# 2 Server Administrator I - RH124

2.1 What is linux ?

Modular operations system where you can add components :

Open Source

• Copyleft lic
  • GNU : General Public License
  • LGNU: Lesser General Public License
• Permissive lic:
  • MIT
  • Simplified BSD
  • Apache Software License

Red Hat contribute and facilitate open source projects, validating the code and support code that Red Hat validate and test.

Sample of Red Hat Products:

* Red Hat Enterprise Linux 8
* Red Hat Open Shift 4
* Red Hat Ansible Automation
* Red Hat Ceph Storage
* Red Hat OpenStack Platform
* Red Hat Virtualization
* Red Hat Gluster Storage

Benefits of open source software:

• Code can survive the loss of orginal developer or distributor
• We can learn from real-world code and develop more effective applications
• Red Hat sponsor and integrate open source projects into the Fedora project and participate in upstream projects

2.2 Command Line

2.2.1 $ and #

Attention on shell when you are on command line:

• Root user : #
• Normal user : $

Sample of using - or --

-all : Shell going to be interpreting all arguments individually

# Shell going to be interpreting all arguments individually  a, l and l
ls -all . 

--all : Shell will interpreting the entire word

# Shell will interpreting the entire word
ls --all .

2.2.2 To login into another computer using ssh

ssh <machine or server>

2.2.3 Executing commands

# List the user that you  login
whoami

# formatting the dates
date +%F

# Executing one command after the other
date +%A; uname -r; whoami

## runner
## 2022-08-02
## Tuesday
## 20.6.0
## runner

Conditional commands execution &&

This is the basic IF statement

The date command must complete successfully than the uname -r command will be running and if execute successfully the whoami command will be executed

# If with and
date +%F && uname -r && whoami

echo 
echo
# If with or
ate +%F || uname -r && whoami

## 2022-08-02
## 20.6.0
## runner
## 
## 
## bash: line 7: ate: command not found
## 20.6.0
## runner

command history

• !! : execute the last command
• !12: execute the command 12 on history
• ctr+r

check type of file

The file command is used to determining the type fo file :

file /etc/issue

echo 
echo

file /bin/bash

## /etc/issue: cannot open `/etc/issue' (No such file or directory)
## 
## 
## /bin/bash: Mach-O universal binary with 2 architectures: [x86_64:Mach-O 64-bit executable x86_64] [arm64e:Mach-O 64-bit executable arm64e]
## /bin/bash (for architecture x86_64): Mach-O 64-bit executable x86_64
## /bin/bash (for architecture arm64e): Mach-O 64-bit executable arm64e

check content of file

• cat or tac
• head or tail
• wc -l : count lines of file

2.3 Managing File system

2.3.1 The file system hierarchy /

Try command tree -d

• / : Top of system file system hierarchy
• /usr : unix system resources, installed software and libraries
  • /usr/bin : regular commands and utilities
• /bin : binaries executable that are usable by normal users
• /sbin : system binaries executable typically used by root user
• /boot : component that are necessary to boot file system, like bootloader called grub 2 and linux kernel
• /dev : device files, represent hardware components
• /etc : extended text configuration configuration files
• /home : home dir for normal users
• /run : runtime data going to be recreate on a reboot
• /var: variable data that should survive a reboot, log files, database and website files
• /root : root user home dir
• /tmp : accommodate temporary files, deleted by 10 days
• /var/tmp : another temp dir , purge every 30 days

2.3.2 Absolute and Relative paths

• Absolute path, is the complete path, no ambiguity, start with / :

#sample 
cat /etc/issue

• Relative path is the path to a file relative the current position, do not start with /

cd /
cat etc/issue

2.3.3 Managing Files

• mkdir -p : create the parents that do not exists

pwd
mkdir -p dirA/dirB/dirc
ls -R dirA

## /Users/runner/work/book-notes-linux-learn/book-notes-linux-learn
## dirB
## 
## dirA/dirB:
## dirc
## 
## dirA/dirB/dirc:

Commands

• cp -r : copy dir
• rm -i : show to you msg to confirm
• rmdir : remove dir
• rm -r : delete dir recursive
• mv : move
• touch : create files

