tar command is useful to combine 1 or more files/directories into a single file. It is good for restore/backup purpose. There are some compress utilities which compress the files to save space on the filesystem.
1. Create a tar file from your current directory ..
# tar -cvf /var/tmp/mytar.tar mytar/ # Relative path
[ # tar -cvf /var/tmp/mytar.tar /home/mytar/ #Absolute path. ]
2. View the content of the tar file.
# tar -tvf mytar.tar
3. Extract the content of the tar file.
# tar -xvf mytar.tar
If you want to compress,
# bzip2 -c mytar.tar > mytar.tar.bz2 ==>> creates a new file mytar.tar.bz2
# bzip2 mytar.tar ==>> Compress the original file and renames with .bz2
# gzip -c mytar.tar > mytar.tar.gz ==>> creates a new file mytar.tar.gz
# gzip myfile ==>> Compress the original file and renames with .gz
# compress myfile # Compresses the file with .Z extension.
4. To extract the tar file
# bzip2 -d mytar.tar.bz2
# gzip -d mytar.tar.gz
# gunzip mytar.tar.gz
# uncompress mytar.tar.Z
and
# tar -xvf mytar.tar
5. Reading/viewing the Compress file.
# bzcat mytar.tar.bz2
# zcat mytar.tar.gz
# tar -tvf mytar.tar
Friday, November 25, 2011
setuid and setgid and sticky bit on Unix/Linux
setuid and setgid and sticky bit on Unix/Linux
When a file had setuid permission set, a user can executes a file or a program being (effective user id of the owner) owner of the file.
When a file has setgid permission set, a user can executes a file or a program being (effective group id of ) the group owner of the program.
You can use setgid on directories as well. When a user creates files on a setgid directory, the group ownership to set to the group owner of the directory. (Note the owner permission for S and s for execute bit)
1. Working with setuid files.
$ ls -l
-rw-r--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
a. myfile does not have setuid set up. To enable setuid
$ chmod u+s myfile
$ ls -l
-rwSr--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
$ chmod u+x myfile # Note: adding execute bit to the owner.
$ ls -l
-rwsr--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
or
$ chmod 4754 myfile
$ ls -l
-rwsr-xr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
Note: You can use 4 instead to u+s with chmod command but you have to use other permission as well.
b. To disable,
$ chmod u-s myfile
$ ls -l
-rwxr--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
---------------------------------------------
2. Working with setgid (Note the group ownership on execute bit for s and S, why?)
a. To enable setgid,
$ chmod g+s myfile
$ ls -l
-rwxr-Sr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
$ chmod g+x myfile # Adding execute permission to the group.
$ ls -l
-rwxr-sr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
b. To disable,
$ chmod g-s myfile
$ ls -l
-rwxr-xr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
$
or use number 2 instead to g+s.
$ chmod 2754 myfile
$ ls -l
-rwxr-sr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
Note: You can use 2 with chmod command to set GID but remember other permissions as well.
---------------------------------------------
3. Working with sticky bit. (Note the execute permission on others for T and t).
Sticky bit is set to the shared directories where public need to have read, write permission such as /tmp, /var/tmp. If you give permissions without sticky bit set on, any user can edit/remove any files created by any user. So, with sticky bit set on the publicly writable directory, user can create file and also execute other files created by other users as well but they can not remove or change it. Only the owner or the root can make changes or delete the file or directories inside sticky bit set directories.
If you encounter any issues with the application failed to load, please make sure that you have proper permission set on /var/tmp and /tmp directories.
# ls -ld /tmp /var/tmp
drwxrwxrwt 19 root root 159288 Dec 1 08:32 /tmp
drwxrwxrwt 12 root sys 1536 Nov 30 19:01 /var/tmp
Note the execute permission for others, you will see t value. If you see upper case T that means execute permission on the directory for others is not enabled. lower case t states that execute permission is placed for the directory.
To set sticky bit
# chmod o+t /tmp
if there is not execute bit set for others then there will be S on execute bit place.
To add execute bit to the directory,
# chmod o+x /tmp
You will see s (lower case now) on others execute permission on the directory.
or
# chmod o=rwxt /tmp
or
# chmod 1777 /tmp
---------------------------------------------
4. setuid/setgid/sticky bit setting together.
# chmod 4755 myfile.sh # always execute the script as the user or owner.
# chmod 2755 /mydir # When created a file, it will inherit the group permission.
# chmod 1777 /tmp # setting sticky bit on the directory.
Note: If you want to apply set uid + gid + sticky bit, you can add the numeric value together. such as,
#chmod 3777 /tmp
here, note the first numeric value of 3 (2+1) which comes off gid + sticky bit set on the directory.
When a file had setuid permission set, a user can executes a file or a program being (effective user id of the owner) owner of the file.
When a file has setgid permission set, a user can executes a file or a program being (effective group id of ) the group owner of the program.
You can use setgid on directories as well. When a user creates files on a setgid directory, the group ownership to set to the group owner of the directory. (Note the owner permission for S and s for execute bit)
1. Working with setuid files.
$ ls -l
-rw-r--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
a. myfile does not have setuid set up. To enable setuid
$ chmod u+s myfile
$ ls -l
-rwSr--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
$ chmod u+x myfile # Note: adding execute bit to the owner.
$ ls -l
-rwsr--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
or
$ chmod 4754 myfile
$ ls -l
-rwsr-xr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
Note: You can use 4 instead to u+s with chmod command but you have to use other permission as well.
b. To disable,
$ chmod u-s myfile
$ ls -l
-rwxr--r-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
---------------------------------------------
2. Working with setgid (Note the group ownership on execute bit for s and S, why?)
a. To enable setgid,
$ chmod g+s myfile
$ ls -l
-rwxr-Sr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
$ chmod g+x myfile # Adding execute permission to the group.
$ ls -l
-rwxr-sr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
b. To disable,
$ chmod g-s myfile
$ ls -l
-rwxr-xr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
$
or use number 2 instead to g+s.
$ chmod 2754 myfile
$ ls -l
-rwxr-sr-- 1 kbhusal kbhusal 13 Nov 25 12:55 myfile
Note: You can use 2 with chmod command to set GID but remember other permissions as well.
---------------------------------------------
3. Working with sticky bit. (Note the execute permission on others for T and t).
Sticky bit is set to the shared directories where public need to have read, write permission such as /tmp, /var/tmp. If you give permissions without sticky bit set on, any user can edit/remove any files created by any user. So, with sticky bit set on the publicly writable directory, user can create file and also execute other files created by other users as well but they can not remove or change it. Only the owner or the root can make changes or delete the file or directories inside sticky bit set directories.
