 part 
  ├─cl-root 253:0    0   70G  0 lvm  /
  ├─cl-swap 253:1    0    2G  0 lvm  [SWAP]
  └─cl-home 253:2    0  127G  0 lvm  /home
sdb           8:16   0    5G  0 disk 
sdc           8:32   0    5G  0 disk 
sdd           8:48   0    5G  0 disk

Using logical volume partitions

[root@C8-196 ~]# pvcreate /dev/sdb /dev/sdc /dev/sdd
  Physical volume "/dev/sdb" successfully created.
  Physical volume "/dev/sdc" successfully created.
  Physical volume "/dev/sdd" successfully created.
[root@C8-196 ~]# vgcreate mt /dev/sdb /dev/sdc /dev/sdd
  Volume group "mt" successfully created
[root@C8-196 ~]# lvcreate -L +5G mt1_lv mt
  Volume group "mt1_lv" not found
  Cannot process volume group mt1_lv
[root@C8-196 ~]# lvcreate -L +5G -n mt1_lv mt
  Logical volume "mt1_lv" created.
[root@C8-196 ~]# lvcreate -L +5G -n mt2_lv mt
  Logical volume "mt2_lv" created.
[root@C8-196 ~]# lvcreate -L +5G -n mt3_lv mt
  Volume group "mt" has insufficient free space (1277 extents): 1280 required.
[root@C8-196 ~]# lvcreate -L +4 -n mt3_lv mt
  Logical volume "mt3_lv" created.
[root@C8-196 ~]# lsblk
NAME        MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sda           8:0    0  200G  0 disk 
├─sda1        8:1    0    1G  0 part /boot
└─sda2        8:2    0  199G  0 part 
  ├─cl-root 253:0    0   70G  0 lvm  /
  ├─cl-swap 253:1    0    2G  0 lvm  [SWAP]
  └─cl-home 253:2    0  127G  0 lvm  /home
sdb           8:16   0    5G  0 disk 
└─mt-mt1_lv 253:3    0    5G  0 lvm  
sdc           8:32   0    5G  0 disk 
├─mt-mt1_lv 253:3    0    5G  0 lvm  
├─mt-mt2_lv 253:4    0    5G  0 lvm  
└─mt-mt3_lv 253:5    0    4M  0 lvm  
sdd           8:48   0    5G  0 disk 
└─mt-mt2_lv 253:4    0    5G  0 lvm  
[root@C8-196 ~]# lvs
  LV     VG Attr       LSize   Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert
  home   cl -wi-ao---- 126.96g                                                    
  root   cl -wi-ao----  70.00g                                                    
  swap   cl -wi-ao----   2.03g                                                    
  mt1_lv mt -wi-a-----   5.00g                                                    
  mt2_lv mt -wi-a-----   5.00g                                                    
  mt3_lv mt -wi-a-----   4.00m

Create a file system on the three newly created logical volume partitions

 mkfs.xfs /dev/mapper/mt-mt1_lv
 mkfs.xfs /dev/mapper/mt-mt2_lv
 mkfs.ext4 /dev/mapper/mt-mt3_lv

View partition status

[root@C8-196 ~]# lsblk -f
NAME        FSTYPE      LABEL UUID                                   MOUNTPOINT
sda                                                                  
├─sda1      xfs               c17463f3-cf5c-4d9e-b7f8-34e1f09de31c   /boot
└─sda2      LVM2_member       5fOYEk-G1PN-Yilu-3PZD-dV92-kCgM-4NQTC4 
  ├─cl-root xfs               e88eb208-9984-4cbe-867b-8a0fee022dec   /
  ├─cl-swap swap              ad9f0577-fb9a-4e61-99d2-b42473a0e815   [SWAP]
  └─cl-home xfs               19122a86-a5b9-43c8-bdb3-6ae3a94702bb   /home
sdb         LVM2_member       vv0NK7-BQXM-Z2dp-QSnp-W3Ff-qdfz-qDkM2K 
└─mt-mt1_lv xfs               540120f8-557d-4a7b-95a0-8f3965e212a5   
sdc         LVM2_member       FQRPl7-L1K5-z6PK-o3wR-rTz9-0tGj-hCMB8s 
├─mt-mt1_lv xfs               540120f8-557d-4a7b-95a0-8f3965e212a5   
├─mt-mt2_lv xfs               01a90cf9-4049-45f3-95c3-acb6f879cba0   
└─mt-mt3_lv ext4              597e10e6-e1ad-4d04-ac95-304b0a61de58   
sdd         LVM2_member       7hKy5j-Yik1-mpdE-NXir-J84W-a4c7-JmTZLE 
└─mt-mt2_lv xfs               01a90cf9-4049-45f3-95c3-acb6f879cba0