Links

Index Node ((inode_): How files are identified, keep track of: * permissions * ownership * date & time stamps * paths to data on file system

ls -li LICENSE

## 12893901114 -rw-r--r--  1 runner  staff  6556 Aug  2 13:21 LICENSE

• 40652348 : file ID (inode)
• 1 : means one name using this iNode right now

Hard Link

ln LICENSE LIC2
ls -li LIC*

## ln: LIC2: File exists
## 12893901113 -rw-r--r--  1 runner  staff  6556 Aug  2 13:21 LIC2
## 12893901114 -rw-r--r--  1 runner  staff  6556 Aug  2 13:21 LICENSE

• Both LIC2 and LICENSE have the same iNode but the number right now is 2, not able to identify which one was create first, this is a hard link

Soft Link

ln -s README.md README_2.md
ls -li READ*

## ln: README_2.md: File exists
## 12893901115 -rw-r--r--  1 runner  staff  741 Aug  2 13:21 README.md
## 12893901116 lrwxr-xr-x  1 runner  staff    9 Aug  2 13:21 README_2.md -> README.md

• Both have different iNode number, so they are two different files, the permission there are a l meaning link, this is a soft link, if we delete file 3 we have a broken link

# create tmp files
mkdir -p tmp ;cd tmp ;touch 1file 2file 3file 4file able alfa baker bravo cast easy echo _src

# List files that match this case
echo
echo "List files that match this case Xfile"
ls ?file
echo "----"

# list files that begin with a or c or e 
echo
echo "List files that begin with a or c or e "
ls [ace]*
echo "----"


# List files that do not start with a c or e
echo
echo "List files that do not start with a c or e"
ls [^ace]*  # or ls [!ace]*
echo "----"


# List all file that begin with an alphabetical character
echo
echo "List all file that begin with an alphabetical character"
ls [[:alpha:]]*

# List all files that begin with digit
echo
echo "List all files that begin with digi"
ls [[:digit:]]*
echo "----"

# List all files that match digit or alphabetical char
echo
echo "List all files that match digit or alphabetical char"
ls [[:alnum:]]*
echo "----"


# List all files that begin with punctuation 
echo
echo " List all files that begin with punctuation"
ls [[:punct:]]*
echo "----"

# clean
cd ..
rm -rf tmp

## 
## List files that match this case Xfile
## 1file
## 2file
## 3file
## 4file
## ----
## 
## List files that begin with a or c or e 
## able
## alfa
## cast
## easy
## echo
## ----
## 
## List files that do not start with a c or e
## 1file
## 2file
## 3file
## 4file
## _src
## baker
## bravo
## ----
## 
## List all file that begin with an alphabetical character
## able
## alfa
## baker
## bravo
## cast
## easy
## echo
## 
## List all files that begin with digi
## 1file
## 2file
## 3file
## 4file
## ----
## 
## List all files that match digit or alphabetical char
## 1file
## 2file
## 3file
## 4file
## able
## alfa
## baker
## bravo
## cast
## easy
## echo
## ----
## 
##  List all files that begin with punctuation
## _src
## ----

Brace expansion

mkdir -p tmp ;cd tmp
echo {Sun,Mon,Tues,Wednes}day.log

# create dirs
mkdir -p RHEL{6,7,8}; ls RHEL*

# create sequence of files using ..
touch song{1..5}.mp3 ; ls *.mp3

#clean
cd ..
rm -rf tmp

## Sunday.log Monday.log Tuesday.log Wednesday.log
## RHEL6:
## 
## RHEL7:
## 
## RHEL8:
## song1.mp3
## song2.mp3
## song3.mp3
## song4.mp3
## song5.mp3

Variable

SOMETHING=value
echo $SOMETHING


# command substitution
echo "Today is $(date +%A)"

## value
## Today is Tuesday

skell

The contenct of path /etc/skel is automatically copied to all users

2.4 Help

All man pages are on /usr/share/man

Commands

• man
• pinfo

man -k cron

## cron(8)                  - daemon to execute scheduled commands (Vixie Cron)
## crontab(1)               - maintain crontab files for individual users (V3)
## crontab(5)               - tables for driving cron
## DateTime::Locale::en_FM(3pm) - Locale data examples for the English Micronesia (en-FM) locale
## cron(8)                  - daemon to execute scheduled commands (Vixie Cron)
## crontab(1)               - maintain crontab files for individual users (V3)
## crontab(5)               - tables for driving cron

To go direct to a section , belo example of command to go direct to session 5

man 5 crontab

2.5 Text Files

Channels

• 0 : stdin , read only used by keyboard
• 1 : stdout, write only, terminal
• 2 : stderr, write only, terminal
• 3+ : file name, read and write