If you encounter any issues with the application failed to load, please make sure that you have proper permission set on /var/tmp and /tmp directories.
# ls -ld /tmp /var/tmp
drwxrwxrwt 19 root root 159288 Dec 1 08:32 /tmp
drwxrwxrwt 12 root sys 1536 Nov 30 19:01 /var/tmp
Note the execute permission for others, you will see t value. If you see upper case T that means execute permission on the directory for others is not enabled. lower case t states that execute permission is placed for the directory.
To set sticky bit
# chmod o+t /tmp
if there is not execute bit set for others then there will be S on execute bit place.
To add execute bit to the directory,
# chmod o+x /tmp
You will see s (lower case now) on others execute permission on the directory.
or
# chmod o=rwxt /tmp
or
# chmod 1777 /tmp
---------------------------------------------
4. setuid/setgid/sticky bit setting together.
# chmod 4755 myfile.sh # always execute the script as the user or owner.
# chmod 2755 /mydir # When created a file, it will inherit the group permission.
# chmod 1777 /tmp # setting sticky bit on the directory.
Note: If you want to apply set uid + gid + sticky bit, you can add the numeric value together. such as,
#chmod 3777 /tmp
here, note the first numeric value of 3 (2+1) which comes off gid + sticky bit set on the directory.
How to limit unauthorized access to grub.conf file.
If you want to protect illegal access to grub.conf file and unauthorized password change or reboot, you can protect the grub.conf file with encrypted password. Use the encrypted password generated by grub-md5-crypt.
1. Generate md5 password.
# /sbin/grub-md5-crypt
Password:
Retype password:
$1$7CzLO0$r4wIx9cb2TN3aTiPvjwaH1
[root@durava ~]#
2. Add md5 password just generated to the grub.conf file right after splashimage entry.
password --md5 $1$7CzLO0$r4wIx9cb2TN3aTiPvjwaH1
==========================================
# cat /boot/grub/grub.conf
# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE: You do not have a /boot partition. This means that
# all kernel and initrd paths are relative to /, eg.
# root (hd0,0)
# kernel /boot/vmlinuz-version ro root=/dev/sda1
# initrd /boot/initrd-version.img
#boot=/dev/sda
default=0
timeout=5
splashimage=(hd0,0)/boot/grub/splash.xpm.gz
hiddenmenu
password --md5 $1$7CzLO0$r4wIx9cb2TN3aTiPvjwaH1
title Red Hat Enterprise Linux Server (2.6.18-238.9.1.el5)
root (hd0,0)
kernel /vmlinuz-2.6.18-238.9.1.el5 ro root=LABEL=/ elevator=deadline audit=1
initrd /initrd-2.6.18-238.9.1.el5.img
title Red Hat Enterprise Linux Server (2.6.18-238.5.1.el5)
root (hd0,0)
kernel /vmlinuz-2.6.18-238.5.1.el5 ro root=LABEL=/ audit=1
initrd /initrd-2.6.18-238.5.1.el5.img
title Red Hat Enterprise Linux Server (2.6.18-194.26.1.el5)
root (hd0,0)
kernel /boot/vmlinuz-2.6.18-194.26.1.el5 ro root=LABEL=/ audit=1
initrd /boot/initrd-2.6.18-194.26.1.el5.img
#
1. Generate md5 password.
# /sbin/grub-md5-crypt
Password:
Retype password:
$1$7CzLO0$r4wIx9cb2TN3aTiPvjwaH1
[root@durava ~]#
2. Add md5 password just generated to the grub.conf file right after splashimage entry.
password --md5 $1$7CzLO0$r4wIx9cb2TN3aTiPvjwaH1
==========================================
# cat /boot/grub/grub.conf
# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE: You do not have a /boot partition. This means that
# all kernel and initrd paths are relative to /, eg.
# root (hd0,0)
# kernel /boot/vmlinuz-version ro root=/dev/sda1
# initrd /boot/initrd-version.img
#boot=/dev/sda
default=0
timeout=5
splashimage=(hd0,0)/boot/grub/splash.xpm.gz
hiddenmenu
password --md5 $1$7CzLO0$r4wIx9cb2TN3aTiPvjwaH1
title Red Hat Enterprise Linux Server (2.6.18-238.9.1.el5)
root (hd0,0)
kernel /vmlinuz-2.6.18-238.9.1.el5 ro root=LABEL=/ elevator=deadline audit=1
initrd /initrd-2.6.18-238.9.1.el5.img
title Red Hat Enterprise Linux Server (2.6.18-238.5.1.el5)
root (hd0,0)
kernel /vmlinuz-2.6.18-238.5.1.el5 ro root=LABEL=/ audit=1
initrd /initrd-2.6.18-238.5.1.el5.img
title Red Hat Enterprise Linux Server (2.6.18-194.26.1.el5)
root (hd0,0)
kernel /boot/vmlinuz-2.6.18-194.26.1.el5 ro root=LABEL=/ audit=1
initrd /boot/initrd-2.6.18-194.26.1.el5.img
#
Tuesday, November 22, 2011
Find if the kernel is 32 or 64 bit.
Find whether the kernel is 32 or 64 bit.
On Redhat run the following commands.
$getconf LONG_BIT
64
$ uname -m
x86_64
$ uname -m
i686
$ file /bin/ls
/bin/ls: ELF 32-bit
The command below gives you if you are running 64 or 32 bit platform.
$ cat /proc/cpuinfo
32
On Solaris
$ isainfo -v
On Redhat run the following commands.
$getconf LONG_BIT
64
$ uname -m
x86_64
$ uname -m
i686
$ file /bin/ls
/bin/ls: ELF 32-bit
The command below gives you if you are running 64 or 32 bit platform.
$ cat /proc/cpuinfo
32
On Solaris
$ isainfo -v
Wednesday, November 16, 2011
step by step guide to Redhat KickStart Installation
step by step guide to Redhat KickStart Installation
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kick start is a hand free installation of Redhat server on multiple machine.
Plan:
- Any successfully installed redhat system creates /root/anaconda-ks.cfg file. Based on this file you can add or remove packages.
- Or use the command System-Config-Kickstart to create a new ks file based on your requirement.
- Define your OS location either using ftp, nfs, or http
- Start the kickstart installation.
================================================
detail to follow,
===============================================
Before setting up your kickstart environment, first you better set up yum repo so that you can install all required softwares.