Create 3 mount points

[root@C8-196 ~]# mkdir -pv /data/mt{1..3}
mkdir: created directory '/data/mt1'
mkdir: created directory '/data/mt2'
mkdir: created directory '/data/mt3'
[root@C8-196 ~]# ll /data
total 0
drwxr-xr-x 2 root root 6 Mar 10 21:05 mt1
drwxr-xr-x 2 root root 6 Mar 10 21:05 mt2
drwxr-xr-x 2 root root 6 Mar 10 21:05 mt3
drwxr-xr-x 2 root root 6 Dec 17 21:43 spugdata

Test 1: Mount multiple partitions on the same mount point (directory)

Mount point /data/mt1 mounts 3 partitions respectively

[root@C8-196 ~]# mount /dev/mapper/mt-mt1_lv /data/mt1
[root@C8-196 ~]# df -Th
Filesystem            Type      Size  Used Avail Use% Mounted on
devtmpfs              devtmpfs  876M     0  876M   0% /dev
tmpfs                 tmpfs     896M     0  896M   0% /dev/shm
tmpfs                 tmpfs     896M   17M  879M   2% /run
tmpfs                 tmpfs     896M     0  896M   0% /sys/fs/cgroup
/dev/mapper/cl-root   xfs        70G  3.6G   67G   6% /
/dev/mapper/cl-home   xfs       127G  2.5G  125G   2% /home
/dev/sda1             xfs      1014M  197M  818M  20% /boot
overlay               overlay    70G  3.6G   67G   6% /var/lib/docker/overlay2/dff1098e902038f4105591060554c84a668a85b076f0596f8b144e77ca7a8633/merged
tmpfs                 tmpfs     180M     0  180M   0% /run/user/0
/dev/mapper/mt-mt1_lv xfs       5.0G   68M  5.0G   2% /data/mt1
[root@C8-196 ~]# mount /dev/mapper/mt-mt2_lv /data/mt1
[root@C8-196 ~]# df -Th
Filesystem            Type      Size  Used Avail Use% Mounted on
##上边的省略
/dev/mapper/mt-mt2_lv xfs       5.0G   68M  5.0G   2% /data/mt1
[root@C8-196 ~]# mount /dev/mapper/mt-mt3_lv /data/mt1
[root@C8-196 ~]# df -Th
Filesystem            Type      Size  Used Avail Use% Mounted on
##上边的省略
/dev/mapper/mt-mt3_lv ext4      2.9M   45K  2.6M   2% /data/mt1

Obviously, it is impossible to mount at the same time,
When the second partition is mounted, the first partition is dropped
When the third partition is mounted, the second partition is dropped
But when I umount something strange happens

[root@C8-196 ~]# umount /data/mt1
[root@C8-196 ~]# df -Th
Filesystem            Type      Size  Used Avail Use% Mounted on
##上边的省略
/dev/mapper/mt-mt2_lv xfs       5.0G   68M  5.0G   2% /data/mt1

Did you see that, after umount, it does not return to the unmounted state, but returns to the state of the previous mount point

Take a look at the partition

The truth is, the same directory can mount multiple partitions

[root@C8-196 ~]# lsblk
NAME        MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sda           8:0    0  200G  0 disk 
├─sda1        8:1    0    1G  0 part /boot
└─sda2        8:2    0  199G  0 part 
  ├─cl-root 253:0    0   70G  0 lvm  /
  ├─cl-swap 253:1    0    2G  0 lvm  [SWAP]
  └─cl-home 253:2    0  127G  0 lvm  /home
sdb           8:16   0    5G  0 disk 
└─mt-mt1_lv 253:3    0    5G  0 lvm  /data/mt1
sdc           8:32   0    5G  0 disk 
├─mt-mt1_lv 253:3    0    5G  0 lvm  /data/mt1
├─mt-mt2_lv 253:4    0    5G  0 lvm  /data/mt1
└─mt-mt3_lv 253:5    0    4M  0 lvm  
sdd           8:48   0    5G  0 disk 
└─mt-mt2_lv 253:4    0    5G  0 lvm  /data/mt1