We can direct errors to a specific file like below I do not have the file or dir abc

ls abc file1 2> errors.log

# output
file1

# on errors.log
ls: cannot access 'abc': No such file or directory

We also can send both to the same file

ls /show /boot &> combine.log
# another option 
# ls /show /boot > combine.log 2>&1

cat combine.log
rm combine.log

## ls: /boot: No such file or directory
## ls: /show: No such file or directory

Another usage when we have an output with lot of errors and can only the the results, all the errors message was sent to /dev/null

find / -iname passwd 2> /dev/null

#output
/usr/share/bash-completion/completions/passwd
/usr/share/doc/passwd
/usr/share/lintian/overrides/passwd
/usr/bin/passwd
/etc/passwd
/etc/pam.d/passwd

A good usage is send successfully output to an output.log file and the errors to errors.log file

ls /shoe /boot >> output.log 2>> errors.log

Send emails with content of file, the < will direct the content of file to email to <user>

mail -s "Subject text" <user>  < file

Pipe

The resolv.conf have 6 lines in total but we can see the lines without comment # and direct the output using pipe |

wc -l /etc/resolv.conf

echo

# check the lines without comment
grep ^[^#] /etc/resolv.conf

echo
# combine the grep with pipe
grep ^[^#] /etc/resolv.conf | wc -l

##       17 /etc/resolv.conf
## 
## nameserver 1.1.1.1
## nameserver 208.67.222.222
## 
##        2

Another way is get the output and save on file using pipe | and tee

find / -iname passwd 2> /dev/null | tee find.out

2.5.1 vim

Modes :

• Insert : i
• Command : default
  • yy : Copy
  • p : paste
  • 5p : paste 5 times
  • dd : delete
  • ZZ : save and quite
  • cw : change word
  • x : delete character
  • r : replace character
  • a : append
• Extend command : :
  • q! : exit without save
  • wq : save and exit
• Visual : v
  • crt + v : block column to manipulate column
  • shift + v: line mode to select line
  • x : delete
  • u : undo

2.5.2 Changing shell

• Setting a editor

# To set
env EDITOR=nano crontab -e

# To unset
expor -n EDITOR

The crotab will open on nano editor instead of vi or vim

Also can set the variable EDITOR to nano and the same will hapens if we call crontab é

# using export
export EDITOR=nano

Some .bash* files and adjust

• All the history are saved on .bash_history
• .bashrc user specific env for example:
  • we can add more lines on history add the variable export HISTFILESIZE=2000
  • add export HISTTIMEFORMAT="%F %T "

2.6 Managing Local Users and Groups

• whoami : Show your user

• id : show your user, groups

• Type of users*

  • Super User root
  • Account users, not used by people
  • Regular users

2.6.1 Users

• /etc/passwd : file with user info

grep bruno /etc/passwd

#output
bruno:x:1000:1000:bruno,,,:/home/bruno:/bin/bash

• User : Bruno

• x : long time ago where password where stored

• 1000 : User ID

• 1000 : group ID

• Comment

• User home dir : /home/bruno:/bin/bash

• The password is stored on /etc/shadow, only root can access this file

  • Sample : 6$CSsXcYG1L/4ZfHr/$2W6evvJahUfzfHpc9X.45Jc6H30E...output omitted..
    • 6 : Hash algorithm used , SHA-512
    • CSsXcYG1L/4ZfHr/ Used by cript info
    • W6evvJahUfzfHpc9X.45Jc6H30E has password

• Another way to list and return user info is using getent command

getent passwd bruno

#output
bruno:x:1000:1000:bruno,,,:/home/bruno:/bin/bash

• useradd : create new user, sample : useradd kano
• userdel : remove user
  • sample 1: userdel kano, but do not remove the data and home dir
  • sample 2: userdel -r kano, delete user and data/home dir
• usermod -c : add comments to user
  • usermod -a -G <user>
    • -a : append to the secondary group membership, if do not use a overwritten the current info
    • -G : group name
  • usermod -g <group> <user>
    • -g : change the primary group info
• usermod -L <user> : Lock a user account
• usermod -U <user> : Unlock a user account

2.6.2 Groups

• /etc/group : file with group info

grep bruno /etc/group 

#output
Show in New Window
/etc/issue: ASCII text


adm:x:4:syslog,bruno
cdrom:x:24:bruno
sudo:x:27:bruno
dip:x:30:bruno
plugdev:x:46:bruno
lpadmin:x:113:bruno
bruno:x:1000:
sambashare:x:130:bruno

• bruno : group name
• x : pwd info that is not used
• 1000 : group id
• Members : of group

Show groups from my users using command groups.