A. a. To set up your yum, just create the following file with the info below.
# createrepo /opt/RHL6.1/Packages
[jay@sama ~]$ cat /etc/yum.repos.d/file.repo
[RHEL-Repository]
name=RHEL repository
baseurl=file:///opt/RHL6.1/Packages
enabled=1
gpgcheck=0
b. Set up nfs server.
[root@sama ~]# yum install nfs*
[jay@sama ~]$ cat /etc/exports
#/opt/RHL6.1/Packages *(rw,sync)
/opt/RHL6.1 *(rw,sync)
[root@sama ~]# service nfs reload
c. set up web server.
[root@sama ~]# yum -y install httpd
[root@sama ~]# service httpd start
[root@sama html]# vi index.html
<html>
<title>This is a test page</title>
<body>
This is a test page
</body>
</html>
confirm that you can access your website,
type http://192.168.10.110/ on your browser and press enter, you should be able to see "This is a test page page".
B. Now, you can copy your OS DVD copy on your web server.
# mkdir /OS; mount -o loop OS_Image.iso /OS; cd /OS
# cp -a * /var/www/html/yum/
verify that you can access http://192.168.10.110/yum
make sure that you can access the site from other systems on the network. if it failed, disable firewall (iptables).
# /etc/init.d/iptables stop
C. Now, real fun begins, Create a kickstart file
a. you can use anaconda-ks.cfg file as a template that is created during the first time installation of the Linux OS. Modify the values based on your requirement or
the environment.
b. Create kickstart config file using the system-config-kickstart command utility and select different parameters based on your environment. Note: make sure to instal
kickstart config package to use this feature.
here is the sample file,
[root@sama html]# cat /var/www/html/ks.cfg
# Kickstart file automatically generated by anaconda.
#version=DEVEL
install
#cdrom
nfs --server=192.168.10.110 --dir=/opt/RHL6.1
lang en_US.UTF-8
keyboard us
#network --onboot no --device eth0 --noipv4 --noipv6
network --device eth0 --bootproto static --ip 192.168.10.150 --netmask 255.255.255.0 --gateway 192.168.10.110 --hostname kickRH6.1.local
rootpw --iscrypted $6$dsXNrQw2LZLmKFLZ$55Pxwi.6bJhI3QCxembF5lRD9hrjd15b5wx3caJZVcxQC8yTEc0cz2GyVeR5s9Ao4ZxnDvFyDLWBVy9Oi2SGC0
firewall --service=ssh
authconfig --enableshadow --passalgo=sha512
selinux --enforcing
timezone --utc America/New_York
bootloader --location=mbr --driveorder=sda --append="crashkernel=auto rhgb quiet"
# The following is the partition information you requested
# Note that any partitions you deleted are not expressed
# here so unless you clear all partitions first, this is
# not guaranteed to work
clearpart --linux --drives=sda
#part --onpart=sda1 --format
part /boot --fstype=ext4 --size=500
part pv.008003 --grow --size=1
volgroup vg_kickRH6.1 --pesize=4096 pv.008003
logvol / --fstype=ext4 --name=lv_root --vgname=vg_kickRH6.1 --grow --size=1024 --maxsize=51200
logvol swap --name=lv_swap --vgname=vg_kickRH6.1 --grow --size=2032 --maxsize=4064
repo --name="Red Hat Enterprise Linux" --baseurl=http://192.168.10.110/yum --cost=100
%packages
@base
@client-mgmt-tools
@console-internet
@core
@debugging
@basic-desktop
@desktop-platform
@directory-client
@fonts
@general-desktop
@graphical-admin-tools
@hardware-monitoring
@internet-browser
@java-platform
@kde-desktop
@large-systems
@mysql-client
@mysql
@nepali-support
@network-file-system-client
@performance
@perl-runtime
@postgresql-client
@postgresql
@remote-desktop-clients
@server-platform
@server-policy
@system-admin-tools
@web-server
@x11
mtools
pax
python-dmidecode
oddjob
sgpio
genisoimage
wodim
qt-mysql
certmonger
pam_krb5
krb5-workstation
perl-DBD-MySQL
perl-DBD-SQLite
rdesktop
crypto-utils
certmonger
perl-CGI
%end
%post
(
# Disable some services and create normal user with encripted password.
chkconfig --level 3 ip6tables off
useradd -p '$6$OY1GbBYr
$MJI1dgQK23CBRzNyAEWGupxMRd2Hhwovr0cID6CtSgX7TSnLpPTR0rYNJ.AgshZzyK6QhPSWQZ0ifP9ky8HFl.' jay
) 2>&1 | tee /root/post-install.log
echo "Installation completed !!"
echo "All unauthorized activities will be monitored">>/etc/issue
%end
D. validate the kickstart configuration file with ksvalidator command
# ksvalidator /var/www/html/ks.cfg
Note: If there is an mistake or typos, you will get an error.
E. Confirm the configuration file is available form the browser.
Either type the full path on the browser or use elinks to verify.
[root@sama html]# elinks 192.168.10.110/ks.cfg
*Make sure ks.cfg file has at least 644 permission.
Now you kickstart configuration is available for installer using the web.
Now, use your installer, if you are using redhat use boot.iso, centos-->> CentOS*netinstall.iso, sciencefic linux SL*boot.iso file to install. You can use any iso file
to install any OS type but make sure to use right 32 or 64 bit image.
F. When you boot the system using any boot image, at the grub menu list, press the tab and type the following values at the prompt
linux ks=http://192.168.10.110/ks.cfg
press enter and you are good to go. You should have system installed with a normal user which you can use to login.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The kickstart file has 4 sections:
pre, configuration, packages and post.
The pre section is generally used to
check for certain hardware types or configurations and act upon that data. The
pre is not used in this scenario.
The configuration, sets up the network,
video, hard disk, locale, etc.
The packages lists which packages to
install or (with a ‘-‘) not install.
The post is how all the third party
software is added, the system is customized, and environment is setup.
The post section executes immediately
after loading the OS packages, but is finished before the OS reboots for the
first time. Because there are some
things which do not load properly until the full OS is booted, a finishing
script is placed on the system to run on first boot.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~Kick start is a hand free installation of Redhat server on multiple machine.
Plan:
- Any successfully installed redhat system creates /root/anaconda-ks.cfg file. Based on this file you can add or remove packages.
- Or use the command System-Config-Kickstart to create a new ks file based on your requirement.
- Define your OS location either using ftp, nfs, or http
- Start the kickstart installation.
