LVM
Components of LVM
- Volume group (VG)
- Physical Volume (PV)
- Logical volume (LV)
- Physical Partition (PP)
- Logical Partition (LP)
- Volume Group Descriptor Area (VGDA)
- Volume Group Status Area (VGSA)
- Logical Volume Control Block (LVCB)
- File system (FS)
- Mount Point
Maximum # PP/disk à 1016
VG Type
VG
type
|
Maximum
PVs
|
Maximum
LVs
|
Maximum
PPs per VG
|
Maximum
PP size
|
Normal VG
|
32
|
256
|
32,512 (1016 * 32)
|
1 GB
|
Big VG
|
128
|
512
|
130,048 (1016 * 128)
|
1 GB
|
Scalable VG
|
1024
|
4096
|
2,097,152
|
128 GB
|
Default PP for big vg :- 32MB
Default max LV :- 256
Pbuf == control the pending disk IO(primary
memory buffer)
# lvmo -a
vgname = rootvg
pv_pbuf_count = 512 ==> buffer value of per pv in vg
total_vg_pbufs = 1024
max_vg_pbufs = 16384
pervg_blocked_io_count
= 4997572
pv_min_pbuf = 512
max_vg_pbuf_count = 0
global_blocked_io_count
= 5325214
#
LTG(Logical track
group)==> variable stands Maximum an IO request for LV
To check the value of
LTG
# lquerypv -M hdisk0
256
#
Or
#lspv hdisk0
PHYSICAL VOLUME: hdisk0 VOLUME GROUP: rootvg
PV IDENTIFIER: 0002cb8982c3200a VG IDENTIFIER 0002cb890000d6000000011c4ef65807
PV STATE: active
STALE PARTITIONS: 0 ALLOCATABLE: yes
PP SIZE: 256 megabyte(s) LOGICAL VOLUMES: 13
TOTAL PPs: 546 (139776 megabytes) VG DESCRIPTORS: 2
FREE PPs: 414 (105984 megabytes) HOT SPARE: no
USED PPs: 132 (33792 megabytes) MAX REQUEST: 256
kilobytes
FREE DISTRIBUTION: 109..67..20..109..109
USED
DISTRIBUTION: 01..42..89..00..00
MIRROR POOL: None
#
Or
#lsvg rootvg
VOLUME GROUP: rootvg VG IDENTIFIER: 0002cb890000d6000000011c4ef65807
VG STATE: active PP SIZE: 256 megabyte(s)
VG PERMISSION: read/write TOTAL PPs: 1092 (279552 megabytes)
MAX LVs: 256 FREE PPs: 840 (215040 megabytes)
LVs: 13 USED PPs: 252 (64512 megabytes)
OPEN LVs: 12 QUORUM: 1 (Disabled)
TOTAL PVs: 2 VG DESCRIPTORS: 3
STALE PVs: 0 STALE PPs: 0
ACTIVE PVs: 2
AUTO ON: yes
MAX PPs per VG: 32512
MAX PPs per PV: 1016 MAX PVs: 32
LTG size (Dynamic): 256 kilobyte(s) AUTO SYNC: no
HOT SPARE: no BB POLICY: relocatable
PV RESTRICTION: none
#
To change the LTG
value using below command for a vg
#varyonvg –M512K
testvg
GLVM (Geographic
Logical Volume Manger)
It extend the LVM
mirroring across the Site using TCP/IP networking.
Creating Normal VG
# mkvg -s 512 -v 1024 -y newvg hdisk2
Creating Big VG
# mkvg –B –s 512 -y newvg hdisk2
Creating the
Scalable VG
# mkvg –S –s 1024 –y
newvg hdisk2
Creating the
concurrent VG for HACMP
It has to create in
HACMP ES using –C option it will support only enhanced concurrent vg only
support 64bit kernel, concurrent will support 32 bit kernel
Types of LV
1. Jfs
2. jfslog
3. jfs2
4. jfs2log
5. paging
6. boot
7. sysdump
8. aio_cache
Default type ==> jfs
Default max lp ==> 512
Default scheduling
policy ==> parallel
Default position ==> middle
Default rang of PV ==> minimum
Default stripe value ==> 128M
Create jfs2 LV
# mklv –t jfs2 –x 1000
–y newlv newvg 600
#chlv –x 2000 testlv ==> To increase the max pp size in lv
# rmlv testlv
File System Type
- JFS
- Enhanced JFS or JFS2
- CD ROM FS
- RAM File system
- NFS
- Name File system(NameFS)
- General Parallel File system(GPFS)
Difference between JFS and JFS2 FS
Function
|
JFS
|
Enhanced JFS
|
Optimization
|
32-bit kernel
|
64-bit kernel
|
Maximum file system size
|
1 terabyte
|
4 petabytes
Note: This
is an architectural limit. AIX® currently only supports up to 16 terabytes.
|
Maximum file size
|
64 gigabytes
|
4 petabytes
Note: This
is an architectural limit. AIX currently only supports up to 16 terabytes.
|
Number of I-nodes
|
Fixed at file system creation
|
Dynamic, limited by disk space
|
Large file support
|
As mount option
|
Default
|
Online defragmentation
|
Yes
|
Yes
|
namefs
|
Yes
|
Yes
|
DMAPI
|
No
|
Yes
|
Compression
|
Yes
|
No
|
Quotas
|
Yes
|
Yes
|
Deferred update
|
Yes
|
No
|
Directory structure
|
Linear organization
|
Binary tree representation
|
·
The main advantage of using
Enhanced JFS over JFS is scaling.