• groupadd : add new group (GID > 1000)
  • -g : specify the GID, groupadd -g 1000 group1
  • -r : create the system group (GID 0-999)
• groupdel : remove group groupdel <group_name>
• groupmod : modify the group
  • -n : change the group name, groupmod -n <new_name> <old_name>

2.6.3 Gaining superuser access

• su : change to root super user, without set the profile, the path still your regular user
• su - : change to root user and set the profile and env of root
• sudo : allow run commands as another user
• sudo visudo : sudoers file
  • Line : if you are member the group wheel you can be logged in from any computer and you can run all cmds as all users

  # Allows people in group wheel to run all commands
  %wheel  ALL=(ALL) ALL  

  • % indicate a group
• sudo -i or sudo su - : assume as root user
• /etc/sudoers.d/<user> : drop config into that on dir /etc/sudores.d as if you are editing the /etc/sudoers

Tip

To create /etc/sudoers.d/admin file and grant all members of admin group total privileges

echo "%admin ALL=(ALL) ALL" >> /etc/sudoers.d/admin

To create just for one user

echo "user ALL=(ALL) ALL" >> /etc/sudoers.d/user

2.6.4 Managing user passwords

We no longer store password on /etc/passwd, the password are stored on /etc/shadow

• chage : used to change the aging info for a user password

  • Sample : chage -m 1 -M 26 -W 4 -I 3 -E (2019-05-31) <user>
    • -m : min wait time before user can change the pwd again, in this case 1 day
    • -M : when they have to changed the pwd, in this case in 26 days need to change
    • -W : warning of 4 days
    • -I : once the pwd expire they have 3 more days to login and change the pwd
    • -E : expiration date

• chage -l : show all the info

• /etc/login.defs : Define info of login such PASS_MAX_DAYS, PASSMIN_DAYS, etc

• Add user with nologin shell

  • useradd -s <user> -s /sbin/nologin

Tip

Change the user and update expiration to more 180 days

chage -E $(date -d + 180days +%Y-%m-%d) <user>

2.7 Controlling access to files

drwxrwxrwx

• d : dir, can be l link, etc
• r | 4 : read
• w | 2 : write
• x | 1 : execute

Owning User : first 3 rwx Owning Group : next 3 Other : last 3

Commands

• chmod : change the permission mod
  • chmod 740 <file or dir> or
  • chmod o+rw <file or dir>
• chown : change ownership
  • chmod user:group <file or dir>
  • chmod :group <file or dir> or
  • chmod user <file or dir>

Tip

Give you read and execute permission do dir but no execute on files inside the dir, i.e, we can list the content but not execute files inside

chmod -R a=rX

2.7.1 Special permissions

Sticky bit : In a collaborative dir you can create files on the dir and only delete files that you have created

• u+s (suid) : Files executes as the user that owns the file, not the user that ran the file

• g+s (guid) : file execute as the group that owner the file, files newly create in directory have their group owner set to match the group owner of the dir.

• o+t (sticky): Users with write access to dir can only remove files that they own

• Sticky bit — directories o+t

  • To set o+t or 1
    • chmod o+t <dir> or
    • chmod 1770 <dir>
  • To remove o-t or 0
    • chmod o-t <dir> or
    • chmod 0770 <dir>

# To set sticky bit
chmod o+t dirA
ls -ld dirA

# To remove sticky bit
#chmod o-t dirA

s : Files created on the dir will have the same owning group of dir

• Set grid to dirs or files g+s
  • To set g+s
    • chmod g+s <dir>
  • To unset g-s
    • chmod g-s <dir>

# To set
chmod g+s dirA
ls -ld dirA

# To remove 
# chmod g-s dirA

• Set uid to files u+s
  • To set u+s
    • chmod u+s <file>
  • To unset u-s
    • chmod u-s <file>

2.7.2 Default permissions - umask

• umask : show umask info
• umask 0000 : set the umask to 0000 or 777 to new dirs

To change the user umask we can update the .bashrc file

0022: * The permissions of files going to be 755

2.8 Monitoring and Managing Linux process

Process state description

2.8.1 Commands to monitor

• top
  • %CPU
  • load average
  • tasks
• ps aux , ps -ef
  • PID : process ID
  • PPID: Parent process ID
  • Time
  • CMD
  • TTY : from where the process is running
• htop

