Linux Kernel
The Linux kernel is the core of the Linux operating system, acting as a bridge between the hardware and software. It interacts with hardware devices through drivers, known as modules, which are essential for the system's functionality.
These drivers include:
- initramfs: The initial RAM drive loaded when Linux boots up.
- systemd-udevd: Recognizes external hardware, such as thumb drives.
To manually load the drivers, we can use modprobe. Once the kernel is loaded with all the drivers, you can access it through the shell. Commands typed into the shell send syscalls to the kernel.
Kernel Modules and modprobe
Kernel modules are essential components that provide the necessary drivers for the system's hardware. They can be loaded and unloaded dynamically, allowing for flexibility and customization of the system.
- Linux drivers are implemented as kernel modules.
- Most are loadeded automatically through
initramfsorsystemd-udevd - Use
modprobeto manually load a kernel module - To load specific modules, edit the `/etc/modprobe.conf' file.
Loading the Module
Loading a module is necessary when a particular hardware device is not automatically recognized by the system or when you need to use a specific feature that is not loaded by default.
You can check if a module is loaded using lsmod and load it using modprobe if it isn't.
lsmod | grep vfat
modprobe vfat
Check for the module again:
$ lsmod | grep vfat
vfat 20480 0
fat 81920 1 vfat
Unloading the Module
Unloading a module is useful when you no longer need a specific driver or want to free up system resources. It helps in maintaining a clean and efficient system environment.
To unload a module, use the modprobe -r command.
$ lsmod | grep vfat
vfat 20480 0
fat 81920 1 vfat
modprobe -r vfat
Check once again:
$ lsmod | grep vfat
Changing Module Parameters
Changing module parameters allows you to customize the behavior of the kernel modules to suit your specific needs. This can include adjusting settings for performance, compatibility, or functionality.
To change module parameters, modify the configuration files.
$ ll /etc/mod
modprobe.d/ modules-load.d/
$ ll /etc/modprobe.d/
total 20
-rw-r--r--. 1 root root 18 May 4 2021 blacklist-nouveau.conf
-rw-r--r--. 1 root root 747 Jul 28 19:19 lockd.conf
-rw-r--r--. 1 root root 1004 May 10 2021 mlx4.conf
-rw-r--r--. 1 root root 92 May 14 2021 truescale.conf
-rw-r--r--. 1 root root 674 Jul 21 18:41 tuned.conf
After modifying the config file, load the module again using modprobe to reflect the changes.
Tuning Kernel Behavior
Tuning the kernel involves adjusting settings to optimize system performance, security, and functionality. The /proc directory contains information about running processes and system settings, allowing for dynamic adjustments.
The /proc directory is a pseudo-filesystem that provides an interface to kernel data structures. It allows users and administrators to query the system and make runtime changes to kernel parameters.
/procprovides access to kernel information, such as PID directories and status files.- Write the parameters to
/etc/sysctl.confto make them persistent.
To display information about filesystems that are mounted on directories related to the proc filesystem:
$ mount | grep proc
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
systemd-1 on /proc/sys/fs/binfmt_misc type autofs (rw,relatime,fd=37,pgrp=1,timeout=0,minproto=5,maxproto=5,direct,pipe_ino=2903)
/proc/meminfo provides detailed information on what's happening in memory, which can be useful for monitoring and debugging.
$ cat /proc/meminfo
MemTotal: 3918724 kB
MemFree: 1816232 kB
MemAvailable: 2543220 kB
Buffers: 265312 kB
Cached: 625512 kB
SwapCached: 1116 kB
Active: 568072 kB
Inactive: 1003320 kB
Active(anon): 2648 kB
Inactive(anon): 681140 kB
Active(file): 565424 kB
Inactive(file): 322180 kB
The tuning interface is in /proc/sys, which allows for dynamic adjustments to various system parameters.
$ ll /proc/sys
total 0
dr-xr-xr-x. 1 root root 0 Jan 3 07:49 abi
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 crypto
dr-xr-xr-x. 1 root root 0 Jan 3 07:49 debug
dr-xr-xr-x. 1 root root 0 Jan 3 07:49 dev
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 fs
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 kernel
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 net
dr-xr-xr-x. 1 root root 0 Jan 3 07:49 sunrpc
dr-xr-xr-x. 1 root root 0 Jan 3 07:49 user
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 vm
IP Settings
You can navigate through the /proc/sys/net/ directory to access and modify network-related settings.
$ ll /proc/sys/net/
total 0
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 core
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 ipv4
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 ipv6
dr-xr-xr-x. 1 root root 0 Jan 3 07:50 mptcp
dr-xr-xr-x. 1 root root 0 Jan 3 07:50 netfilter
dr-xr-xr-x. 1 root root 0 Jan 2 16:19 unix
$ ll /proc/sys/net/ipv4/ip*
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_default_ttl
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_dynaddr
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_early_demux
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_forward
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_forward_update_priority
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_forward_use_pmtu
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ipfrag_high_thresh
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ipfrag_low_thresh
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ipfrag_max_dist
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ipfrag_secret_interval
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ipfrag_time
-rw-r--r--. 1 root root 0 Jan 3 00:00 /proc/sys/net/ipv4/ip_local_port_range
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_local_reserved_ports
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_nonlocal_bind
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_no_pmtu_disc
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_unprivileged_port_start
Changing ip_forward Setting
The ip_forward setting determines whether the system can forward packets between network interfaces. By default, it is set to 0 (disabled).
$ ll /proc/sys/net/ipv4/ip_forward
-rw-r--r--. 1 root root 0 Jan 3 07:50 /proc/sys/net/ipv4/ip_forward
$ cat /proc/sys/net/ipv4/ip_forward
0
If we want our machine to act as a router and be able to forward packets, we can set ip_forward to 1..
echo 1 > /proc/sys/net/ipv4/ip_forward
$ cat /proc/sys/net/ipv4/ip_forward
1
Verify the current setting:
[root@server ~]# sysctl -a | grep ip_forward
net.ipv4.ip_forward = 1
net.ipv4.ip_forward_update_priority = 1
net.ipv4.ip_forward_use_pmtu = 0
Note that these changes will be wiped out when system is restarted. To make them persistent across reboots, they need to be added to /etc/sysctl.conf.
$ ll /etc/sysctl.
sysctl.conf sysctl.d/
$ ll /etc/sysctl.conf
-rw-r--r--. 1 root root 449 Dec 10 09:31 /etc/sysctl.conf
Edit /etc/sysctl.conf:
$ vim /etc/sysctl.conf
# sysctl settings are defined through files in
# /usr/lib/sysctl.d/, /run/sysctl.d/, and /etc/sysctl.d/.
#
# Vendor settings live in /usr/lib/sysctl.d/.
# To override a whole file, create a new file with the same name in
# /etc/sysctl.d/ and put new settings there. To override
# only specific settings, add a file with a lexically later
# name in /etc/sysctl.d/ and put new settings there.
#
# For more information, see sysctl.conf(5) and sysctl.d(5).
#
net.ipv4.ip_forward = 1
Updating the Kernel
Updating the kernel ensures you have the latest features and security patches. This is crucial for maintaining system security and stability.
- Linux kernels are not technically updated, a new kernel is install beside the old one.
- This allows sysadmins to boot the old kernels in case anything goes wrong.
Use either of the command below to update the kernel:
yum update kernel
yum install kernel
By keeping your kernel up to date, you benefit from performance improvements, bug fixes, and enhanced hardware support, which collectively contribute to a more secure and efficient system.