================================================
detail to follow,
===============================================
Before setting up your kickstart environment, first you better set up yum repo so that you can install all required softwares.
A. a. To set up your yum, just create the following file with the info below.
# createrepo /opt/RHL6.1/Packages
[jay@sama ~]$ cat /etc/yum.repos.d/file.repo
[RHEL-Repository]
name=RHEL repository
baseurl=file:///opt/RHL6.1/Packages
enabled=1
gpgcheck=0
b. Set up nfs server.
[root@sama ~]# yum install nfs*
[jay@sama ~]$ cat /etc/exports
#/opt/RHL6.1/Packages *(rw,sync)
/opt/RHL6.1 *(rw,sync)
[root@sama ~]# service nfs reload
c. set up web server.
[root@sama ~]# yum -y install httpd
[root@sama ~]# service httpd start
[root@sama html]# vi index.html
<html>
<title>This is a test page</title>
<body>
This is a test page
</body>
</html>
confirm that you can access your website,
type http://192.168.10.110/ on your browser and press enter, you should be able to see "This is a test page page".
B. Now, you can copy your OS DVD copy on your web server.
# mkdir /OS; mount -o loop OS_Image.iso /OS; cd /OS
# cp -a * /var/www/html/yum/
verify that you can access http://192.168.10.110/yum
make sure that you can access the site from other systems on the network. if it failed, disable firewall (iptables).
# /etc/init.d/iptables stop
C. Now, real fun begins, Create a kickstart file
a. you can use anaconda-ks.cfg file as a template that is created during the first time installation of the Linux OS. Modify the values based on your requirement or
the environment.
b. Create kickstart config file using the system-config-kickstart command utility and select different parameters based on your environment. Note: make sure to instal
kickstart config package to use this feature.
here is the sample file,
[root@sama html]# cat /var/www/html/ks.cfg
# Kickstart file automatically generated by anaconda.
#version=DEVEL
install
#cdrom
nfs --server=192.168.10.110 --dir=/opt/RHL6.1
lang en_US.UTF-8
keyboard us
#network --onboot no --device eth0 --noipv4 --noipv6
network --device eth0 --bootproto static --ip 192.168.10.150 --netmask 255.255.255.0 --gateway 192.168.10.110 --hostname kickRH6.1.local
rootpw --iscrypted $6$dsXNrQw2LZLmKFLZ$55Pxwi.6bJhI3QCxembF5lRD9hrjd15b5wx3caJZVcxQC8yTEc0cz2GyVeR5s9Ao4ZxnDvFyDLWBVy9Oi2SGC0
firewall --service=ssh
authconfig --enableshadow --passalgo=sha512
selinux --enforcing
timezone --utc America/New_York
bootloader --location=mbr --driveorder=sda --append="crashkernel=auto rhgb quiet"
# The following is the partition information you requested
# Note that any partitions you deleted are not expressed
# here so unless you clear all partitions first, this is
# not guaranteed to work
clearpart --linux --drives=sda
#part --onpart=sda1 --format
part /boot --fstype=ext4 --size=500
part pv.008003 --grow --size=1
volgroup vg_kickRH6.1 --pesize=4096 pv.008003
logvol / --fstype=ext4 --name=lv_root --vgname=vg_kickRH6.1 --grow --size=1024 --maxsize=51200
logvol swap --name=lv_swap --vgname=vg_kickRH6.1 --grow --size=2032 --maxsize=4064
repo --name="Red Hat Enterprise Linux" --baseurl=http://192.168.10.110/yum --cost=100
%packages
@base
@client-mgmt-tools
@console-internet
@core
@debugging
@basic-desktop
@desktop-platform
@directory-client
@fonts
@general-desktop
@graphical-admin-tools
@hardware-monitoring
@internet-browser
@java-platform
@kde-desktop
@large-systems
@mysql-client
@mysql
@nepali-support
@network-file-system-client
@performance
@perl-runtime
@postgresql-client
@postgresql
@remote-desktop-clients
@server-platform
@server-policy
@system-admin-tools
@web-server
@x11
mtools
pax
python-dmidecode
oddjob
sgpio
genisoimage
wodim
qt-mysql
certmonger
pam_krb5
krb5-workstation
perl-DBD-MySQL
perl-DBD-SQLite
rdesktop
crypto-utils
certmonger
perl-CGI
%end
%post
(
# Disable some services and create normal user with encripted password.
chkconfig --level 3 ip6tables off
useradd -p '$6$OY1GbBYr
$MJI1dgQK23CBRzNyAEWGupxMRd2Hhwovr0cID6CtSgX7TSnLpPTR0rYNJ.AgshZzyK6QhPSWQZ0ifP9ky8HFl.' jay
) 2>&1 | tee /root/post-install.log
echo "Installation completed !!"
echo "All unauthorized activities will be monitored">>/etc/issue
%end
D. validate the kickstart configuration file with ksvalidator command
# ksvalidator /var/www/html/ks.cfg
Note: If there is an mistake or typos, you will get an error.
E. Confirm the configuration file is available form the browser.
Either type the full path on the browser or use elinks to verify.
[root@sama html]# elinks 192.168.10.110/ks.cfg
*Make sure ks.cfg file has at least 644 permission.
Now you kickstart configuration is available for installer using the web.
Now, use your installer, if you are using redhat use boot.iso, centos-->> CentOS*netinstall.iso, sciencefic linux SL*boot.iso file to install. You can use any iso file
to install any OS type but make sure to use right 32 or 64 bit image.
F. When you boot the system using any boot image, at the grub menu list, press the tab and type the following values at the prompt
linux ks=http://192.168.10.110/ks.cfg
press enter and you are good to go. You should have system installed with a normal user which you can use to login.
boot options linux ks=ftp://myserver.com/ks.conf linux ks=http://domain.com/path/ks.conf linux ks=nfs:servername:/path/ks.conf
================================
for the post installation entry
--------------------------------
%post --nochroot ###### NFS mounts ###### mkdir /mnt/sysimage/opt/users mkdir /mnt/sysimage/home echo "192.168.10.110 sama sama.expanor.local >>/mnt/sysimage/etc/hosts" echo "sama:/home/users /opt/users nfs rw,bg,intr,soft 1 2" >> /mnt/sysimage/etc/fstab ##echo "d2:/vol/root/home /home nfs rw,bg,intr,soft 0 0" >> /mnt/sysimage/etc/fstab
some selinux info
======================================
List and identify SELinux file and process context.
