mjhjmtu

Im trying to remove the added disk to the VM, as the disk space is not needed anymore.
However, it looks to be that the VM has in the VG almost all the data used what is available there.

However, i've managed to resize the zoneminder--vg to 6G almost from 1,5TB
with resiz2fs but lvdisplay says differently.

Here is the output of some commands.

root@zoneminder:~# lvs
 LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
 root zoneminder-vg -wi-ao---- 1.52t
 swap_1 zoneminder-vg -wi-ao---- 976.00m
root@zoneminder:~# pvs
 PV VG Fmt Attr PSize PFree
 /dev/sda3 zoneminder-vg lvm2 a-- 900.00g 0
 /dev/sda5 zoneminder-vg lvm2 a-- 699.52g 46.57g
root@zoneminder:~# vgs
 VG #PV #LV #SN Attr VSize VFree
 zoneminder-vg 2 2 0 wz--n- 1.56t 46.57g

DF - H

Filesystem Size Used Avail Use% Mounted on
udev 7.9G 0 7.9G 0% /dev
tmpfs 1.6G 8.9M 1.6G 1% /run
/dev/mapper/zoneminder--vg-root 5.6G 4.9G 431M 92% /
tmpfs 7.9G 0 7.9G 0% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 7.9G 0 7.9G 0% /sys/fs/cgroup
/dev/sda1 472M 108M 340M 25% /boot
tmpfs 1.6G 0 1.6G 0% /run/user/1000

pvdisplay

 --- Physical volume ---
 PV Name /dev/sda5
 VG Name zoneminder-vg
 PV Size 699.52 GiB / not usable 2.00 MiB
 Allocatable yes
 PE Size 4.00 MiB
 Total PE 179077
 Free PE 11922
 Allocated PE 167155
 PV UUID SVGqoc-SQ42-tDzp-Qc7H-n90f-1g9n-x0eLWe

 --- Physical volume ---
 PV Name /dev/sda3
 VG Name zoneminder-vg
 PV Size 900.00 GiB / not usable 0
 Allocatable yes (but full)
 PE Size 4.00 MiB
 Total PE 230400
 Free PE 0
 Allocated PE 230400
 PV UUID Cdanv0-2pLJ-Yp2n-3zsl-JvjH-72QS-Ciwhaj

lvdisplay

 --- Logical volume ---
 LV Path /dev/zoneminder-vg/root
 LV Name root
 VG Name zoneminder-vg
 LV UUID poThtY-v96W-e2Ai-nan7-ckqn-aeBm-T0Kqji
 LV Write Access read/write
 LV Creation host, time zoneminder, 2018-08-01 22:22:13 +0200
 LV Status available
 # open 1
 LV Size 1.52 TiB
 Current LE 397311
 Segments 2
 Allocation inherit
 Read ahead sectors auto
 - currently set to 256
 Block device 252:0

 --- Logical volume ---
 LV Path /dev/zoneminder-vg/swap_1
 LV Name swap_1
 VG Name zoneminder-vg
 LV UUID SXQ36r-5Kum-Z3Wa-m9DE-CBVb-h9Wx-kmctKT
 LV Write Access read/write
 LV Creation host, time zoneminder, 2018-08-01 22:22:13 +0200
 LV Status available
 # open 2
 LV Size 976.00 MiB
 Current LE 244
 Segments 1
 Allocation inherit
 Read ahead sectors auto
 - currently set to 256
 Block device 252:1

vgdisplay

 --- Volume group ---
 VG Name zoneminder-vg
 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 1.56 TiB
 PE Size 4.00 MiB
 Total PE 409477

 Alloc PE / Size 397555 / 1.52 TiB
 Free PE / Size 11922 / 46.57 GiB

 VG UUID lTo8U0-dIL9-Yye3-RVYk-rJu6-w6WQ-zIpL8f

How can i make the /dev/sda3 remove without hitting any data on the LVM?
And then rebuilt the VGroup back again, to 100G or something?

fdisk -l

Disk /dev/sda: 1.6 TiB, 1717986918400 bytes, 3355443200 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
Disklabel type: dos
Disk identifier: 0x21880f4a

Device Boot Start End Sectors Size Id Type
/dev/sda1 * 2048 999423 997376 487M 83 Linux
/dev/sda2 1001470 1468004351 1467002882 699.5G 5 Extended
/dev/sda3 1468004352 3355443199 1887438848 900G 8e Linux LVM
/dev/sda5 1001472 1468004351 1467002880 699.5G 8e Linux LVM

Partition table entries are not in disk order.