2.8.2 Controlling jobs

sample : gnome-calculator &

Command : * jobs : list all jobs * fg %<job_number> : Bring the job to foreground : * Ctrl+z : suspend/stop the job * ps j : show the info relate jobs * bg %<job_number> : restart job * Ctrl+c : terminate job

2.8.3 Killing process

Signals

kill -l

##  1) SIGHUP    2) SIGINT   3) SIGQUIT  4) SIGILL
##  5) SIGTRAP   6) SIGABRT  7) SIGEMT   8) SIGFPE
##  9) SIGKILL  10) SIGBUS  11) SIGSEGV 12) SIGSYS
## 13) SIGPIPE  14) SIGALRM 15) SIGTERM 16) SIGURG
## 17) SIGSTOP  18) SIGTSTP 19) SIGCONT 20) SIGCHLD
## 21) SIGTTIN  22) SIGTTOU 23) SIGIO   24) SIGXCPU
## 25) SIGXFSZ  26) SIGVTALRM   27) SIGPROF 28) SIGWINCH
## 29) SIGINFO  30) SIGUSR1 31) SIGUSR2 

• default kill -15 <process>
• die right now kill -9 <process>
• stop kill -19 <process>
• continue kill -18 <process>
• kill with -15 several with same name killall <name>
• pkill -t pts/2 : terminate the user logged on pts/2*
  
• pkill -SIGTERM tail : will kill the tail process running
• pstree : display tree view of process
• pgrep -l -u <user> : identify the process that going to be killed by pkill

2.8.4 Monitor process activity

Load Average

• uptime command review the load average

uptime

## 13:22  up 3 mins, 1 user, load averages: 9.20 8.06 3.70

• 1,24 : Last min
• 1,07 : Last 5 min
• 0,93 : Last 15min

To analyze the load average need to know how many CPU’s do we have

lscpu | grep -i 'CPU(s)'

#or
echo
echo "========Number of CPUs=====" : 
cat /proc/cpuinfo | grep "model name" | wc -l

#output

CPU(s):                          8
On-line CPU(s) list:             0-7
NUMA node0 CPU(s):               0-7

========Number of CPUs===== :
8

So 1,24 / 8 is 0,155 , what means that my CPU is busy 15% of the time on last min

On last 15min my cpu is 11% (0,93/8) last 15min and 13%(1,07/8) last 5min.

To know if the CPU is overload the results show be 1.13 what means that my CPU is overload by 13% on that particular time

Another example

#From lscpu, the system has four logical CPUs, so divide
#                                   load average: 2.92, 4.48, 5.20
#divide by number of logical CPUs:                  4     4     4
                                                   ---- ---- ----
                          per-CPU load average:    0.73 1.12 1.30
                              
# This system's load average appears to be decreasing.
# With a load average of 2.92 on four CPUs, all CPUs were in use ~73% of the time.
# During the last 5 minutes, the system was overloaded by ~12%.
# During the last 15 minutes, the system was overloaded by ~30%

• top is another command that can be used to monitor the system
  • k : ask by the PID to be terminated
  • M : sort by memory
  • h : help
  • shift + w : write the top config on /home/<user>/.config/procps/toprc
• On top :
  • PID : process ID
  • PR : Priority
  • VIRT : Virtual memory that process is using
  • RES : Physical RAM used
  • SHR : Shared memory
  • %CPU
  • %MEN
  • Time : how long it is running
  • Command

2.9 Controlling Services and Daemons

The systemd is responsable for initializing the system and uses units that represent daemons: * service : database, web service, tc * target : collection of units * device * socket

systemctl list-units  | head -n 10

# or
# systemctl
# systemctl list-units --type=service
# systemctl list-units --type=target

status

• Status of sshd : systemctl status sshd

#systemctl list-units --type=service

systemctl status bluetooth.service


#output

● bluetooth.service - Bluetooth service
     Loaded: loaded (/lib/systemd/system/bluetooth.service; enabled; vendor preset: enabled)
     Active: active (running) since Sun 2021-08-22 08:50:54 -03; 5 days ago
       Docs: man:bluetoothd(8)
   Main PID: 906 (bluetoothd)
     Status: "Running"
      Tasks: 1 (limit: 18970)
     Memory: 2.7M
     CGroup: /system.slice/bluetooth.service
             └─906 /usr/lib/bluetooth/bluetooth

