resize2fs: Bad magic number in super-block while trying to open

Question

I am trying to resize a logical volume on CentOS7 but am running into the following error:

resize2fs 1.42.9 (28-Dec-2013)
resize2fs: Bad magic number in super-block while trying to open /dev/mapper/centos-root
Couldn't find valid filesystem superblock.

I have tried adding a new partition (using fdisk) and using vgextend to extend the volume group, then resizing. Resize worked fine for the logical volume using lvextend, but it failed at resize2fs.

I have also tried deleting an existing partition (using fdisk) and recreating it with a larger end block, then resizing the physical volume using lvm pvresize, followed by a resize of the logical volume using lvm lvresize. Again everything worked fine up to this point.

Once I tried to use resize2fs, using both methods as above, I received the exact same error.

Hopefully some of the following will shed some light.

fdisk -l

[root@server~]# fdisk -l

Disk /dev/xvda: 32.2 GB, 32212254720 bytes, 62914560 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk label type: dos
Disk identifier: 0x0009323a

Device Boot      Start         End      Blocks   Id  System
/dev/xvda1   *        2048     1026047      512000   83  Linux
/dev/xvda2         1026048    41943039    20458496   8e  Linux LVM
/dev/xvda3        41943040    62914559    10485760   8e  Linux LVM

Disk /dev/mapper/centos-swap: 2147 MB, 2147483648 bytes, 4194304 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes


Disk /dev/mapper/centos-root: 29.5 GB, 29532094464 bytes, 57679872 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes

pvdisplay

[root@server ~]# pvdisplay
--- Physical volume ---
PV Name               /dev/xvda2
VG Name               centos
PV Size               19.51 GiB / not usable 2.00 MiB
Allocatable           yes (but full)
PE Size               4.00 MiB
Total PE              4994
Free PE               0
Allocated PE          4994
PV UUID               7bJOPh-OUK0-dGAs-2yqL-CAsV-TZeL-HfYzCt

--- Physical volume ---
PV Name               /dev/xvda3
VG Name               centos
PV Size               10.00 GiB / not usable 4.00 MiB
Allocatable           yes (but full)
PE Size               4.00 MiB
Total PE              2559
Free PE               0
Allocated PE          2559
PV UUID               p0IClg-5mrh-5WlL-eJ1v-t6Tm-flVJ-gsJOK6

vgdisplay

[root@server ~]# vgdisplay
--- Volume group ---
VG Name               centos
System ID
Format                lvm2
Metadata Areas        2
Metadata Sequence No  6
VG Access             read/write
VG Status             resizable
MAX LV                0
Cur LV                2
Open LV               2
Max PV                0
Cur PV                2
Act PV                2
VG Size               29.50 GiB
PE Size               4.00 MiB
Total PE              7553
Alloc PE / Size       7553 / 29.50 GiB
Free  PE / Size       0 / 0
VG UUID               FD7k1M-koJt-2veW-sizL-Srsq-Y6zt-GcCfz6

lvdisplay

[root@server ~]# lvdisplay
--- Logical volume ---
LV Path                /dev/centos/swap
LV Name                swap
VG Name                centos
LV UUID                KyokrR-NGsp-6jVA-P92S-QE3X-hvdp-WAeACd
LV Write Access        read/write
LV Creation host, time localhost, 2014-10-09 08:28:42 +0100
LV Status              available
# open                 2
LV Size                2.00 GiB
Current LE             512
Segments               1
Allocation             inherit
Read ahead sectors     auto
- currently set to     8192
Block device           253:0

--- Logical volume ---
LV Path                /dev/centos/root
LV Name                root
VG Name                centos
LV UUID                ugCOcT-sTDK-M8EV-3InM-hjIg-2nwS-KeAOnq
LV Write Access        read/write
LV Creation host, time localhost, 2014-10-09 08:28:42 +0100
LV Status              available
# open                 1
LV Size                27.50 GiB
Current LE             7041
Segments               2
Allocation             inherit
Read ahead sectors     auto
- currently set to     8192
Block device           253:1

I've probably done something stupid, so any help would be greatly appreciated!

Dispatch

possible duplicate of Trying to resize2fs EB volume fails – Virology

Get the location of the backup superblock: newfs -N /dev/device-name & Restoring a bad superblock via FSCK: fsck -F ufs -o b=<Superblock_stored_location> /dev/device-name Source: docs.oracle.com/cd/E19455-01/805-7228/6j6q7uf0i/index.html – Thrombocyte

Answer 1

After a bit of trial and error... as mentioned in the possible answers, it turned out to require xfs_growfs rather than resize2fs.