Let's mount mt-mt3 on /data/mt1
Obviously, the same mount point can mount multiple partitions at the same time
Obviously multiple partitions can be mounted on the same mount point at the same time

reading and writing test

Since they are all hung up, we should write data to all partitions at the same time when we write data to this mount point.
We now write data to the /data/mt1 folder

[root@C8-196 ~]# touch /data/mt1/mt{1..10}.log 
[root@C8-196 ~]# ll /data/mt1
total 12
drwx------ 2 root root 12288 Mar 10 21:02 lost+found
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt10.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt1.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt2.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt3.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt4.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt5.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt6.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt7.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt8.log
-rw-r--r-- 1 root root     0 Mar 10 21:20 mt9.log
[root@C8-196 ~]# ll /data/mt2
total 0
[root@C8-196 ~]# ll /data/mt3
total 0

At the same time, it can be seen that /data/mt2 and /data/mt3 are empty
At this time, the three partitions are all mounted on /data/mt1, which means that the three disks should have the same content.
We dismantle the two partitions and mount them separately

[root@C8-196 ~]# mount /dev/mapper/mt-mt2_lv /data/mt2
[root@C8-196 ~]# mount /dev/mapper/mt-mt3_lv /data/mt3

Strange things happened again, this time only mt1 and mt3 have data, but there is no data in mt2
The mount situation at this time is like this
At this time, the partition situation is like this
Since there is a supernatural event, we all take it off and remount it
This time the world is clean
Re-mount to the corresponding directory to see the data

[root@C8-196 ~]# mount /dev/mapper/mt-mt1_lv /data/mt1
[root@C8-196 ~]# mount /dev/mapper/mt-mt2_lv /data/mt2
[root@C8-196 ~]# mount /dev/mapper/mt-mt3_lv /data/mt3

remount

As a result, there is only data on mt3
It also means that the data is written on the partition /dev/mapper/mt-mt3_lv
That is to say, the last mounted disk /dev/mapper/mt-mt3_lv has actually written data.
That is to say, one mount point cannot be mounted on multiple partitions at the same time.

Test 2: Mount the same partition to multiple mount points (directories)

Initialize all mount points and partitions
Mount mt-mt1_lv to /data/mt1 /data/mt2 /data/mt3 respectively
We successively mounted 3 directories in the same partition without umount,
But df shows the same directory, no change
But when I run lsblk, I see that the partition has been mounted to mt3
- We know that the data actually exists on the partition, so let's see if there is any content in the current directory
All directories are currently empty
Then let's write the data and see

write 123 in directory mt1
touch /data/mt1/mt{1..3}.log
write 456 in directory mt2
touch /data/mt3/mt{4..6}.log

The time to witness the miracle is here, let's watch mt2

Witness the moment of miracle

Although neither df nor lsblk shows the mount status of /data/mt2,
But we saw that the contents of mt1 and mt3 were written in the original empty directory mt2.
Then we can be sure that mt1 and mt3 now have the same data as mt2

mt1 and mt3 have the same data

My eyes are rulers, no need to watch playback! ! !

Summarize

The same directory cannot mount multiple partitions at the same time
. Even if it seems to be mounted on lsblk, only the last mounted one will take effect
Multiple partitions cannot be mounted on the same directory at the same time.
Even if they can be mounted, the same data will not be written.
The same partition can be mounted to multiple directories at the same time.
Even if df and lsblk look different, as long as there is no umount, it is considered to be mounted.
Multiple directories can mount the same partition at the same time
. The data written to any directory is actually written to the same partition.

This is the same data in the same partition!

make a metaphor,

A partition or disk or nfs can be imagined as a house
The directory, the mount point, can be like a door

A room can have multiple doors, whichever door enters and exits to this house.
A door can only be installed on one house, not two houses, only one room can be entered through one door and multiple rooms cannot be entered through one door.

Multiple mount points mount experiment under Linux