RH133

Redhat Enterprise Linux System

Administration
Unit 1
 Installation
Hardware Overview
 Kernel Support
 Core support: CPU, Memory, Process
 Management , Interrupt/Exception Handling etc.
 Dynamically Loadable Kernel Modules
 Device Drivers
 Additional Functionality
 User Mode Access to kernel facilities
 System Calls and Signals
 Filesystem Device Nodes
 Network Interfaces
 Are not accessed through a device node but instead are
accessed through a “network interface” abstraction.
CPU and Memory
 Seven Supported Architectures: x86, Itanium2,
AMD64/EM64T, S/390, zSeries, iSeries, pSeries.
 CPU Support on x86
 Technical support for more than 2 physical CPUs only on
AS variant (may use Hyper-Threading)
 Up to 32 Physical CPUs with SMP or hugemem kernel.
 Memory support on x86
 Technical support for more than 16 GB on AS or WS
 Standard i686/athlon kernel: 4GB
 SMP i686/athlon kernel: 16GB
 Hugemem SMP kernel: 64GB
Preparing to Install
 Read the RELEASE-NOTES file on the first
CD or at http://www.redhat.com
 Check Hardware Compatibility
 Redhat Supported Hardware List
 Hardware compatible with Redhat Linux
 http://hardware.redhat.com/hcl
 XFree86 supported video cards.
 http://xorg.freedesktop.org
 http://www.x.org/wiki
Multiboot systems
 Redhat Enterprise Linux and the GRUB boot
loader can co-exist with other operating systems,
including the following:
 Windows NT/2000/XP/2003
 DOS, Windows 3.x/9x/ME
 NetBSD, FreeBSD and other open systems.
 Two major issues arise when implementing
multiboot systems:
 Partitioning and the boot process.
 A boot loader such as System Commander or
NTLDR is already on the system and will launch
GRUB as a secondary boot loader.
Device Node Examples
 Block devices:
 hd[a-t] IDE devices
 sd[a-z]+ SCSI devices
 fd[0-7] Standard floppy drives
 md[0-31] software RAID metadisks
 loop[0-15] loopback devices
 ram[0-9] ramdisks
 Character Devices:
 tty[0-31] virtual consoles
 ttyS[0-9]+ Serial ports
 lp[0-3] Parallel Ports
 null infinite sink ( the bit bucket)
 zero infinite source of zeros
 [u]random sources of random information
 fb[0-31] framebuffer devices
 Symbolic Links:
 /dev/cdrom - - > /dev/hd[a-t], /dev/sd[a-z]+
 /dev/modem - - > /dev/ttyS[0-9]+
 /dev/pilot - - > /dec/ttyS[0-9]+
The RHEL Installer
 First Stage Installer Images
 diskboot.img – VFAT filesystem image for
bootable media larger than a
floppy
 You will need to use the dd command to move this image to you media. For
instance:
dd <diskboot.img > /dev/sda
 Floppy installation is no longer supported
 boot.iso -- ISO9660 bootable CD image
 Booting form boot.iso is the same as passing the askmethod argument to
the installer when booting from CD 1.
 You can create a bootable CD using the cdrecord command. For instance

cdrecord dev=/dec/hdc boot.iso

 pxeboot Directory
 Pre-boot Execution Environment (PXE) provides for a diskless installation.
 Read /usr/share/doc/syslinux-2.11/prelinux.doc
 Second Stage Installer
 Graphical or textual
 Can be invoked in noprobe or Kickstart mode
 Once located and loaded by the first stage, drives the remainder of the
installation process.
Installer Features
 noprobe and Kickstart modes available
 mediacheck tests media integrity
 Multiple Interfaces:
 Graphical
 Starts X server and a GUI installer
 Can be started in lowers mode.
 Works with hard drive, CDROM, NFS Installation
 Graphical is the default
 Text
 Menu-based terminal interface
 Works with all installation methods (ftp and http)
RHEL Installation Overview
 Language, Keyboard and mouse selection
 Media selection if applicable
 Disk partitioning
 Bootloader configuration
 Network and firewall configuration
 Authentication Setup
 Package Selection
 X server configuration
Partitioning Hard Drives
 Hard drives are divided into partitions.
 Partitions normally contain file systems.
 Primary, extended and logical partitions
 The default filesystem is ext3
 Multiple partitions may be assembled into a
larger virtual partitions: software RAID and
LVM
 Filesystems are accessed via a mount
point, which is a designed directory in the
file system hierarchy.
Software RAID
 Redundant Array of Inexpensive Disks
 Multiple partitions on different disks combined
into one RAID device
 Fault tolerance, larger disk size, performance
 Install-time RAID levels:
 RAID 0: Striping (no redundancy)
 RAID 1: Mirroring
 RAID 5: Striping with distributed parity
Configuring File Systems
 Must select mount points, partition sizes,
and file system types in the installer
 Can set up manually or automatically
 There are many layouts which may be
used
 / mast include /etc, /lib, /bin, /sbin, /dev
 Swap space is typically 2x physical RAM
 Typical mount points: /boot, /home, /usr, /var,
/tmp, /usr/local, /opt
Network Configuration
 Can configure each NIC independently
 DHCP or static IP configuration
 Determine if automatically activated on boot
LVM: Logical Volume Manager
 Manages storage on one or more
partitions as virtual partitions, or logical
volumes
 Real partitions are physical volumes and are
assigned to a volume group (a virtual disk)
 Disk space in the volume group is divided into
extends which are assigned to a logical volume
 Easy to resize logical volumes
 Add a physical volume to the volume group
and assign the new extents to the logical
volume.