CentOS 7,

fdisk /dev/xvda

Create new primary partition, set type as linux lvm.

n
p
3
t
8e
w

Create a new primary volume and extend the volume group to the new volume.

partprobe
pvcreate /dev/xvda3
vgextend /dev/centos /dev/xvda3

Check the physical volume for free space, extend the logical volume with the free space.

vgdisplay -v
lvextend -l+288 /dev/centos/root

Finally perform an online resize to resize the logical volume, then check the available space.

xfs_growfs /dev/centos/root
df -h

Dispatch

Works for RHEL 7 too. Thanks! – Pallid

How to shrink one partition and add the shrinked space to another partition? xfs_growfs seem doesn't work for this. – Allegraallegretto

you are a legend, ty so much for this guide, didn't realise ubuntu/centos uses different resize method – Malenamalet

Here you can find more information: tecmint.com/extend-and-reduce-lvms-in-linux – Bicyclic

I realize that we don't need to use pvcreate explicitely before vgextend command. When using vgextend, it creates pv automatically. – Righteous

this is work on centos 7 when installed directadmin . – Dyad

It should be mentioned that the filesystem is displayed in /etc/fstab. That's how it's determined to use xfs_growfs rather than resize2fs. – Axenic

xfs_growfs is for XFS filesystem. If someone were to install the CentOS with another filesystem (ext4), xfs_growfs will fail. Use fsadm which handles xfs, ext4 and a few filesystem. – Whiplash

This answer is mostly correct, but xfs_growfs expects a MOUNT POINT, not a DEVICE! So "xfs_growfs /dev/centos/root" is not correct, it'd be "xfs_growfs /wherever/you/mounted/it" This took me a while to figure out: linux.die.net/man/8/xfs_growfs At least on Fedora 31 specifying DEVICE did not work for me. – Anglicist

Had this issue in Cloud9 environment (AWS EC2 AMI Amazon Linux 2) - this solution worked for me :) – Pacificas

Answer 2

In Centos 7 default filesystem is xfs.

xfs file system support only extend not reduce. So if you want to resize the filesystem use xfs_growfs rather than resize2fs.

xfs_growfs /dev/root_vg/root 

Note: For ext4 filesystem use

resize2fs /dev/root_vg/root

Goblin

One can find their filesystem type using mount | column -t (among many other ways). – Boxboard

You need to spevify mount point, not block device. xfs_growfs version 5.1.0, Fedora 31, Usage: xfs_growfs [options] mountpoint – Anile

Answer 3

resize2fs Command will not work for all file systems.

Please confirm the file system of your instance using below command.

Please follow the procedure to expand volume by following the steps mentioned in Amazon official document for different file systems.

https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/recognize-expanded-volume-linux.html

Default file system in Centos is xfs, use the following command for xfs file system to increase partition size.

sudo xfs_growfs -d /

then "df -h" to check.

Logician

My partition was created using parted->mkpart. Finally this one worked for me. – Secondclass

Correct. Check type. – Opinicus

Answer 4

I ran into the same exact problem around noon today and finally found a solution here --> Trying to resize2fs EB volume fails

I skipped mounting, since the partition was already mounted.

Apparently CentOS 7 uses XFS as the default file system and as a result resize2fs will fail.

I took a look in /etc/fstab, and guess what, XFS was staring me in the face... Hope this helps.

Toadflax

I ran lvextend --resize which resizes for you in centos 7. This seemed to fix the issue. – Naji

man page of lvextend says it uses fsadm to resize filesystem. So i tried fsadm resize /dev/centos/root and it worked. – Asseverate

Technically, this is not really an answer, and the information on that URL did not work for me. Best to answer the question here, where we can discuss it, and add links to supporting documentation if you think that's helpful. – Selwyn

Answer 5

On centos and fedora work with fsadm

fsadm resize /dev/vg_name/root

Fokos

Answer 6

CentOS7 + VM

Ive made it with:

1. Gparted-live extend the volume
2. pvresize -v /dev/sda2
3. lvresize -r -l+100%FREE centos/root

Kiln

Answer 7

On Centos 7, in answer to the original question where resize2fs fails with "bad magic number" try using fsadm as follows:

fsadm resize /dev/the-device-name-returned-by-df

Then:

df 

... to confirm the size changes have worked.

