Infrastructure at your Service

Soufiane Benmalek

How to reduce the size of an LVM partition formatted with xfs filesystem on CentOS7?

DISCLAIMER: I know it exists other solutions to do it

Pre-requisites:
– a virtual machine (or not) with CentOS7 installed
– a free disk or partition

I use a VBox machine and I added a 5GiB hard disk

We list the disk and partition to check if our new hard is added.

[root@deploy ~]$ lsblk
NAME                       MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
sda                          8:0    0   20G  0 disk
├─sda1                       8:1    0    1G  0 part /boot
└─sda2                       8:2    0   19G  0 part
  ├─cl-root                253:0    0   21G  0 lvm  /
  └─cl-swap                253:1    0    2G  0 lvm  [SWAP]
sdb                          8:16   0   10G  0 disk

Good, we can continue..

Let’s partition the disk using fdisk

[root@deploy ~]$ fdisk /dev/sdb
Welcome to fdisk (util-linux 2.23.2).

Changes will remain in memory only, until you decide to write them.
Be careful before using the write command.

Device does not contain a recognized partition table
Building a new DOS disklabel with disk identifier 0x76a98fa2.

Command (m for help): n

[root@deploy ~]$ fdisk /dev/sdb
Welcome to fdisk (util-linux 2.23.2).

Changes will remain in memory only, until you decide to write them.
Be careful before using the write command.

Device does not contain a recognized partition table
Building a new DOS disklabel with disk identifier 0x76a98fa2.

Command (m for help): n
Partition type:
   p   primary (0 primary, 0 extended, 4 free)
   e   extended
Select (default p): p
Partition number (1-4, default 1): 1
First sector (2048-20971519, default 2048): 
Using default value 2048
Last sector, +sectors or +size{K,M,G} (2048-20971519, default 20971519): +5G
Partition 1 of type Linux and of size 5 GiB is set

Command (m for help): w
The partition table has been altered!

Calling ioctl() to re-read partition table.
Syncing disks.

Now, we need to inform the kernel that the partition table has changed. To do that, either we reboot the server or we run partprobe

[root@deploy ~]$ partprobe /dev/sdb1
[root@deploy ~]$

We create a physical volume

[root@deploy ~]$ pvcreate /dev/sdb1
Physical volume "/dev/sdb1" successfully created.
[root@deploy ~]$ pvs
  PV         VG Fmt  Attr PSize  PFree
  /dev/sda2  cl lvm2 a--  19.00g       0
  /dev/sdb1     lvm2 ---   5.00g    5.00g
  /dev/sdc2  cl lvm2 a--   5.00g 1020.00m

We create a volume group

[root@deploy ~]$ vgcreate  vg_deploy /dev/sdb1
  Volume group "vg_deploy" successfully created

We check that the volume group was created properly

[root@deploy ~]$ vgdisplay vg_deploy
  --- Volume group ---
  VG Name               vg_deploy
  System ID
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  1
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                0
  Open LV               0
  Max PV                0
  Cur PV                1
  Act PV                1
  VG Size               5.00 GiB
  PE Size               4.00 MiB
  Total PE              1279
  Alloc PE / Size       0 / 0
  Free  PE / Size       1279 / 5.00 GiB
  VG UUID               5ZhlvC-lpor-Ti8x-mS9P-bnxW-Gdtw-Gynocl

Here, I set the size of the logical volume with PE (Physical Extent). One PE represents 4.00 MiB

We create a logical volume on our volume group

[root@deploy ~]$ lvcreate -l 1000 -n lv_deploy vg_deploy
  Logical volume "lv_deploy" created.

We have a look to check how our logical volume “lv_deploy” looks like

[root@deploy ~]$ lvdisplay /dev/vg_deploy/lv_deploy
  --- Logical volume ---
  LV Path                /dev/vg_deploy/lv_deploy
  LV Name                lv_deploy
  VG Name                vg_deploy
  LV UUID                2vxcDv-AHfB-7c2x-1PM8-nbn3-38M5-c1QoNS
  LV Write Access        read/write
  LV Creation host, time deploy.example.com, 2017-12-05 08:15:59 -0500
  LV Status              available
  # open                 0
  LV Size                3.91 GiB
  Current LE             1000
  Segments               1
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     8192
  Block device           253:3

Let’s create our file system on the new logical volume

[root@deploy ~]$ mkfs.xfs  /dev/vg_deploy/lv_deploy
meta-data=/dev/vg_deploy/lv_deploy isize=512    agcount=4, agsize=256000 blks
         =                       sectsz=512   attr=2, projid32bit=1
         =                       crc=1        finobt=0, sparse=0
data     =                       bsize=4096   blocks=1024000, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0 ftype=1
log      =internal log           bsize=4096   blocks=2560, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0