Firewall Setup
 Installer can set up a kernel mode stateful
packet filter
 Choice of two settings: “Enabled” and “No
Firewall”
 “Trusted Devices” can bypass the firewall
 Can allow access to arbitrary services.
Security Enhanced Linux
 Access control determines what actions
processes can perform on what objects
 Discretionary Access Control (Traditional Linux)
 Users control permissions on objects
 Mandatory Access Control (SELinux)
 System policy restricts permission which can be
granted.
SELinux Installation Options
 Installation Options:
 Disabled
 Warn (Permissive)
 Active (default) (Enforcing)
Package Selection
 Package Selection
 Universally (“Everything”)
 By predefined components
 Defined in RedHat/base/comps.xml
 Individually
Validating Installation
 Virtual consoles during installation
 Post-boot validation
 dmesg and /var/log/dmesg
 /var/log/messages
 /root/install.log
 GRUB drops to a prompts if there is a
problem loading files.
noprobe Mode and Driver Disks
 Method for supporting hardware newer
than the install program
 Used at install time for less common
hardware
 Prompt for Driver Disk
 When run in noprobe mode
 When started with: linux dd
 When no PCI devices are detected.
Post-Install Configuration
 Setup Agent (firstboot)
 Configure X window System if necessary
 Set date and time
 Register with Redhat Network and get updated
RPMs
 Install additional RPMs or Redhat
Documentation from CDROM
 Setup users
 system-config-* configuration tools
Unit 2
 System Initialization and Services
Boot Sequence Overview
 BIOS initialization
 Boot Loader
 Kernel Initialization
 init starts and enters desired run level by
executing:
 /etc/rc.d/rc.sysinit
 /etc/rc.d/rc and /etc/rc.d/rc?.d
 /etc/rc.d/rc.local
 X Display Manager if appropriate
BIOS initialization
 Peripheral detected
 Boot device selected
 First sector of boot device read and
executed
Boot Loader Components
 Boot Loader
 1st Stage – small, resides in MBR or boot sector
 2nd Stage – loaded from boot partition
 Minimum Specifications for Linux:
 Label, kernel location, OS root filesystem and
Location of the initial ramdisk (initrd)
 Minimum specification for other OS:
 Boot device, label
GRUB and grub.conf
 GRUB – The Grand Unified Bootloader
 Command-line interface available at boot prompt
 Boot from ext2/ext3, ReiserFS, JFS, FAT, minix, or FFS
filesystems
 Support MD5 password protection
 /boot/grub/grub.conf
 Changes to grub.conf take effect immediately
 If MBR on /dev/had is corrupted, reinstall the first
stage bootloader with:
 /sbin/grub-install /dev/hda
Starting the Boot Process: GRUB
 Image selection
 Select with space followed by up/down arrows
on the boot splash screen
 Argument passing
 Change an exiting stanza in menu editing
mode
 Issue boot commands interactively on the
GRUB command line
init Initialization
 init reads its config: /etc/inittab
 Initial run level
 System initialization scripts
 Run level specific script directories
 Trap certain key sequences
 Define UPS power fall/restore scripts
 Spawn gettys on virtual consoles
 Initialize X in run level 5
Kernel Initialization
 Kernel boot time functions
 Device detection
 Device driver initialization
 Mounts root filesystem read only
 Loads initial process (init)
/etc/rc.d/rc.sysinit
 Important tasks include:
 Activate udev and selinux
 Sets kernel parameters in /etc/sysctl.conf
 Sets the system clock
 Loads keymaps
 Enables swap partitions
 Sets hostname
 Root filesystem check and remount
 Active RAID and LVM devices
 Enable disk quotas
 Check and mount other filesystems
 Cleans up stale locks and PID files.
System V run levels
 Run level defines which services to start
 Each run level has a corresponding directory
 /etc/rc.d/rcX.d
 The system V init scripts reside in:
 /etc/rc.d/init.d
 Symbolic links in the run level directories call
the init.d scripts with a start or stop
arguments.
Daemon Processes
 A daemon process is a program that is run
in the background, providing some sytem
service
 Two types of daemons:
 Standalone
 Transient – Controlled by the “Super-daemon”
xinetd
/etc/rc.d/rc
 initiallzes the default run level per the
/etc/inittab file initdefault line such as
 id:3:initdefault

 10:0:wait:/etc/rc.d/rc 0
 11:1:wait:/etc/rc.d/rc 1
 12:2:wait:/etc/rc.d/rc 2
 13:3:wait:/etc/rc.d/rc 3 <--- (run level 3)
 14:4:wait:/etc/rc.d/rc 4
 15:5:wait:/etc/rc.d/rc 5
 16:6:wait:/etc/rc.d/rc 6
 17:7:wait:/etc/rc.d/rc 7
/etc/rc.d/rc.local
 Run after the run level specific scripts
 Common place for custom modification
 In most cases it is recommended that you
create a System V init script in
 /etc/rc.d/init.d unless the service you are
starting is so trivial it doesn’t warrant it.
Existing scripts can be used as a starting
point.
Virtual Consoles
 Multiple independent VT100-like terminals
 Defined in /etc/inittab
 Accessed with Ctrl-Alt-F_key from an X session
 /dev/ttyn: virtual console n
 /dev/tty0: the current virtual console
 Default RedHat Enterprise Linux Configuration
 12 consoles defined
 Consoles 1-6 accept logins
 X server starts on the first available console, usually 7.