View SELinux contexts of processes:
# ps -eZ, ps -axZ, ps -Zc <process name>, etc.
View SELinux contexts of files and directories:
# ls -Zd /path/to/dir/, ls -Z /path/to/file, etc.
View SELinux contexts of users:
id -Z
Policy context rules are stored in
/etc/selinux/targeted/context/files/file_contexts and
/etc/selinux/targeted/context/files/file_contexts.local
# semanage fcontext -[a|d|m] -f <ftype> -t <context> ‘<regex>’
e.g.: semanage fcontext -a -t virt_image_t “/virtstorage(/.*)?”
Restore default file contexts.
# restorecon -R -v /dir/ : note the last slash. -R = recursive (all child files and directories) -v = verbose.
Use boolean settings to modify system SELinux settings.
- Booleans are plain text files located in /selinux/booleans
- semanage boolean -l : List booleans with basic descriptions (very useful with grep)
- setsebool [-P] <boolean_name> : set SE boolean, -P to make permanent (survive reboot)
- Use the graphical tool: system-config-selinux
- Diagnose and address routine SELinux policy violations.
Many targeted services have specialised man pages dealing with SELinux configuration.
Display these pages with:
# man -k ‘_selinux’
Installing setroubleshoot-server sends SELinux error messages to /var/log/messages. These can be further parsed with sealert.
audit2why and audit2allow can be used to parse the messages in /var/log/audit/audit.log and explain why access was denied, and how to modify your configuration to allow it.
# semanage port -l : list SELinux port settings.
SELinux Packages and utilities
coreutils : Always installed. Provides some default elements of SELinux.
policycoreutils : Provides restorecon, secon, setfiles, et al.
libselinux-utils : Provides getenforce, setenforce, getsebool, setsebool, et al.
policycoreutils-gui : Provides system-config-selinux and sepolgen, et al.
policycoreutils-python : Provides semanage, audit2allow, audit2why, et al.
setroubleshoot : Provides seapplet
setroubleshoot-server : Provides sealert, sedispatch, setroubleshootd, et al.
==========================
SET ENFORCING AND PERMISSIVE MODES FOR SELINUX
Persistent change:
/etc/selinux/config:
SELINUX=enforcing|permissiveCurrent session:
Non persistent change:
setenforce 1|0|enforcing|permissive
VIEW SELINUX STATUS:
Sestatus
LIST AND IDENTIFY SELINUX FILE AND PROCESS CONTEXT
ls -lZ
ps -efZ
RESTORE DEFAULT FILE CONTEXTS
restorecon -R*file
USE BOOLEAN SETTINGS TO MODIFY SYSTEM SELINUX SETTINGS
View booleans:
getsebool -a | grep keyword
OR*semanage boolean -l | grep keyword
Change booleans:
setsebool -P boolean on|off
DIAGNOSE AND ADDRESS ROUTINE SELINUX POLICY VIOLATIONS
Diagnose:
/var/log/audit/audit.log
/var/log/messages
view service logs
sealert
Fix:
audit2allow
setsebool -P boolean on|off
========================
List and identify SELinux file and process context.
View SELinux contexts of processes:
# ps -eZ, ps -axZ, ps -Zc <process name>, etc.
View SELinux contexts of files and directories:
# ls -Zd /path/to/dir/, ls -Z /path/to/file, etc.
View SELinux contexts of users:
id -Z
Policy context rules are stored in
/etc/selinux/targeted/context/files/file_contexts and
/etc/selinux/targeted/context/files/file_contexts.local
# semanage fcontext -[a|d|m] -f <ftype> -t <context> ‘<regex>’
e.g.: semanage fcontext -a -t virt_image_t “/virtstorage(/.*)?”
Restore default file contexts.
# restorecon -R -v /dir/ : note the last slash. -R = recursive (all child files and directories) -v = verbose.
Use boolean settings to modify system SELinux settings.
- Booleans are plain text files located in /selinux/booleans
- semanage boolean -l : List booleans with basic descriptions (very useful with grep)
- setsebool [-P] <boolean_name> : set SE boolean, -P to make permanent (survive reboot)
- Use the graphical tool: system-config-selinux
- Diagnose and address routine SELinux policy violations.
Many targeted services have specialised man pages dealing with SELinux configuration.
Display these pages with:
# man -k ‘_selinux’
Installing setroubleshoot-server sends SELinux error messages to /var/log/messages. These can be further parsed with sealert.
audit2why and audit2allow can be used to parse the messages in /var/log/audit/audit.log and explain why access was denied, and how to modify your configuration to allow it.
# semanage port -l : list SELinux port settings.
SELinux Packages and utilities
coreutils : Always installed. Provides some default elements of SELinux.
policycoreutils : Provides restorecon, secon, setfiles, et al.
libselinux-utils : Provides getenforce, setenforce, getsebool, setsebool, et al.
policycoreutils-gui : Provides system-config-selinux and sepolgen, et al.
policycoreutils-python : Provides semanage, audit2allow, audit2why, et al.
setroubleshoot : Provides seapplet
setroubleshoot-server : Provides sealert, sedispatch, setroubleshootd, et al.
==========================
SET ENFORCING AND PERMISSIVE MODES FOR SELINUX
Persistent change:
/etc/selinux/config:
SELINUX=enforcing|permissiveCurrent session:
Non persistent change:
setenforce 1|0|enforcing|permissive
VIEW SELINUX STATUS:
Sestatus
LIST AND IDENTIFY SELINUX FILE AND PROCESS CONTEXT
ls -lZ
ps -efZ
RESTORE DEFAULT FILE CONTEXTS
restorecon -R*file
USE BOOLEAN SETTINGS TO MODIFY SYSTEM SELINUX SETTINGS
View booleans:
getsebool -a | grep keyword
OR*semanage boolean -l | grep keyword
Change booleans:
setsebool -P boolean on|off
DIAGNOSE AND ADDRESS ROUTINE SELINUX POLICY VIOLATIONS
Diagnose:
/var/log/audit/audit.log
/var/log/messages
view service logs
sealert
Fix:
audit2allow
setsebool -P boolean on|off
========================
Step by Step guide to Create and manage Logical Volume Manager in Redhat .
Step by Step guide to Create and manage Logical Volume Manager in Redhat Linux.
1. Creating an LVM Logical Volume on Three Disks (or more if you have available).
2. Creating a Striped Logical Volume
3. Splitting a Volume Group
4. Removing a Disk from a Logical Volume
1. Creating an LVM Logical Volume on Three Disks.
Plan:
logical volume name: new_logical_volume
Disks: /dev/sda1, /dev/sdb1, and /dev/sdc1
partition type: Linux LVM which is 8e.