We now create a new directory “mysqldata” for example

[root@deploy ~]$ mkdir /mysqldata

We add the new entry for our new logical volume

[root@deploy ~]$ echo "/dev/mapper/vg_deploy-lv_deploy     /mysqldata      xfs     defaults      0 0" >> /etc/fstab

We mount it

[root@deploy ~]$ mount -a

We check the filesystem is mounted properly

[root@deploy ~]$ df -hT
Filesystem                      Type      Size  Used Avail Use% Mounted on
/dev/mapper/cl-root             xfs        21G  8.7G   13G  42% /
devtmpfs                        devtmpfs  910M     0  910M   0% /dev
tmpfs                           tmpfs     920M     0  920M   0% /dev/shm
tmpfs                           tmpfs     920M  8.4M  912M   1% /run
tmpfs                           tmpfs     920M     0  920M   0% /sys/fs/cgroup
/dev/sda1                       xfs      1014M  227M  788M  23% /boot
tmpfs                           tmpfs     184M     0  184M   0% /run/user/0
/dev/loop2                      iso9660   4.3G  4.3G     0 100% /media/iso
/dev/mapper/vg_deploy-lv_deploy xfs       3.9G   33M  3.9G   1% /mysqldata

We add some files to the /mysqldata directory (a for loop will help us)

[root@deploy mysqldata]$ for i in 1 2 3 4 5; do dd if=/dev/zero  of=/mysqldata/file0$i bs=1024 count=10; done
10+0 records in
10+0 records out
10240 bytes (10 kB) copied, 0.000282978 s, 36.2 MB/s
10+0 records in
10+0 records out
10240 bytes (10 kB) copied, 0.000202232 s, 50.6 MB/s
10+0 records in
10+0 records out
10240 bytes (10 kB) copied, 0.000255617 s, 40.1 MB/s
10+0 records in
10+0 records out
10240 bytes (10 kB) copied, 0.000195752 s, 52.3 MB/s
10+0 records in
10+0 records out
10240 bytes (10 kB) copied, 0.000183672 s, 55.8 MB/s
[root@deploy mysqldata]$ ls -l
total 60
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file01
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file02
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file03
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file04
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file05

NOW the interesting part is coming because we are going to reduce our /mysqldata filesystem
But first let’s make a backup of our current /mysqldata FS

[root@deploy mysqldata]$ yum -y install xfsdump
Loaded plugins: fastestmirror
Loading mirror speeds from cached hostfile

Bad news! we cannot reduce an xfs partition directly so we need:
– to backup our filesystem
– umount the filsesytem && delete the logical volume
– re-partition tle logical volume with xfs FS
– restore our data

Backup the file system

[root@deploy mysqldata]$ xfsdump -f /tmp/mysqldata.dump /mysqldata
xfsdump: using file dump (drive_simple) strategy
xfsdump: version 3.1.4 (dump format 3.0) - type ^C for status and control

 ============================= dump label dialog ==============================

please enter label for this dump session (timeout in 300 sec)
 -> test
session label entered: "test"

 --------------------------------- end dialog ---------------------------------

xfsdump: level 0 dump of deploy.example.com:/mysqldata
xfsdump: dump date: Tue Dec  5 08:36:20 2017
xfsdump: session id: f010d421-1a34-4c70-871f-48ffc48c29f2
xfsdump: session label: "test"
xfsdump: ino map phase 1: constructing initial dump list
xfsdump: ino map phase 2: skipping (no pruning necessary)
xfsdump: ino map phase 3: skipping (only one dump stream)
xfsdump: ino map construction complete
xfsdump: estimated dump size: 83840 bytes

 ============================= media label dialog =============================

please enter label for media in drive 0 (timeout in 300 sec)
 -> test
media label entered: "test"

 --------------------------------- end dialog ---------------------------------

xfsdump: creating dump session media file 0 (media 0, file 0)
xfsdump: dumping ino map
xfsdump: dumping directories
xfsdump: dumping non-directory files
xfsdump: ending media file
xfsdump: media file size 75656 bytes
xfsdump: dump size (non-dir files) : 51360 bytes
xfsdump: dump complete: 5 seconds elapsed
xfsdump: Dump Summary:
xfsdump:   stream 0 /tmp/mysqldata.dump OK (success)
xfsdump: Dump Status: SUCCESS