Controlling Services
 Utilities to control default service startup
 system-config-services: graphical utility that requires
and X interface
 ntsysv : ncurses based utility usuable in virtual
consoles
 chkconfig: a fast, versatile command line utility that
works well and is usable with scripts and Kickstart
installations
 Utilities to control services manually
 service: immediately Start or stop a standalone service
 chkconfig: immediately starts and stop xinetd-
managed service.
System Shutdown
 Shutting down the system
 shutdown –h now
 halt
 poweroff
 init 0
System Reboot
 Rebooting rarely fixes problem in Linux
 If you feel a reboot is necessary try bringing
the system down to runlevel 1 and the back up
to runlevel 3 or 5. This is much faster than a
reboot.
 Rebooting the system:
 shutdown –r now
 reboot
 init 6
Unit 3
 Kernel Services and Configuration
Kernel Modules
 Modular kernel components
 Components that need not be resident in the
kernel for all configurations and hardware
 Peripheral device drivers
 Supplementary filesystems
 Modules configurable at load time
 /lib/modules
 Controlling Modules
 lsmode, modprobe
 Kernel Tainting
Kernel Module Configuration
 Module examination: /sbin/modinfo
 Parameters, license
 Module Configuration: /etc/modprobe.conf
 Aliases, parameters, actions
 Module Dependencies: modules.dep,
depmod
 Manual control: insmod, rmmod
The /proc filesystem
 /proc is a vital filesystem containing
information about the running kernel
 Contens of “files” under /proc may be
viewed using cat
 Example
 cat /proc/interrupts
 Provides information on system hardware,
networking settings and activity, memory
usage, and more.
The /proc filesystem, cont’d
 /proc subdirectories
 The /proc/sys subdirectory allows
administrators to modify certain
parameters of a running kernel.
/proc/sys configuration with sysctl
 /proc/sys modifications are temporary and
not saved at system shutdown
 The sysctl command manages such
settings in a static and centralized fashion:
 /etc/sysctl.conf
 sysctl is called at boot time by rc.sysinit
and uses setting sin /etc/sysctl.conf
General Hardware Resources
 dmesg and /var/log/dmesg
 kudzu
 /etc/sysconfig/hwconf
 /usr/share/hwdata/
 /proc filesystem
 hwbrowser
System Bus Support
 PCI Bus
 /sbin/lspci
 /proc/bus/pci
 ISA Bus
 /proc/isapnp
Hotswappable Bus Support
 USB and IEEE 1394 Buses
 /sbin/hotplug, (/etc/hotplug/)
 Information in /proc/bus subdirectories
 /sbin/lsusb and /sbin/usbmodules utilities
 USB devices in /dev/usb
 PCMCIA Bus
 /sbin/cardmgr, (/etc/pcmcia/)
 Information in /proc/bus/pccard
 /sbin/cardctl utility
System Monitoring and Process
Control
 top, gnome-system-monitor  display
snapshot of processes
 ymstat – reports virtual memory stats
 iostat – lists information on resource
usage, including I/O statistics
 free – summary of system memory usage
 renice – change priority of a process
 kill – send system signal to a process
Unit 4
 Filesystem Management
System Initialization: Device
Recognition
 Master Boot Record (MBR) contains:
 Executable code to load operating system
 Space for partition table information, including:
 Partition id and type
 Starting cylinder for partition
 Number of cylinder for partition
Disk Partitioning
 An extended partition points to additional
partition descriptors
 Total maximum number of partitions
supported by the kernel:
 63 for IDE drives
 15 for SCSI drives
 Why partition drives?
 Containment, performance, quotas, recovery
Managing Partitions
 Create partition using:
 fdisk
 sfdisk
 GNU parted – Advanced partition manipulation
(create, copy, resize, etc)
 Partprobe – reinitializes the kernel’s in
memory version of the partition table.
Managing Data: Filesystem
creation
 mkfs
 mkfs.ext2, mkfs.ext3, mkfs.minix,
mkfs.msdos
 Specific filesystem utilities may be called
directly
 mke2fs [options] device
Journaling for ext2 filesystems:
ext3
 ext3 is essentially an ext3 filesystem that
uses a journal for file transaction
automatically.
 ext3 filesystems can be created natively or
easily converted from ext2
 Ext3 has three journaling modes:
 Ordered – the default, journals only meta-data
 Journaled – Journals data as well as meta-data
 Writeback – Journals updates are not
automatic, but gives better performance at
possible expense of data integrity.
Managing data: mount
 mount [options] [device] [mount_point]
 device (or filesystem label) points to the
filesystem to mount.
 mount_point is the directory under which
the files on the filesystem will be located.
Managing Data: mount options
 -t vfstype (vfat, ext2, ext3, iso9660, etc.)
 Not normally needed
 -o options
 Default options for the ext2/ext3 filesystem:
 rw, suid, dev, exec, auto, nouse, and async
Managing Data: Unmounting
Filesystems
 umount [options] device | mnt_point
 A filesystem “in use” may not be
unmounted
 Use fuser to check and/or kill processes
 Use the remount option to change a
mounted filesystem’s options
“automatically”
 mount –o remount,ro /data
Managing Data: Filesystem Labels
 Alternate way to refer to devices
 Device independent
 e2lable <special_dev_file>
 mount [options] LABEL=fslabel mount_point
Managing Data: mount, by
example
 Sample filesystem requirements met using
options:
 Disabling execute access
 Mounting a filesystem image
 Mounting a pc-compatible filesytem.
 Disabling access time updates.