Disk /dev/mapper/zoneminder--vg-root: 1.5 TiB, 1666443116544 bytes, 3254771712 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/zoneminder--vg-swap_1: 976 MiB, 1023410176 bytes, 1998848 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
root@zoneminder:~#

Update#1:

@telcoM : Thank you for you kind and detailed answer

All the actions mentioned by you where done.
I came across also that the data was ofcourse spread all over the disk, so I had to physically move the data also:

root@zoneminder:~# pvresize --setphysicalvolume 101G -v /dev/sda5
 Using physical volume(s) on command line.
 Archiving volume group "zoneminder-vg" metadata (seqno 9).
 /dev/sda5: Pretending size is 211812352 not 1467002880 sectors.
 Resizing volume "/dev/sda5" to 211812352 sectors.
 Resizing physical volume /dev/sda5 from 0 to 25855 extents.
 /dev/sda5: cannot resize to 25855 extents as later ones are allocated.
 0 physical volume(s) resized / 1 physical volume(s) not resized


root@zoneminder:~# pvs -v --segments /dev/sda5
 Using physical volume(s) on command line.
 Wiping cache of LVM-capable devices
 PV VG Fmt Attr PSize PFree Start SSize LV Start Type PE Ranges
 /dev/sda5 zoneminder-vg lvm2 a-- 699.52g 598.57g 0 25600 root 0 linear /dev/sda5:0-25599
 /dev/sda5 zoneminder-vg lvm2 a-- 699.52g 598.57g 25600 141311 0 free
 /dev/sda5 zoneminder-vg lvm2 a-- 699.52g 598.57g 166911 244 swap_1 0 linear /dev/sda5:166911-167154
 /dev/sda5 zoneminder-vg lvm2 a-- 699.52g 598.57g 167155 11922 0 free

root@zoneminder:~# sudo pvmove --alloc anywhere /dev/sda5:166911-167154 /dev/sda5:25601-25845
 /dev/sda5: Moved: 0.4%
 /dev/sda5: Moved: 100.0%

root@zoneminder:~# pvresize --setphysicalvolume 101G -v /dev/sda5
 Using physical volume(s) on command line.
 Archiving volume group "zoneminder-vg" metadata (seqno 12).
 /dev/sda5: Pretending size is 211812352 not 1467002880 sectors.
 Resizing volume "/dev/sda5" to 211812352 sectors.
 Resizing physical volume /dev/sda5 from 0 to 25855 extents.
 Updating physical volume "/dev/sda5"
 Creating volume group backup "/etc/lvm/backup/zoneminder-vg" (seqno 13).
 Physical volume "/dev/sda5" changed
 1 physical volume(s) resized / 0 physical volume(s) not resized

root@zoneminder:~# pvs -v --segments /dev/sda5
 Using physical volume(s) on command line.
 Wiping cache of LVM-capable devices
 PV VG Fmt Attr PSize PFree Start SSize LV Start Type PE Ranges
 /dev/sda5 zoneminder-vg lvm2 a-- 101.00g 44.00m 0 25600 root 0 linear /dev/sda5:0-25599
 /dev/sda5 zoneminder-vg lvm2 a-- 101.00g 44.00m 25600 1 0 free
 /dev/sda5 zoneminder-vg lvm2 a-- 101.00g 44.00m 25601 244 swap_1 0 linear /dev/sda5:25601-25844
 /dev/sda5 zoneminder-vg lvm2 a-- 101.00g 44.00m 25845 10 0 free

Actually what I'm trying to achieve is the following (post #18):
https://communities.vmware.com/message/2723540#2723540

I'm now stuck at that /dev/sda3 has not been removed, still seeing in fdisk -l and the size of /dev/sda5 is still 700G.

root@zoneminder:~# fdisk -l
Disk /dev/sda: 1.6 TiB, 1717986918400 bytes, 3355443200 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
Disklabel type: dos
Disk identifier: 0x21880f4a

Device Boot Start End Sectors Size Id Type
/dev/sda1 * 2048 999423 997376 487M 83 Linux
/dev/sda2 1001470 1468004351 1467002882 699.5G 5 Extended
/dev/sda3 1468004352 3355443199 1887438848 900G 8e Linux LVM
/dev/sda5 1001472 1468004351 1467002880 699.5G 8e Linux LVM

Partition table entries are not in disk order.