Headwaiter

Worked for me too on CentOS VM with LVM partitions resized. – Fanfani

Answer 8

After reading about LVM and being familiar with PV -> VG -> LV, this works for me :

0) #df -h

Filesystem               Size  Used Avail Use% Mounted on
devtmpfs                 1.9G     0  1.9G   0% /dev
tmpfs                    1.9G     0  1.9G   0% /dev/shm
tmpfs                    1.9G  824K  1.9G   1% /run
tmpfs                    1.9G     0  1.9G   0% /sys/fs/cgroup
/dev/mapper/fedora-root   15G  2.1G   13G  14% /
tmpfs                    1.9G     0  1.9G   0% /tmp
/dev/md126p1             976M  119M  790M  14% /boot
tmpfs                    388M     0  388M   0% /run/user/0

1) # vgs

  VG     #PV #LV #SN Attr   VSize   VFree   
  fedora   1   2   0 wz--n- 231.88g 212.96g

2) # vgdisplay

  --- Volume group ---
  VG Name               fedora
  System ID
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  3
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                2
  Open LV               2
  Max PV                0
  Cur PV                1
  Act PV                1
  VG Size               231.88 GiB
  PE Size               4.00 MiB
  Total PE              59361
  Alloc PE / Size       4844 / 18.92 GiB
  Free  PE / Size       54517 / 212.96 GiB
  VG UUID               9htamV-DveQ-Jiht-Yfth-OZp7-XUDC-tWh5Lv

3) # lvextend -l +100%FREE /dev/mapper/fedora-root

  Size of logical volume fedora/root changed from 15.00 GiB (3840 extents) to 227.96 GiB (58357 extents).
  Logical volume fedora/root successfully resized.

4) #lvdisplay

5) #fd -h

6) # xfs_growfs /dev/mapper/fedora-root

meta-data=/dev/mapper/fedora-root isize=512    agcount=4, agsize=983040 blks
         =                       sectsz=512   attr=2, projid32bit=1
         =                       crc=1        finobt=1 spinodes=0 rmapbt=0
         =                       reflink=0
data     =                       bsize=4096   blocks=3932160, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0 ftype=1
log      =internal               bsize=4096   blocks=2560, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0
data blocks changed from 3932160 to 59757568

7) #df -h

Filesystem               Size  Used Avail Use% Mounted on
devtmpfs                 1.9G     0  1.9G   0% /dev
tmpfs                    1.9G     0  1.9G   0% /dev/shm
tmpfs                    1.9G  828K  1.9G   1% /run
tmpfs                    1.9G     0  1.9G   0% /sys/fs/cgroup
/dev/mapper/fedora-root  228G  2.3G  226G   2% /
tmpfs                    1.9G     0  1.9G   0% /tmp
/dev/md126p1             976M  119M  790M  14% /boot
tmpfs                    388M     0  388M   0% /run/user/0

Best Regards,

Edholm

Fedora Server 35 on Raspberry Pi, working as a charm. No other solution worked, thanks for the detailed steps! – Dakota

works for Centos8 as well. thanks – Darsie

Answer 9

os: rhel7

After gparted, # xfs_growfs /dev/mapper/rhel-root did the trick on a living system.

$ df -h
Filesystem             Size  Used Avail Use% Mounted on
/dev/mapper/rhel-root   47G   47G   20M 100% /
devtmpfs               1.9G     0  1.9G   0% /dev
tmpfs                  1.9G     0  1.9G   0% /dev/shm
tmpfs                  1.9G  9.3M  1.9G   1% /run
tmpfs                  1.9G     0  1.9G   0% /sys/fs/cgroup
/dev/sda1             1014M  205M  810M  21% /boot
tmpfs                  379M  8.0K  379M   1% /run/user/42
tmpfs                  379M     0  379M   0% /run/user/1000


# lvresize -l +100%FREE /dev/mapper/rhel-root
  Size of logical volume rhel/root changed from <47.00 GiB (12031 extents) to <77.00 GiB (19711 extents).
  Logical volume rhel/root successfully resized.