 Setting up a mount alias
Managing Data: Connecting Network
Resources
 Mounting NFS resources
 Requires hostname or address of server
 Requires name of exported directory
 Mounting SMB resources
 Requires hostname and address of server
 Requires share name
 May require username and password
Managing Data: /etc/fstab
 Configuring of the filesystem hierarchy
 Used by mount, fsck, and other programs
 Maintains the hierarchy between system
reboots
 May use filesystem volume labels in the
device field
Managing Data: The auto-Mounter
 System administrator specifies mount
points to be controlled by the automounter
daemon process.
 The automounter monitors access to these
directories and mount the filesystem on
request.
 Filesystems automatically unmounted
after a specified interval of inactivity.
 Enable /etc/auto.net to “browse” all NFS
exports on the network.
ext2/ext3 Filesystem Attributes
 ext2 and ext3 support attributes that
affect the manipulation of the file data.
 lsattr display file attributes
 chattr changes file attributes
 Some attributes are not currently supported by the
Linux kernel.
Virtual Memory
 Swap space is supplement to system RAM
 Basic setup involves:
 Create swap partition or file
 Write special signature using mkswap
 Add appropriate entries to /etc/fstab
 Activate swap space with swapon -a
Filesystem Maintenance
 Maintaining consistency with fsck
 Filesystems checked at boot up
 sulogin session started if errors are sever
 lost+found
 tune2fs
 dump2fs
 debugfs
 parted
Adding a Drive
 Physically connect the new drive
 Create partitions
 If required, reread partition table with
partprobe
 Verify with fdisk –l and cat /proc/partitions
 Create filesystems for new partitions, or
 Write signature to new swap partitions
 Optionally create disk label
 Create any needed mount points
 Add new entries to /etc/fstab
Unit 5
 Network Configuration
Device Recognition
 All drivers for network interface cards are
built as module
 Networking scripts reference logical
interface names, eg:
 eth0
 /etc/modprobe.conf maps logical names to
specific module name
 Example:
 Alias eth0 3c59x
Network Interfaces
 Interface Names:
 Ethernet : eth0, eth1, ethN
 Token Ring : tr0, tr1, trN
 FDDI : fddi0, fddi1, fddiN
 PPP : ppp0, ppp1, pppN
 Data link layer addresses
 ifconfig
mii-tool
 Views and controls the negotiated media
speed (100baseTX, 10baseT) of some
ethernet cards.
 Useful for forcing specific ethernet speed
and duplex settings
 Changes with mii-tools should be made on
inactive interfaces.
ifconfig
 Used to configure and set IP address on
network interfaces
 Not Usually called directly, but by other scripts
 Also used to view properties of active and
inactive network interfaces.
ifup/ifdown
 if (up | down) interface
 Start and Stop network interfaces
 Take care of details specific to interface
 Changing/adding/deleting routes
 Obtains addresses as needed
 BOOTP, DHCP
Interface configuration file
 ifcfg-xxx
 Located in:
 /etc/sysconfig/network-scripts/
 Configuration methods
 Static
 dhcp
 bootp
Configuration Utilities
 netconfig
 Text-based network configuration tool
 Only writes config files. Does not activate
device or changes. Use ifup/ifdown to activate
changes
 Used by kudzu when new network card found
at boot time.
 system-config-network
 GNOME-based network configuration tool
 Can be launched by a non-privileged user, but
requires authentication as root.
Binding multiple IP addresses
 Use multiple IP addresses on a NIC
 Virtual interface (s)
 For a small number of IPs, create an ifcfg
file for each virtual interface
 ifcfg-ethX:xxx
 For a large number of IPs, create an ifcfg
range file
 ifcfg-ethx-rangeX
DHCP/BOOTP
 The dhclient daemon manages client-side
DHCP and BOOTP
 For DHCP, dhclient:
 Obtains a lease
 Performs automatic lease renewal
 Normally run by ifup/ifdown
 Can be run manually to force renewal or
release of a lease
Global Network Parameters
 /etc/sysconfig/network
 NETWORKING=yes|no
 HOSTNAME=<fqdn by default>
 GATEWAY=<gateway IP>
 NISDOMAIN=<nis domain name>
Default Route
 Global default defined in:
 /etc/sysconfig/network
 GATEWAY=xxx.xxx.xxx.xxx
 Default gateway can also be defined in
 /etc/sysconfig/network-scripts/ifcfg-XXX
 ifcfg-xxx default overrides Global default routes
 GATEWAY=xxx.xxx.xxx.xxx
Static Routes
 Connected networks
 Linux kernel automatically creates a network
route for connected networks
 Static routes defines per interface
 /etc/sysconfig/network-scripts/route-eth0
 /etc/sysconfig/networking/devices/eth0.route
 Display with:
 route –n
 netstat -rn
Name Resolution
 hostname – display or set the system’s
name
 Is initially set by rc.sysinit from $HOSTNAME
variable
 /etc/sysconfig/network
 /etc/hosts – local database of hostname to
IP address mappings
 Checked before DNS
 Useful for small isolated networks
DNS client configuration
 /etc/resolv.conf
 Defines which name servers to use
 Servers are checked in order listed
DNS Utilites
 Useful utilites in bind-utils RPM package
include:
 host : gather host/domain information
 host ns1.redhat.com
 host –a redhat.com
 dig: send queries to name server directly
 Dig @ns1.redhat.com mx redhat.com
 nslookup
Network diagnostics
 ping:
 Network packet loss and latency measurement
tool
 traceroute, mtr
 Display network path to a destination
 netstat
 Multi-purpose network information tool
Unit 6
 RPM and Kickstart
The RPM Way
 Package installation is never interactive
 Applies to all software (core OS and add-
ons)
 No such thing as a patch to a package
RPM Package Manager
 RPM Components
 local database
 rpm and related executables
 package files
 Primary functions
 install/remove
 query
 verify
 build
Installing and Removing Software
 Primary RPM Options:
 Install : rpm –i, -- install
 Upgrade : rpm –U, --upgrade
 Freshen : rpm –F, --freshen
 Erase : rpm –e, --erase
 Output Options: -v, -h
 URL support: ftp:// (with globbing),
http://
 Many other install-options are available to
address special cases.