·
Journaling for disable Meta
data logging using nonitergrity mount option in JFS it will give more
performance than JFS2.
Create jfs2 file
system with already created lv
# crfs -v jfs2 -d
testlv -A yes -m /testfs
Create jfs2 file
sytem without existing lv
# crfs -v jfs2 -g
itlogsvg -m /test1fs -a size=1G
Create jfs2 with
NFS4 ACL type
# crfs –v jfs2 –g
itlogvg –m /test –a size=10G –a ea=v2
Attributes:-
Agblksize=512, 1024,
2048, 4096 ==> 4096 is default.
Cmtype=gc or ac ==> gc reduce the data loss during
system crash, ac will give better performance vice versa
Logname=logLVname ==>
It can be INLINE its
default is 0.4% of FS size. And its cannot be more then 2047MB or 10% of FS.
Options=mountoptions
Size=
fs block size
(byte) MAX fssize (TB)
===========================================
512 4
1024 8
2048 16
4096 32
ea= (v1 or v2) ==> v1 is default and compatible with
prior version of AIX.
v2 is support of scalable named as well as
support for NFS4 ACLs.
# rmfs /test1fs à it will also remove the LV
#extendvg testvg
hdisk10 ==> To increase the vg
#chfs –a size=+10G
/testfs ==> To adding the space (-) for reduce
Mount
# mount –a or # mount
all ==> To mount all file systems
# umount /test or
unmounts /test ==> Te unmounts the file system
# mount host1:/test
/test1 ==> to mount nfs file system
# mount –v cdrfs –o ro
/dev/cd0 /mnt ==> mount the cd rom
# umount –f /test ==> force unmounts
# fuser –kuc /test ==> to check the FS using process
# fuser –kucx /test ==> to kill the Fs using process
# mount –o ro /test ==> mount as read only
Mirroring rootvg
- # extendvg rootvg hdisk1
- # mirrorvg rootvg
- # bootlist –o –m normal ==> to check the boot order
- # bootlist –m normal hdisk0 hdisk1 ==> recreate the boot order
- # bosboot –ad hdisk0
- # bosboot –ad hdisk1
Replace the failed HDD in Mirrored rootvg
- # unmirrorvg rootvg hdisk1
- # mirgrate hdisk1 hdisk0 ==> if having any issues
- # reducevg rootvg hdisk1
- # rmdev –Rdl hdisk1
- # cfgmgr
- # cfgmgr -v ==> Afger replacing the disk
- # lspv
- Steps of mirroring the rootvg
Creating lv copy
- Can make the Copies with the VG only.
- Max no.of copies 3
# lslv –m lv01 ==> to check the existing copies of lv
# mklvcopy –k lv01 2
hdisk3 ==> to create the 2 copies and sync
# rmlvcopy lv01 2 ==> to remove the lvcopy
# lquerypv –M hdisk0
(or) # lspv hdisk0 ==> To show the LTG size.
# sync –v datavg ==> to sysn the steal PP
RAM Disk
Default block size is
512 bytes
Max ramdisk 64
To crate the jfs2
ramdisk
# mkramdisk 2G
# /sbin/helpers/jfs2/mkfs
–V jfs2 /dev/ramdisk0
# mkdir /ramdisk0
# mount –V jfs2 –o
log=NULL /dev/ramdisk0 /ramdisk0
# rmramdisk ramdisk0 ==> To remove the ramdisk
# ls –l /dev|grep ram ==> To check the ramdisk
Dump device
# sysdumpdev –e ==> To check the estimated size of dump
# sysdumpdev –l ==> To check the dump dev status
# mklv –y dumplv –t
sysdump rootvg 5 ==> to create the sysdump
# sysdump –p
/dev/dumplv ==> To change the primary dump
# sysdump –s
/dev/sysdumpnull ==> To change the secondary dump
# sysdump –k ==> To change the always allow dump to
TRUE
# sysdump –Pp /dev/lg_dumplv ==> To change permanently
# rmlv dumplv ==> To remove system dump
Renaming FS/LV
# umount /testfs ==> Un mount FS
# chfs –m testfs newfs ==> Change the FS name
# chlv -n newlv oldlv ==> To change the LV name
LVM Management
# defragfs /home ==> To defrag the FS
# lsvg –o ==> list the active vg
# lsvg –l rootvg ==> list the lv in vg and mirror numbers
# chvg –S datavg ==> changing Vg to scalable
# exportvg testvg ==> To remove or export the VG
# importvg –y newvg
hdisk2 ==> import or import with new
name(rename)
Rootvg LV names à hd1-/home, hd2-/usr, hd3-/tmp, hd4-/
, hd5-boot, hd6-pageing, hd7(lg_dumplv)-dumpdev, hd8-jfs2log, hd9var-/var,
hd10opt-/opt.