• Stop : systemctl stop <service>
• Start : systemctl start <service>
• Restart: systemctl restart <service>
• Reload : systemctl reload <service>
• ask : systemctl is-active <service> or ...is-enable..., is-failed
• disable: systemctl disable <service>, the service will not start when the system is started
• list dependency : systemcl list-dependecies <service>
• mask : systemctl mask <service> prevent service to be started
• unmask: systemctl unmask <service>

2.10 Configuring and Securing SSH

On RHEL8 we have OpenSSH which implement not only the SSH daemon but also SSH command line tool

• ~/.ssh/known_hosts stores the fingerprint sent by server to future communication

• To define the StrictHostKeyChecking we can edit the file ~/.ssh/config or /etc/ssh/ssh_config

2.10.1 Configure SSH Key-based Authentication

• private - decrypt
  • When you connect to ssh server using private key the remote machine will generate a challenge (encrypted with your public key), if you are able to decrypt that encrypted challenge then yo are allowed to make a connection
• public - encrypt
  • when you ssh to a server the public key is saved on the SSH server

2.10.2 Create a public and private keypair

1. Issue to start and create the Key : ssh-keygen

2. Enter the file name

3. Create a paraphrase

Result: We going to have a public and private key created

4. Install the key on server

ssh-copy-id -i .ssh/<public key name> <hostname>

5. Test

ssh -i .ssh/<public key name> <hostname>

6. Add the private key to agent

# start the agent
eval $(ssh-agent)  

# add the key
ssh-add .ssh/<public key name>

2.10.3 Customizing OpenSSH Service Config

• Disable the root ability to login on /etc/ssh/sshd_config to modify the daemon config and change the PermitRootLogin to no

Tecniques to avoid password

• Create a sshusers group and configure /etc/ssh/sshd_config 4 allowGroups, or
• Create a private key and public key

2.11 Analysing and Storing Logs

Systemd is the heart of RHEL8 system and need to analyse how it is working

The journal collects messages from several sources (booting, daemons, etc) and we can query using journalctl but this is not persistent by default.

There are another process rsyslog that read syslog and receive systemd-journal and save it on /var/log

Facility is one of the following keywords : auth, authpriv, cron, daemon, kern, lpr, mail, mark, news, security, syslog, user, uucp, local0 to local7.

Priority is one of the following keywords : debug, info, notice, warning, warn, err, error, crit, alert, emerg, panic

• /var/log/messages : Store most of syslog messages

• /var/log/secure : store syslog messages related to security and authentication operations

• rsyslog : service that organize syslog messages into /var/log

• /var/log : directory of syslog files

• /var/log/maillog : store syslog messages related mail server

• /var/log/cron : store syslog messages related to the schedule jobs

• /var/log/boot.log : store console message related to system startup

  • Syslog codes

    • 0 - emerg
    • 1 - alert
    • 2 - crit
    • 3 - err
    • 4 - warning
    • 5 - notice
    • 6 - info
    • 7 - debug

  • The message is organized as facility.priority and we can configure to direct it to particular file like authpriv.notice and send to /var/log/foo

  • Config file : /etc/rsyslog.conf, where we can add all rules

    • Avoid edit the main config file /etc/rsyslog.conf and use drop-in directory /etc/rsyslog.d

  • Log rotate prune framework

    • /etc/logrotate.conf : Config file
    • /etc/logrotate.d : drop-in dir

  • Manual messages to syslog sample :

    logger -p local7.notice "log entry created on host"

2.11.1 Reviewing System Hournal Entries

Journal is not persistent and can find the logs on /run/log/journal

• systemd-journald - command journalctl
  • journalctl -r : reverse the log message, last page of msg show first
  • journalctl -u <unit> : show logs from unit such as sshd.service
  • journalctl -u <unit> --since today : show today message of particular unit
  • journalctl -u <unit> --since "2019-04-15 09:00:00" --until "2019-04-15 11:00:00" -p warning : Show the warning message of unit in particular time
  • journalctl --since "2019-04-15 09:00:00" --until "2019-04-15 11:00:00" -p warning : Show ALL warning message in particular time
  • since request all warning we also going to see the err, crit, alert and emerg messages
  • journalctl -p err : show the errors
  • journalctl -f : to monitor the system
  • journalctl -n 50 : to show the last 50 messages
  • journalctl _SYSTEM_UNIT=sshd.service _PID=xx : will show to you message about this unit and PID
  • journalctl -b -1 : show the data related one boot if the journal is persistent