Updating a Kernel RPM
 Make sure to install kernel updates
 Do not use rpm –U or rpm –F !
 rpm –ivh kernel-version.arch.rpm
 Boot new kernel to test
 Revert to old kernel if a problem arises
 rpm –e kernel-olderversion if no problems
rpm queries
 Syntax:
 rpm –q what_packages what_information
 Installed Package options:
 rpm –qa List installed packages
 rpm –qi filename shows owning package
 rpm –qi package_name general information
 rpm –qi package_name lists files in package
 Uninstalled Package Options
 rpm -qip packages_file.i386.rpm
 rpm –qlp packages_file.i686.rpm
rpm verification
 Installed RPM file Verification:
 rpm –V package_name
 rpm –Vp package_file.i386.rpm
 rpm –Va
 Signature verification BEFORE package
install:
 rpm –import gpg_key
 rpm –checksig package_file.i386.rpm
Other RPM Utilities and Features
 rpm2cpio: file extraction
 rpmdb-redhat: distribution database
 rpm –redhatprovides filename
 rpm –redhatprovides capability
 system-config-packages
Automatic Dependency Resolution
 Automatic installation of dependent
packages
 Invokes with –aid option
 Use in conjunction with rpmdb-redhat
 Macro can indicate where packages files
found.
RedHat Network (RHN)
 RHN Components
 RHN Account
 System identity
 /usr/sbin/up2date
 rhnsd daemon and queued actions
 Advantages
 Errata concurrency
 Collective and remote administration
 Base metal provisioning
RHN in the Enterprise
 Management Entitlements
 System grouping
 Multiple administrators
 Proxy Server
 Updates cached locally conserving bandwidth
 Private channels
 Satellite Server
 Client profiles stored locally
 Custom channel management
 Provisioning Module
RHN Registration
 /usr/sbin/up2date
 username, password, system name
 Remote Information
 Hardware Profile
 Software Profile (RPM list)
 Subscribed Channel
 Local Digital Certificate
 /etc/sysconfig/rhn/systemid
The up2date utility
 Interactive or batch invocations
 Functions
 Freshen with published errata/updates
 Install new packages
 Resolve package dependencies
 /usr/sbin/up2date-config
 Install or download only
 Cache dir: /var/spool/up2date
Remote Administration
 Web based administration
 https://rhn.redhat.com
 Queuing of actions
 Local polling: rhnsd
 Every 4 hours by default
 Tuned in /etc/sysconfig/rhnsd
 /usr/sbin/rhn_check does the hard work.
Network Installation Server
 Necessary for network-based Installs
 Often faster than CDROM-based
installation methods
 Provides an easy distribution platforms for
the enterprise
 Shares the Redhat directory via NFS, FTP
and/or HTTP
Using Kickstart to automate
Installation
 Kickstart is a component of the installer that
automates in installation
 Kickstart supports all installation methods.
 The installer reads information from an ASCII file
rather than prompting for it
 Kickstart files can be made available via floppy,
cdrom, hard disk, initrd, nfs, ftp and http. They
can also be dynamically generated using cgi
scripts and specified using dhcp/pxe.
Kickstart: Commands Sections
 Constructs arguments that are passed to
configuration utilites (“commands”)
 The absence of required specifications
(e.g., keyboard) will raise the appropriate
utility.
 Commands section must come first.
Kickstart: %packages
 %packages specifies components groups
and RPMs to install.
 Component groups in the comps.xml file
are specified with @ component-group
 Third party RPMs cannot be specified
without modifying hdlist
 Package names only (not version).
Kickstart: %pre, %post
 %pre gives you the first word
 Executes as a bash shell script
 Executes after kickstart file is parsed
 %post gives you the final word
 Can specify interpreter (bash is default)
 chroot’ed by default, but may be run without
chroot.
Unit 7
 User Administration
User Policy Considerations
 Amount of system access outside of user’s
account
 Determine “need to know”
 Expiration of passwords and accounts
 Disk usage and CPU limits
User Account Database: /etc/passwd
 Contains account information used at login
and by other programs
 One account per line with seven colon-
delimited
 Should have permission rw-r-- r--
Adding a New User Account
 Most common method is useradd:
 useradd username
 Running useradd is equivalent to :
 Edit /etc/passwd, /etc/shadow, /etc/group
 Create and populate home directory
 Set permissions and ownership
 Set account password and using passwd
 Accounts may be added in a batch with
newusers.
User Private Groups
 When user accounts are created, a private
group is also created with the same name.
 Users are assigned to this private group.
 User’s new files affiliated with this group.
 Advantage: Prevents new files from
belonging to a “Public” group.
 Disadvantage: may encourage making
files “world-accessible”
Group Administration
 Entries to /etc/group
 groupadd
 groupmod
 groupdel
Modifying/Deleting Accounts
 To change files in a user’s /etc/passwd
entry you can:
 Edit the file by hand
 Use usermod [options] username
 To remove a user either:
 Manually remove the user from /etc/passwd
 /etc/shadow, /etc/group, /var/spool/mail
 Use userdel [-r] username
Password Aging Policies
 By default, passwords do not expire.