Disk /dev/mapper/zoneminder--vg-root: 100 GiB, 107374182400 bytes, 209715200 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/zoneminder--vg-swap_1: 976 MiB, 1023410176 bytes, 1998848 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

UPDATE #2:

I've managed via an LIVE-CD to use GPARTED and PARTED to remove the disk and resize the /dev/sda2 wherein also /dev/sda5 is with the LV(M)

Because of the snapshots in ESXi I've made, I need to redo all the steps again. Because the nesteling of the vmdk's are a bit nasty as it seems to be.

It takes a while untill the orignal VM has been copied (1,6TB). After this I will do a full dump of all the steps such that in the future, someone else can use it also.

Great, you've successfully shrunk the filesystem inside the root LV.

The next step is to shrink the LV to match the new size of the filesystem. For the sake of safety, you may want to leave a bit of slack, as accidentally cutting off too much would be a far worse problem.

Filesystem shrinking operations are always bit more risky than extensions, especially if you are not yet quite familiar with them. So backup anything you might need, just in case something goes wrong.

Then use tune2fs -l to get the exact block count from the filesystem:

tune2fs -l /dev/mapper/zoneminder--vg-root | grep "Block "
Block count: NNNNNNNNN
Block size: XXXX

Multiply these two numbers together to get the exact size of the filesystem, then divide by 1024 to get binary kilobytes, and again to get binary megabytes. Add one to protect against rounding errors:

expr NNNNNNNNN * XXXX / 1024 / 1024 + 1
SSSSSS

Now, shrink the LV:

lvreduce -L SSSSSS /dev/mapper/zoneminder--vg-root

Now you should have plenty of free space in your Zoneminder VG. Use the pvs to confirm whether /dev/sda3 is now completely unused or not:

If, in the pvs output, the PFree value is not equal to PSize for /dev/sda3, there are still some parts of the root LV on that PV, and you'll need to move them out of there. pvmove can easily do that. If /dev/sda3 is now completely fbree, you can skip this step.

pvmove /dev/sda3

This essentially says "make sda3 empty by moving all the LV data that's still in ot to other PVs belonging to this same VG."

pvmove works by mirroring a piece of the data to be moved to its new location, then "removing the mirror" from the old side. So if pvmove gets interrupted by a system crash, it's not catastrophic. Just run pvmove with no parameters to continue from where it was.

Now, the sda3 PV should be completely empty. Remove it from the VG:

vgreduce zoneminder-vg /dev/sda3

At this point, /dev/sda3 will be an unattached, completely free LVM PV. You can wipe the PVID from it if you wish:

pvremove /dev/sda3

Now you'll be free to reuse the /dev/sda3 partition any way you like.
(If you plan to do something that causes the partition to be overwritten anyway, the pvremove command won't be strictly necessary.)

Now, if you want to extend the root LV to 100 GiB, here are the steps:

lvextend -L 100G /dev/mapper/zoneminder--vg-root
resize2fs /dev/mapper/zoneminder--vg-root

And you're done.

Note that I didn't say "unmount the filesystem" or "reboot the system" at any point here. It isn't necessary.

The LVM system has no idea of how you use your logical volumes. Even if you reduces the filesystem that you use in your LV, it does not change the size of the LV, nor the free space in your VG and PV (all space allocated to a LV is considered in use).

If you reduced you filesystem, you can reduce your LV with the lvreduce command, but you should be extra careful not to reduce it to small to fit your filesystem, or you may lose some data.

Once you have reduced your logical volume, you may move the used space from one PV to another one, with the pvmove command, and then remove the PV from the VG with the vgreduce command.