Warning: Verify that disks does not contain any important data.
a. Creating the Physical Volumes
# pvcreate /dev/sda1 /dev/sdb1 /dev/sdc1
# pvdisplay # command to display physical volume detail.
Note: Each physical volume has a UUID.
b. Creating the Volume Group
# vgcreate new_vol_group /dev/sda1 /dev/sdb1 /dev/sdc1
# vgs or # Displays new volume group attributes
# vgdisplay volume_name
Note: volume group name is new_vol_group
c. Creating the Logical Volume
# lvcreate -L 2G -n new_logical_volume new_vol_group # Creates 2 GB Logical volume
# lvcreate -L 3GB -n mylvvol new_vol_group # Creates 3 GB Logical volume
# lvdisplay # Displays Logical Volume attributes.
Here, L flag is sued to size while n flag is for new logical volume name.
Note: creates 2gb of logical volume new_logical_volume from the volume group new_vol_group.
You can create multiple logical volumes on a single volume group.
d. Creating the File System using mkfs command.
# mkfs.gfs2 -p lock_nolock -j 1 /dev/new_vol_group/new_logical_volume or
# mkfs.ext4 /dev/new_vol_group/new_logical_volume
Note: You can use mke2fs, mkfs.ext3, mkfs.ext4 commands with different options.
e. mount the logical volume
# mount /dev/new_vol_group/new_logical_volume /mnt
f. Add entry to /etc/fstab to be mounted across the reboot.
g. Resize the logical volume to expand or shrink the filesystem.
# lvresize -L 5GB /dev/new_vol_group/new_logical_volume
# resize2fs /dev/new_vol_group/new_logical_volume 5G
# lvresize -L 3GB /dev/new_vol_group/new_logical_volume
# resize2fs /dev/new_vol_group/new_logical_volume 3G
Note: To shrink you have to unmount the filesystem. To resize with new storage space added, use resize2fs command. The filessytem must be ext3 and up to resize the volume.
# resize2fs /dev/new_vol_group/new_logical_volume 5G
h. Remove the logical volume
# lvremove /dev/new_vol_group/new_logical_volume
i. Rename Logical Volume
# lvrename new_vol_group new_logical_volume my_logical_volume
# lvresize -L 10GB /dev/new_vol_group/my_logical_volume
Note: Only renames logical Volume not the volume group.
j. Rename Volume Group.
# vgrename new_vol_group my_vol_group
Note: Once you rename Logical volume or the volume group, make sure to change your mount point and the /etc/fstab entry.
=================================================================
2. Creating a Striped Logical Volume
Plan:
LVM striped logical volume: striped_logical_volume
Disk for stripping: /dev/sda1, /dev/sdb1, and /dev/sdc1
Warning: Any data on the disk will be lost.
a. Creating the Physical Volumes
# pvcreate /dev/sda1 /dev/sdb1 /dev/sdc1
b. Creating the Volume Group
# vgcreate volgroup01 /dev/sda1 /dev/sdb1 /dev/sdc1
Verify the volume group attributes,
# vgs
c. Creating the Logical Volume
# lvcreate -i3 -I4 -L2G -n striped_logical_volume volgroup01
The command creates striped logical volume striped_logical_volume with 2 gigabytes in size, with three
stripes and a stripe size of 4 kilobytes from the volume group volgroup01.
d. Creating the File System
# mkfs.gfs2 -plock_nolock -j 1 /dev/volgroup01/striped_logical_volume
e. Mount the filesystem.
# mount /dev/volgroup01/striped_logical_volume /mnt
# df -h /mnt
3. Splitting a Volume Group
Plan:
modify ?? below...
If there is enough unused space on the physical volumes, a new volume group can be created without adding
new disks.
In the initial set up, the logical volume mylv is carved from the volume group myvol, which in turn
consists of the three physical volumes, /dev/sda1, /dev/sdb1, and /dev/sdc1.
After completing this procedure, the volume group myvg will consist of /dev/sda1 and /dev/sdb1. A second
volume group, yourvg, will consist of /dev/sdc1.
a. Determining Free Space
determine how much free space is currently available in the volume group using pvscan
# pvscan - change the output. ..
PV /dev/sda1 VG myvg lvm2 [17.15 GB / 0 free]
PV /dev/sdb1 VG myvg lvm2 [17.15 GB / 12.15 GB free]
PV /dev/sdc1 VG myvg lvm2 [17.15 GB / 15.80 GB free]
Total: 3 [51.45 GB] / in use: 3 [51.45 GB] / in no VG: 0 [0 ]
b. Moving the Data
You can move all the used physical extents in /dev/sdc1 to /dev/sdb1 with the pvmove command. The pvmove
command can take a long time to execute.
# pvmove /dev/sdc1 /dev/sdb1
Verify that the space on /dev/sdc1 is free using the pvscan command.
# pvscan
PV /dev/sda1 VG myvg lvm2 [17.15 GB / 0 free]
PV /dev/sdb1 VG myvg lvm2 [17.15 GB / 10.80 GB free]
PV /dev/sdc1 VG myvg lvm2 [17.15 GB / 17.15 GB free]
Total: 3 [51.45 GB] / in use: 3 [51.45 GB] / in no VG: 0 [0 ]
c. Splitting the Volume Group
Using vfsplit command you can create new volume group yourvg, to split the volume group myvg.
Make sure your logical volume is inactive. If there are any mounts, unmount before deactiviting the logical
volume.
Use lvchange or vgchange command to deactivate the logical volume.
# lvchange -a n /dev/myvg/mylv
# vgsplit myvg yourvg /dev/sdc1
The above command deactivates the logical volume mylv and splits the volume group yourvg from volume group
myvg, moving physical volume /etc/sdc1 into new volume group yourvg.
Verify the volume group attributes.
# vgs
d. Creating the New Logical Volume
Now, create new logical volume yourlv after creating the new volume group.
# lvcreate -L5G -n yourlv yourvg
e. Create a File System and mount the New Logical Volume
# mkfs.gfs2 -plock_nolock -j 1 /dev/yourvg/yourlv
f. Activating and Mounting the Original Logical Volume
# lvchange -a y mylv
# mount /dev/myvg/mylv /mnt
# df
4. Removing a Disk from a Logical Volume
You can remove a disk from existing logical volume to use the disk for other volume or to replace the
failed disk. To process, you must have to move the extents on the LVM physical volume to a different disk
or the set of disks.