 Forcing passwords to expire is part of a
strong security policy.
 Modify default expiration settings in
 /etc/login.defs
 To modify password aging for exiting
users, use the chage command
 chage [options] username
Login Shell Scripts
 /etc/profile
 /etc/profile.d/*.sh
 ~/.bash_profile
 ~/.bashrc
 /etc/bashrc
Non Login Shell Scripts
 ~/.bashrc
 /etc/bashrc
 /etc/profile.d/*.sh
Switching Accounts
 Syntax:
 su [-] [user]
 su [-] [user] –c command
 Allows the user to temporarily before
another user.
 Default user is root
 The “-” option makes the new shell a login
shell.
sudo
 Users listed in /etc/sudoers execute
commands with:
 An effective user id of 0
 Group id of root’s group
 An administrator will be contracted if a
user not listed in /etc/sudoers attempts to
use sudo.
Network Users
 Information about users may be centrally
stored and managed on a remote server.
 Two types of information must always be
provided for each user account.
 Account Information: UID number, default
shell, home directory, group memberships, and
so on.
 Authentication: a way to tell that the password
provided on login for an account is correct.
Authentication Configuration
 system-config-authentication
 GUI tool to configure authentication
 For text-based tool, use –nox option
 Supported account information services:
 (local files), NIS, LDAP, Hesoid, Windbind
 Supported authentication mechanisms:
 (NSS), kerberos, LDAP, SMB, Winbind
Example: NIS Configuration
 Must install ypbind and portmap RPMs
 Run system-config-authentication
 Enable NIS to provide User Information
 Specify NIS Server and NIS domain name
 Keep default authentication (through NSS)
 What does this actually do?
 Four text-based configuration files are
changed.
Example: LDAP Configuration
 Must install nss-ldap and openldap RPMs
 Run system-config-authentication
 Enable LDAP to provide User Information
 Specify server, the search base DN and TLS
 Enable LDAP to provide Authentication
 What does this actually do?
 Four text-based configuration files are
changed.
File Ownership
 Every file has both user and group
“ownership”
 A newly created file will be owned by:
 The user who creates it
 The current primary group of that user
 SGID directories may change this behavior
 The chown command can be used by root
to change ownership.
Linux File Permissions
 Access levels
 Access modes
 Flags indicate access mode for each access
level
 File mode is a concise collective
expression of flags’ values.
SUID/SGID Executables
 Normally processes started by a user run
under the user and group security context
of that user.
 SUID and/or SGID bit set on an
executable file cause it to run under the
user and/or group security context of the
file’s owner and/or group.
Default File Permissions
 Read and write for all is the default for files.
 Read, write and execute is the default for
directories.
 umask can be used to withhold permissions on
file creation.
 Non-system users’ umask is 002
 Files will have permission of 664
 Directories will have permission of 775
 Supports users private groups
 System User’s umask 022
The Setgid Access Mode
 Normally, files created in a directory
belong to the default group of the user.
 When a file is created in a directory with
the setgid bit set, it belongs to the same
group as the directory.
SELinux
 Each process or object (file, directory,
network socket also has a SELinux
context.
 identity:role:domain/type
 The SELinux policy controls
 What identities can use which roles
 What roles can enter which domains
 What domains can access which types.
Access Control Lists (ACLs)
 Grant RWX access files to multiple users
or groups
 mount –o acl
 getfacl file|directory
 setfacl –m u:gandolf:rwx
 setfacl –m g:nazgul:rw
 setfacl –m d:u:frodo:rw
 setfacl –x u:samwise
Controlling SELinux
 system-config-securitylevel
 setneforce and setsebool
 /etc/sysconfig/selinux
 enforcing=0
 /selinux virtual file system
SELinux Contexts
 List process contexts: ps –Z
 List file contexts: ls –Z
 Change file contexts: chcon
 chron –t httpd_sys_content_t index.html
 chron –reference=/var/www/html index.html
Troubleshooting SELinux
 What is the error?
 Check /var/log/messages for AVC denials
 Is the process doing something it shouldn’t?
 Does the target have the right context?
 Does a Boolean setting need adjustment?
Unit 8
 Printing and Administration Tools
CUPS Overview
 New IPP protocol based on HTTP/1.1
 Web administration interface on port 631
 Can communicate with LPD print servers
 System V and BSD command interface
 Classes support automatic job redirection and
printer pooling
 Authentication by user/host/digital certificate
 Log files in web server Common Log Format
 Print Queue Design
 program  lp  cupsd  filter  printer
CUPS Configuration Files
 /etc/cups/cupsd.conf
 cupsd server configuration file
 Similar syntax to Apache httpd.conf file
 /etc/cups/printers.conf
 Print queue configuration file
 Automatically generated by lpdadmin, system-
config-printer or the CUPS web administration
interface.
CUPS Queue Management
 system-config-printer
 system-config-printer-tui
 Web interface: http://localhost:631/
 To authenticate, user must be a member of the
SystemGroup (sys by default) listed on
/etc/cups/cupsd.conf
 Connection is not encrypted
 lpadmin – command line tool for printer
administration
cron
 Used to schedule recurring events
 Use crontab to edit, install, and view job
schedules
 Syntax
 crontab [-u user] file
 crontab [-l|-r|e]
 -l lists crontab
 -r removes crontab
 -e edit crontab using $EDITOR
Controlling Access to cron
 Restrict/allow user access to cron
 /etc/cron.allow
 /etc/cron.deny
 Contain usernames to allow/deny access.