# xfs_growfs /dev/mapper/rhel-root
meta-data=/dev/mapper/rhel-root  isize=512    agcount=7, agsize=1900032 blks
         =                       sectsz=512   attr=2, projid32bit=1
         =                       crc=1        finobt=0 spinodes=0
data     =                       bsize=4096   blocks=12319744, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0 ftype=1
log      =internal               bsize=4096   blocks=3711, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0
data blocks changed from 12319744 to 20184064


# df -h
Filesystem             Size  Used Avail Use% Mounted on
/dev/mapper/rhel-root   77G   47G   31G  62% /
devtmpfs               1.9G     0  1.9G   0% /dev
tmpfs                  1.9G     0  1.9G   0% /dev/shm
tmpfs                  1.9G  9.3M  1.9G   1% /run
tmpfs                  1.9G     0  1.9G   0% /sys/fs/cgroup
/dev/sda1             1014M  205M  810M  21% /boot
tmpfs                  379M  8.0K  379M   1% /run/user/42
tmpfs                  379M     0  379M   0% /run/user/1000

Aldehyde

Answer 10

How to resize root partition online :  

1)  [root@oel7 ~]# df -h  
Filesystem                         Size  Used Avail Use% Mounted on  
/dev/mapper/root_vg-root           5.0G  4.5G  548M  90% /

2)   

PV /dev/sda2   VG root_vg     lvm2 [6.00 GiB / 0    free]

as here it shows that there is no space left on root_vg volume group, so first i need to extend VG 

3)  
[root@oel7 ~]# vgextend root_vg /dev/sdb5  

  Volume group "root_vg" successfully extended

4)  

[root@oel7 ~]# pvscan  
    PV /dev/sda2   VG root_vg     lvm2 [6.00 GiB / 0    free]  
    PV /dev/sdb5   VG root_vg     lvm2 [2.00 GiB / 2.00 GiB free]

5)  Now extend the logical volume   

[root@oel7 ~]# lvextend -L +1G /dev/root_vg/root  

  Size of logical volume root_vg/root changed from 5.00 GiB (1280 extents) to 6.00 GiB (1536 extents).
  Logical volume root successfully resized  

3)  [root@oel7 ~]# resize2fs /dev/root_vg/root  

resize2fs 1.42.9 (28-Dec-2013)  
resize2fs: Bad magic number in super-block while trying to open /dev/root_vg  /root  
Couldn't find valid filesystem superblock.  

as root partition is not a ext* partiton so , you resize2fs will not work for you.  

4)  to check the filesystem type of a partition 
[root@oel7 ~]# df -Th  
Filesystem                        Type      Size  Used Avail Use% Mounted on  
/dev/mapper/root_vg-root          xfs       6.0G  4.5G  1.6G  75% /  
devtmpfs                          devtmpfs  481M     0  481M   0% /dev  
tmpfs                             tmpfs     491M   80K  491M   1% /dev/shm  
tmpfs                             tmpfs     491M  7.1M  484M   2% /run  
tmpfs                             tmpfs     491M     0  491M   0% /sys/fs  /cgroup
/dev/mapper/data_vg-home          xfs       3.5G  2.9G  620M  83% /home  
/dev/sda1                         xfs       497M  132M  365M  27% /boot  
/dev/mapper/data_vg01-data_lv001  ext3      4.0G  2.4G  1.5G  62% /sybase  
/dev/mapper/data_vg02-backup_lv01 ext3      4.0G  806M  3.0G  22% /backup  

above command shows that root is an xfs filesystem , so we are sure that we need to use xfs_growfs command to resize the partition.

6)  [root@oel7 ~]# xfs_growfs /dev/root_vg/root  
meta-data=/dev/mapper/root_vg-root isize=256    agcount=4, agsize=327680 blks
         =                       sectsz=512   attr=2, projid32bit=1
         =                       crc=0        finobt=0
data     =                       bsize=4096   blocks=1310720, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0 ftype=0
log      =internal               bsize=4096   blocks=2560, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0
data blocks changed from 1310720 to 1572864  

[root@oel7 ~]# df -Th  

Filesystem                        Type      Size  Used Avail Use% Mounted on
/dev/mapper/root_vg-root          xfs       6.0G  4.5G  1.6G  75% /  

Lightning

Answer 11

To resize the existing volume mounted

sudo mount -t xfs /dev/sdf /opt/data/

mount: /opt/data: /dev/nvme1n1 already mounted on /opt/data.

sudo xfs_growfs /opt/data/

Saavedra

Worked on my ec2 instance, in my case I wanted to align the main volume to the new increased size. After using sudo growpart /dev/xvda 1, just issue sudo xfs_growfs /dev/xvda1. Now df -h will reflect the size of your volume, without need to refresh. Of course change your mount path accordingly – Rime

Answer 12

In my case I could to fix the superblock location with these command:

yum install gdisk
parted -l /dev/mapper/centos-root
growpart /dev/mapper/centos-root 1
xfs_growfs /

Unnerve