A. Moving Extents to Existing Physical Volumes
Plan:
We have a logical volume that is distributed across four physical volumes in the volume group myvg.
a.
# pvs -o+pv_used
PV VG Fmt Attr PSize PFree Used
/dev/sda1 myvg lvm2 a- 17.15G 12.15G 5.00G
/dev/sdb1 myvg lvm2 a- 17.15G 12.15G 5.00G
/dev/sdc1 myvg lvm2 a- 17.15G 12.15G 5.00G
/dev/sdd1 myvg lvm2 a- 17.15G 2.15G 15.00G
We want to move the extents off of /dev/sdb1 so that we can remove it from the volume group.
You need to have free extents on other physical volumes in volume group. The extents will be distributed to
the other devices.
# pvmove /dev/sdb1
/dev/sdb1: Moved: 2.0%
...
/dev/sdb1: Moved: 79.2%
...
/dev/sdb1: Moved: 100.0%
Check the distribution of extents using pvs command.
# pvs -o+pv_used
b. Remove the physical volume /dev/sdb1 from the volume group using vgreduce command.
# vgreduce myvg /dev/sdb1
# pvs
Now, disk can be physically remove or use for other purpose.
B. Moving Extents to a New Disk
Plan:
The logical volume is distributed across three physical volumes in the volume group myvg as follows:
# pvs -o+pv_used
PV VG Fmt Attr PSize PFree Used
/dev/sda1 myvg lvm2 a- 17.15G 7.15G 10.00G
/dev/sdb1 myvg lvm2 a- 17.15G 15.15G 2.00G
/dev/sdc1 myvg lvm2 a- 17.15G 15.15G 2.00G
We want to move the extents of /dev/sdb1 to a new device, /dev/sdd1.
a. Creating the New Physical Volume
scenario: Create a new physical volume from /dev/sdd1.
# pvcreate /dev/sdd1
b. Adding the New Physical Volume to the Volume Group
# vgextend myvg /dev/sdd1
# pvs -o+pv_used
c. Moving the Data
Use the pvmove command to move the data from /dev/sdb1 to /dev/sdd1.
# pvmove /dev/sdb1 /dev/sdd1
# pvs -o+pv_used
d. Removing the Old Physical Volume from the Volume Group
After you have moved the data off /dev/sdb1, you can remove it from the volume group.
# vgreduce myvg /dev/sdb1
Creating and using snapshop for backup
1. Create a new snapshot volume called snaplvname of /dev/vgname/lvname that is 20 mb in size
# lvcreate -s -n snaplv -L 20M /dev/vgname/lvname
2. If backup software requires it, mount the snapshot and point that back program to the new mountpoint.
# mkdir /snapmount
# mount -o ro /dev/vgname/snaplv /snapmount
3. Verify the status of the snapshot logical volume
# lvs /dev/vgname/snaplv
4. When done with snapshot, unmount and remove it.
# umount /snapmount
# lvremove /dev/vgname/snaplv
Based on the online doc @redhat....
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html-single/Logical_Volume_Manager_Administration/index.html#troubleshooting
1. Creating an LVM Logical Volume on Three Disks (or more if you have available).
2. Creating a Striped Logical Volume
3. Splitting a Volume Group
4. Removing a Disk from a Logical Volume
1. Creating an LVM Logical Volume on Three Disks.
Plan:
logical volume name: new_logical_volume
Disks: /dev/sda1, /dev/sdb1, and /dev/sdc1
partition type: Linux LVM which is 8e.
Warning: Verify that disks does not contain any important data.
a. Creating the Physical Volumes
# pvcreate /dev/sda1 /dev/sdb1 /dev/sdc1
# pvdisplay # command to display physical volume detail.
Note: Each physical volume has a UUID.
b. Creating the Volume Group
# vgcreate new_vol_group /dev/sda1 /dev/sdb1 /dev/sdc1
# vgs or # Displays new volume group attributes
# vgdisplay volume_name
Note: volume group name is new_vol_group
c. Creating the Logical Volume
# lvcreate -L 2G -n new_logical_volume new_vol_group # Creates 2 GB Logical volume
# lvcreate -L 3GB -n mylvvol new_vol_group # Creates 3 GB Logical volume
# lvdisplay # Displays Logical Volume attributes.
Here, L flag is sued to size while n flag is for new logical volume name.
Note: creates 2gb of logical volume new_logical_volume from the volume group new_vol_group.
You can create multiple logical volumes on a single volume group.
d. Creating the File System using mkfs command.
# mkfs.gfs2 -p lock_nolock -j 1 /dev/new_vol_group/new_logical_volume or
# mkfs.ext4 /dev/new_vol_group/new_logical_volume
Note: You can use mke2fs, mkfs.ext3, mkfs.ext4 commands with different options.
e. mount the logical volume
# mount /dev/new_vol_group/new_logical_volume /mnt
f. Add entry to /etc/fstab to be mounted across the reboot.
g. Resize the logical volume to expand or shrink the filesystem.
# lvresize -L 5GB /dev/new_vol_group/new_logical_volume
# resize2fs /dev/new_vol_group/new_logical_volume 5G
# lvresize -L 3GB /dev/new_vol_group/new_logical_volume
# resize2fs /dev/new_vol_group/new_logical_volume 3G
Note: To shrink you have to unmount the filesystem. To resize with new storage space added, use resize2fs command. The filessytem must be ext3 and up to resize the volume.
# resize2fs /dev/new_vol_group/new_logical_volume 5G
h. Remove the logical volume
# lvremove /dev/new_vol_group/new_logical_volume
i. Rename Logical Volume
# lvrename new_vol_group new_logical_volume my_logical_volume
# lvresize -L 10GB /dev/new_vol_group/my_logical_volume
Note: Only renames logical Volume not the volume group.
j. Rename Volume Group.
# vgrename new_vol_group my_vol_group
Note: Once you rename Logical volume or the volume group, make sure to change your mount point and the /etc/fstab entry.
=================================================================
2. Creating a Striped Logical Volume
Plan:
LVM striped logical volume: striped_logical_volume
Disk for stripping: /dev/sda1, /dev/sdb1, and /dev/sdc1
Warning: Any data on the disk will be lost.
a. Creating the Physical Volumes
# pvcreate /dev/sda1 /dev/sdb1 /dev/sdc1
b. Creating the Volume Group
# vgcreate volgroup01 /dev/sda1 /dev/sdb1 /dev/sdc1
Verify the volume group attributes,
# vgs
c. Creating the Logical Volume
# lvcreate -i3 -I4 -L2G -n striped_logical_volume volgroup01
The command creates striped logical volume striped_logical_volume with 2 gigabytes in size, with three
stripes and a stripe size of 4 kilobytes from the volume group volgroup01.
d. Creating the File System
# mkfs.gfs2 -plock_nolock -j 1 /dev/volgroup01/striped_logical_volume
e. Mount the filesystem.