System crontab files
 Different format than user crontab files
 Master crontab file /etc/crontab runs
executables in
 /etc/cron.hourly
 /etc/cron.daily
 /etc/cron.weekly
 /etc/cron.monthly
 /etc/cron.d/ directory contains additional
system crontab files.
System cron job : tmpwatch
 Cleans old files out specified directories
 Useful for keeping /tmp directory from
filling up
 tmpwatch is run daily in /etc/cron.daily
System cron Job: logwatch
 Monitor with logwatch
 Helps catch problem issues
 Detects suspicious behavior
 logwatch is run daily in /etc/cron.daily
 Configuration file:
 /etc/log.d/conf/logwatch.conf
 Sends nightly email report
 Other tools
System Cron Job: logrotate
 Maintain log files from getting too large
 Keeps log files from getting too large
 Keeps filesystem from filling up
 logrotate is run daily in /etc/cron.daily
 Highly configurable
 Configure all logs in /etc/logrotate.conf
 Configure individual log files in files within
/etc/logrotate.d
syslog Configuration
 syslog System V initialization script in
 /etc/rc.d/init.d controls both the syslogd
and the klogd daemons
 /etc/syslog.conf
 Configures system logging
 /etc/sysconfig/syslog
 Sets switches used when starting syslogd and klogd
from the System V initialization Scripts
Tape Drives
 SCSI tape devices (i.e, DDS, DLT)
 /dev/[n]st0, /dev/[n]st1, etc.
 Devices with ‘n’ do not automatically rewind
 Use the mt utility to control tape drive
 mt –f /dev/st0 rewind (rewind)
 mt –f /dev/st0 fst 50 (Position)
 mt –f /dev/st0 offline (Eject)
 mt –f /dev/st0 erase (Erase)
 mt –f /dev/st0 rewoff (Rewind, Eject)
Using tar/star
 Archives to tapes or other media or files
 star backs up SELinux context and ACL attributes
 Parameter:
 c create x extra
 t list v verbose
 z gzip compression j bzip2 compression
 Examples:
 cd /tmp && tar xvf ~/archive.tar
 tar cvf /dev/st0 /data /foo /bar
Using dump/restore
 Back up and restore ext2/3 filesystems
 Does not work with other filesystems
 dump should only be used on unmounted
filesystems or filesystems that are read only
 Can do full or incremental backups
 Examples
 dump -0u –f /dev/nst0 /dev/hda2
 restore –fr /dev/nst0
Using cpio
 Similar to tar
 Does no recurse directories by itself
 Can archive special files
 Piping output from find into cpio is common
 Examples:
 find /data | cpio –ocv > /dev/nst0
 cpio -icdvm < /dev/nst0
 cpio -tvf < mybackup.cpio
Remote Backups
 Dump and tar call use rmt (remote tape
mgr)
 dump -0uf joe@svr:/dev/nst0 /home
 Use user@host:path format to specify the
remote user, host and device.
 dump can use ssh for secure backups
when RSH environment variable to set to
ssh.
Other backup software
 Higher-level applications for tape backup
include:
 Amanda
 Highly-scalable command-line client-server
archiver included with RHEL
 Commertial applications
 Arkeia, Bru, Tivoli, Veritas (client), UNiBACK,
ArcServe
Unit 9
 The X Window System
Xorg: The X11 Server
 Foundation for the Redhat Enterprise Linux
graphical user interface (GUI)
 Open Source implementation of X11
 Client/Server Architecture
 Relies on networking
 IP or Local UNIX domain-sockets
 Designed as one server to many clients
 Highly flexible protocol
Xorg Server Design
 System video hardware I/O Management
 Display, video and input device coordination
 Core server: /usr/X11r6/bin/Xorg
 Enhanced by dynamically loaded modules
 Drivers: ati, nv, mouse, keyboard, etc.
 Extensions: dri, glx and extmod
 Font Rendering
 Native server: xfs
 Fontconfig/Xlf libraries
XOrg Server Configuration
 Typically configured after installation
 Post-install configuration:
 Best results while in runlevel 3!
 system-config-display
 Options:
 --noui
 --reconfig
 Stored in /etc/X11/xorg.conf
XOrg Modularity
 The X server and it’s client may be
individually configured and combined
 Server extensions provide enhanced rendering
capabilities
 To view server capabilities: xdpyinfo
 Display Managers
 gdm, kdm and xdm
 Window Managers
 metacity, kwin and twm
Server and Client Relationship

Window Manager

Application
Display
Manager

Xorg
Console
Server
Xorg in runlevel 3
 Two methods to establish the environment
 /usr/X11R6/bin/xinit
 /usr/X11R6/bin/startx
 Environment configuration
 /etc/X11/xinit/xinitrc and ~/.xinitrc
 /etc/X11/xinit/Xclients and ~/.Xclients
 /etc/sysconfig/desktop
XOrg in runlevel 5
 Environment established by /sbin/init
 Environment configuration
 /etc/inittab
 /etc/X11/prefdm
 /etc/sysconfig/desktop
 DESKTOP defines the window manager
 DIPLAYMANGER defines the display manager
 /etc/X11/xdm/Xsession
 /etc/X11/xinit/xinitrc.d/*
 ~/.xsession or ~/.Xclients
Configuration Utilites
 Server:
 system-config-display, mouseconfig
 Fonts and Typefaces
 xfs, chkfontpath, fc-cache
 Display and Window Managers
 switchdesk, /etc/sysconfig/desktop, gconftool-
2
Remote X sessions
 X protocol communication is unencrypted
 Host-based sessions implemented through
the xhost command
 User-based sessions implemented through
the Xauthority mechanism.