Then, we unmount the filesystem and delete the logical volume

[root@deploy ~]$ umount /mysqldata/

[root@deploy ~]$ df -hT
Filesystem          Type      Size  Used Avail Use% Mounted on
/dev/mapper/cl-root xfs        21G  8.7G   13G  42% /
devtmpfs            devtmpfs  910M     0  910M   0% /dev
tmpfs               tmpfs     920M     0  920M   0% /dev/shm
tmpfs               tmpfs     920M  8.4M  912M   1% /run
tmpfs               tmpfs     920M     0  920M   0% /sys/fs/cgroup
/dev/sda1           xfs      1014M  227M  788M  23% /boot
tmpfs               tmpfs     184M     0  184M   0% /run/user/0
/dev/loop2          iso9660   4.3G  4.3G     0 100% /media/iso

[root@deploy ~]$ lvremove /dev/vg_deploy/lv_deploy
Do you really want to remove active logical volume vg_deploy/lv_deploy? [y/n]: y
  Logical volume "lv_deploy" successfully removed

We recreate the logical volume with a lower size (from 1000 PE to 800 PE)

[root@deploy ~]$ lvcreate -l 800 -n lv_deploy vg_deploy
WARNING: xfs signature detected on /dev/vg_deploy/lv_deploy at offset 0. Wipe it? [y/n]: y
  Wiping xfs signature on /dev/vg_deploy/lv_deploy.
  Logical volume "lv_deploy" created.

We build the XFS filesystem

[root@deploy ~]$ mkfs.xfs /dev/mapper/vg_deploy-lv_deploy
meta-data=/dev/mapper/vg_deploy-lv_deploy isize=512    agcount=4, agsize=204800 blks
         =                       sectsz=512   attr=2, projid32bit=1
         =                       crc=1        finobt=0, sparse=0
data     =                       bsize=4096   blocks=819200, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0 ftype=1
log      =internal log           bsize=4096   blocks=2560, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0

We remount the filesystem

[root@deploy ~]$ mount -a
[root@deploy ~]$
[root@deploy ~]$
[root@deploy ~]$ df -hT
Filesystem                      Type      Size  Used Avail Use% Mounted on
/dev/mapper/cl-root             xfs        21G  8.7G   13G  42% /
devtmpfs                        devtmpfs  910M     0  910M   0% /dev
tmpfs                           tmpfs     920M     0  920M   0% /dev/shm
tmpfs                           tmpfs     920M  8.4M  912M   1% /run
tmpfs                           tmpfs     920M     0  920M   0% /sys/fs/cgroup
/dev/sda1                       xfs      1014M  227M  788M  23% /boot
tmpfs                           tmpfs     184M     0  184M   0% /run/user/0
/dev/loop2                      iso9660   4.3G  4.3G     0 100% /media/iso
/dev/mapper/vg_deploy-lv_deploy xfs       3.2G   33M  3.1G   2% /mysqldata

We list the content of /mysqldata directory

[root@deploy ~]$ ls -l /mysqldata
total 0

Let’s restore our data

[root@deploy ~]$ xfsrestore -f /tmp/mysqldata.dump /mysqldata
xfsrestore: using file dump (drive_simple) strategy
xfsrestore: version 3.1.4 (dump format 3.0) - type ^C for status and control
xfsrestore: searching media for dump
xfsrestore: examining media file 0
xfsrestore: dump description:
xfsrestore: hostname: deploy.example.com
xfsrestore: mount point: /mysqldata
xfsrestore: volume: /dev/mapper/vg_deploy-lv_deploy
xfsrestore: session time: Tue Dec  5 08:36:20 2017
xfsrestore: level: 0
xfsrestore: session label: "test"
xfsrestore: media label: "test"
xfsrestore: file system id: 84832e04-e6b8-473a-beb4-f4d59ab9e73c
xfsrestore: session id: f010d421-1a34-4c70-871f-48ffc48c29f2
xfsrestore: media id: 8fda43c1-c7de-4331-b930-ebd88199d0e7
xfsrestore: using online session inventory
xfsrestore: searching media for directory dump
xfsrestore: reading directories
xfsrestore: 1 directories and 5 entries processed
xfsrestore: directory post-processing
xfsrestore: restoring non-directory files
xfsrestore: restore complete: 0 seconds elapsed
xfsrestore: Restore Summary:
xfsrestore:   stream 0 /tmp/mysqldata.dump OK (success)
xfsrestore: Restore Status: SUCCESS

Our data are back

[root@deploy ~]$ ls -l /mysqldata/
total 60
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file01
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file02
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file03
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file04
-rw-r--r--. 1 root root 10240 Dec  5 08:28 file05

Hope this helps :-)

 

Leave a Reply


3 × seven =

Soufiane Benmalek
Soufiane Benmalek

Consultant