# mount /dev/volgroup01/striped_logical_volume /mnt
# df -h /mnt
3. Splitting a Volume Group
Plan:
modify ?? below...
If there is enough unused space on the physical volumes, a new volume group can be created without adding
new disks.
In the initial set up, the logical volume mylv is carved from the volume group myvol, which in turn
consists of the three physical volumes, /dev/sda1, /dev/sdb1, and /dev/sdc1.
After completing this procedure, the volume group myvg will consist of /dev/sda1 and /dev/sdb1. A second
volume group, yourvg, will consist of /dev/sdc1.
a. Determining Free Space
determine how much free space is currently available in the volume group using pvscan
# pvscan - change the output. ..
PV /dev/sda1 VG myvg lvm2 [17.15 GB / 0 free]
PV /dev/sdb1 VG myvg lvm2 [17.15 GB / 12.15 GB free]
PV /dev/sdc1 VG myvg lvm2 [17.15 GB / 15.80 GB free]
Total: 3 [51.45 GB] / in use: 3 [51.45 GB] / in no VG: 0 [0 ]
b. Moving the Data
You can move all the used physical extents in /dev/sdc1 to /dev/sdb1 with the pvmove command. The pvmove
command can take a long time to execute.
# pvmove /dev/sdc1 /dev/sdb1
Verify that the space on /dev/sdc1 is free using the pvscan command.
# pvscan
PV /dev/sda1 VG myvg lvm2 [17.15 GB / 0 free]
PV /dev/sdb1 VG myvg lvm2 [17.15 GB / 10.80 GB free]
PV /dev/sdc1 VG myvg lvm2 [17.15 GB / 17.15 GB free]
Total: 3 [51.45 GB] / in use: 3 [51.45 GB] / in no VG: 0 [0 ]
c. Splitting the Volume Group
Using vfsplit command you can create new volume group yourvg, to split the volume group myvg.
Make sure your logical volume is inactive. If there are any mounts, unmount before deactiviting the logical
volume.
Use lvchange or vgchange command to deactivate the logical volume.
# lvchange -a n /dev/myvg/mylv
# vgsplit myvg yourvg /dev/sdc1
The above command deactivates the logical volume mylv and splits the volume group yourvg from volume group
myvg, moving physical volume /etc/sdc1 into new volume group yourvg.
Verify the volume group attributes.
# vgs
d. Creating the New Logical Volume
Now, create new logical volume yourlv after creating the new volume group.
# lvcreate -L5G -n yourlv yourvg
e. Create a File System and mount the New Logical Volume
# mkfs.gfs2 -plock_nolock -j 1 /dev/yourvg/yourlv
f. Activating and Mounting the Original Logical Volume
# lvchange -a y mylv
# mount /dev/myvg/mylv /mnt
# df
4. Removing a Disk from a Logical Volume
You can remove a disk from existing logical volume to use the disk for other volume or to replace the
failed disk. To process, you must have to move the extents on the LVM physical volume to a different disk
or the set of disks.
A. Moving Extents to Existing Physical Volumes
Plan:
We have a logical volume that is distributed across four physical volumes in the volume group myvg.
a.
# pvs -o+pv_used
PV VG Fmt Attr PSize PFree Used
/dev/sda1 myvg lvm2 a- 17.15G 12.15G 5.00G
/dev/sdb1 myvg lvm2 a- 17.15G 12.15G 5.00G
/dev/sdc1 myvg lvm2 a- 17.15G 12.15G 5.00G
/dev/sdd1 myvg lvm2 a- 17.15G 2.15G 15.00G
We want to move the extents off of /dev/sdb1 so that we can remove it from the volume group.
You need to have free extents on other physical volumes in volume group. The extents will be distributed to
the other devices.
# pvmove /dev/sdb1
/dev/sdb1: Moved: 2.0%
...
/dev/sdb1: Moved: 79.2%
...
/dev/sdb1: Moved: 100.0%
Check the distribution of extents using pvs command.
# pvs -o+pv_used
b. Remove the physical volume /dev/sdb1 from the volume group using vgreduce command.
# vgreduce myvg /dev/sdb1
# pvs
Now, disk can be physically remove or use for other purpose.
B. Moving Extents to a New Disk
Plan:
The logical volume is distributed across three physical volumes in the volume group myvg as follows:
# pvs -o+pv_used
PV VG Fmt Attr PSize PFree Used
/dev/sda1 myvg lvm2 a- 17.15G 7.15G 10.00G
/dev/sdb1 myvg lvm2 a- 17.15G 15.15G 2.00G
/dev/sdc1 myvg lvm2 a- 17.15G 15.15G 2.00G
We want to move the extents of /dev/sdb1 to a new device, /dev/sdd1.
a. Creating the New Physical Volume
scenario: Create a new physical volume from /dev/sdd1.
# pvcreate /dev/sdd1
b. Adding the New Physical Volume to the Volume Group
# vgextend myvg /dev/sdd1
# pvs -o+pv_used
c. Moving the Data
Use the pvmove command to move the data from /dev/sdb1 to /dev/sdd1.
# pvmove /dev/sdb1 /dev/sdd1
# pvs -o+pv_used
d. Removing the Old Physical Volume from the Volume Group
After you have moved the data off /dev/sdb1, you can remove it from the volume group.
# vgreduce myvg /dev/sdb1
Creating and using snapshop for backup
1. Create a new snapshot volume called snaplvname of /dev/vgname/lvname that is 20 mb in size
# lvcreate -s -n snaplv -L 20M /dev/vgname/lvname
2. If backup software requires it, mount the snapshot and point that back program to the new mountpoint.
# mkdir /snapmount
# mount -o ro /dev/vgname/snaplv /snapmount
3. Verify the status of the snapshot logical volume
# lvs /dev/vgname/snaplv
4. When done with snapshot, unmount and remove it.
# umount /snapmount
# lvremove /dev/vgname/snaplv
Based on the online doc @redhat....
http://docs.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/6/html-single/Logical_Volume_Manager_Administration/index.html#troubleshooting
Subscribe to:
Posts (Atom)