 sshd may automatically install xauth keys
on remote machine
 Tunnels x protocol over secure encrypted ssh
connection
Unit 10
 Advanced Filesystem Managerment
Software RAID Configuration
 Create and define RAID device using
mdadm
 mdadm –C /dev/md0 -1 0 –n 2 /dev/hda5
/dev/hda7
 Format each RAID device with a filesystem
 mke2fs –j /dev/md0
 Test the RAID devices
 mdadm allows you to check the status of
your RAID devices
 mdadm –detail /dev/md0
Software RAID Recovery
 Simulating disk failure
 mdadm /dev/md0 –f /dev/sda1
 Recovering from a software RAID disk
failure
 Replace the failed hard drive and power on
 Reconstruct partitions on the replacement
drive
 mdadm /dev/md0 –a /dev/sda1
 mdadm, /proc/mdstat, and syslog
messages
Converting LVM1 to LVM2
 RHEL4 Uses the LVM2 format for metadata
 More compact
 Supports transactional changes and replication
 Human readable and editable in an emergency
 Existing LVM1 volumes can be converted
to LVM2 with the vgconvert command
 vgconvert –M2 vgo
 Converts the volume group vg0 from LVM1 to
LVM2
Creating Logical Volumes
 Create physical volumes
 pvcreate /dev/hda3
 Assign physical volumes to volume groups
 vgcreate vg0 /dev/hda3
 Create logical volumes from volume
groups
 lvcreate –L 256M –n data vg0
 mke2fs –j /dev/vg0/data
Resizing Logical Volumes
 lvextend and ext2online can extend
mounted ext2/3 filesystems.
 lvextend first grow the logical volume
 You can not shirnk mounted filesystems.
 Physical volumes may be added to or
removed
 vgextend vg0 /dev/sdb1
 pvmode /dev/hda3
 vgreduce vg0 /dev/hda3
The Linux Quota System
 Overview
 Implemented within kernel
 Enabled on a per-filesystem basis
 Individual policies for groups or users
 Limit by number of blocks or inodes
 Implement both soft and hard limits
 Initialization
 Partition mount options: usrquota, grpquota
 Initialize database: quotacheck
The Linux Quota System (cont.)
 Implementation
 Start or stop quotas: quotaon, quotaoff
 Edit quotas directly: edquota username
 From a shell
 setquota username 4086 5120 40 50 /foo
 Define prototypical users:
 edquota –p user1 user2
The Linux Quota System (cont.)
 Reporting
 User inspection : quota
 Quota overviews: repquota
 Miscellaneous utilites: wantquota
Unit 11
 Troubleshooting
Unit 11: Agenda
 Troubleshooting Strategies
 Things to check
 Boot procedures
 Rescue Environment
Troubleshooting
 Treat the problem as a symptom
 Gather data by identifying other problems
 Identify what still works
 From a hypothesis about what is wrong
 Check log files for supporting evidence
 Backup config files before editing them
Things to Check: X
 Never debug X while in runlevel 5!
 Try system-config-display first
 X –probeonly
 Is /home or /tmp full, or has the user
reached a hard quota?
 Is xfs running?
Things to Check : Networking
 Hostname resolution
 dig www.redhat.com
 IP configuration
 ifconfig
 Default gateway
 route –n
 Module specification
 Device activation
Order of the Boot Process
 Bootloader configuration
 Kernel
 /sbin/init
 Starting init
 /etc/rc.d/rc.sysinit
 /etc/rc.d/rc, /etc/rc.d/rc?.d
 Entering runlevel X
/etc/rc.d/rc.local
X
Filesystem Corruption
 Common after crash or improper
shutdown
 ext2 mounted for writing marked “dirty”
 If not mounted or mounted read only, “clean”
 If not mounted and “dirty”, may be corrupted
 Repair requires exhaustive check
 ext3 usually marked “clean”
 Journal indicates if recovery is needed
 Only need to check files recorded in journal
Filesystem recovery
 If / has journal, kernel examines it at boot
 /etc/rc.d/rc.sysinit runs fsck on
filesystems marked in the /etc/fstab
 Fack is a front end to other programs
 A “failed” fsck must be run manually
Recovery Run-Levels
 Pass run-level to init
 On boot from GRUB splash screen
 Runlevel 1
 Process rc.sysinit and rc1.d scripts
 Runlevel s,S or single
 Process only rc.sysinit
 Emergency
 Run sulogin only
Rescue Environment
 Required when root filesystem is
unavailable
 Non-system specific
 Boot from CDROM (boot.iso or CD #1)
 Boot from diskboot.img on USB key
Rescue Environment Utilities
 Disk Maintenance Utilities
 Networking Utilities
 Miscellaneous Utilities
 Logging : /tmp/syslog or
/tmp/anaconda.log
Rescue Environment Details
 Filesystem reconstruction
 Anaconda will ask if filesystems should be
mounted
 Watch for error messages
 /mnt/sysimage/*
 /mnt/source
 $PATH includes hard drive’s directories
 Filesystem nodes
 System-specific device files provided
 Mknod knows major/minor #’s
End of Unit 11
 Questions and Answers
 Summary
 What are some things to check for
 X problems?
 Service problems?
 Networking problems?
 Boot Problems?
 How might you repair an ext2 filesystem?
 What are some alternate boot methods?