2.11.2 Preserving the system journal

By default the journal is not persistent on /run/log/journal and after reboot we lose all data. To preserve the journal we need to create the dir /var/log/journal, also by default journal are not allow to get more than 10% of file system or leave less than 15% of the file system free. The configuration are on /etc/journald.conf

2.11.3 Maitaining Accurate time

Have system in sync is important

Checking time

timedatectl

#output

               Local time: Fri 2021-08-27 13:24:50 -03   
           Universal time: Fri 2021-08-27 16:24:50 UTC   
                 RTC time: Fri 2021-08-27 16:24:50       
                Time zone: America/Sao_Paulo (-03, -0300)
System clock synchronized: yes                           
              NTP service: n/a                           
          RTC in local TZ: no  

To list the timezones we can use the command timedatectl list-timezones and set with command:

timedatectl set-timezone <timezone> 

The system will use NTP (Network Time Protocol) to synchronize the time with a machine to perform that we can set it to True

timedatectl set-ntp true

RedHat 7 and 8 use the chronyd.service to synchronize the time and inside of /etc/chrony.conf we going to find the list of servers that are in sync and can check the sources with chronyc sources

2.12 Managing Networking

TCP/IP layers :

• Application : How clients communicate across plataforms, sample Web browser talking withWeb server
  • SSH : remote login
  • HTPS : web
  • NFS or CIFS : file share
  • SMTP : emal
• Transport : How packets are sent and received like
  • TCP : connection-oriented bi-direction form of guaranteed messaging
  • UDP : connectionless unidirectional and non-guaranteed messaging
• Internet : Specify how packets are routed acrosss network
• Link layers : physical askpects of networking
  • MAC

Devices :

The name dependes on where and how the device is connected

• EN : Ethernet devices
  • ENO1 : Onboard Ethernet Interface with index number 1
  • ENS3 : Ethernet device in hotplug slot 3
• WLAN : Wireless LAN devices
• WWAN : Wireless AN devices
• WLP4S0 : PCI buss number 4 connect on slot 0

2.12.1 TCPIP

IPV4

• IP
  • IP Address : 32bits divided up into 4 octets, this is the interface unique identity on the network
  • Subnet mask : divide the IP address into the host portion as well as the network portion and that is used to facilitate routing
    • Notation : 255.255.255.0 or /24

IPV6

“New implementation of IP” : Allow every single person to have its own unique IP Address

• IPV6 is 128-bit number with 8 colon separated groups of 4 hexadecimal nibbles, also use subnet mask and normally make use of /64, i.e, 64 bits

• Common ipv6 address

  • ::1/28 : localhost
  • ::/0 : default route
  • fe80::/10 : link-local (start with fe80), those link are not routable, when we have IPV6 enable the link-local will be allocated automatically and allow to communicate with other machines using IPV6 on the same local segment

ip a

#output
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: enp3s0f1: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc fq_codel state DOWN group default qlen 1000
    link/ether 80:fa:5b:4d:16:6d brd ff:ff:ff:ff:ff:ff
3: wlp4s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
    link/ether f8:94:c2:74:dc:04 brd ff:ff:ff:ff:ff:ff
    inet 192.168.15.8/24 brd 192.168.15.255 scope global dynamic noprefixroute wlp4s0
       valid_lft 35492sec preferred_lft 35492sec
    inet6 fe80::19da:6be7:d851:7ea9/64 scope link noprefixroute 
       valid_lft forever preferred_lft forever

• The IPV6 : inet6 fe80::19da:6be7:d851:7ea9/64 scope link
• Interface : enp3s0f1
• MAC : link/ether 80:fa:5b:4d:16:6d brd ff:ff:ff:ff:ff:ff
• IPV4 : inet 192.168.15.8/24 brd 192.168.15.255 scope global dynamic

Commands

• PING tool for IPV6 : ping6 <ipv6>%<interface>
• Show IPv4 table : ip route
• Show IPv6 table : ip -6 route
• Show specific interface : ip a s <interface
• Properties of interface : ip link show
• Link statistics : ip -s link show

Files

• /etc/hosts : There are number of names that resolve ip address
• /etc/services : List of commonly used services
• /etc/resolv.conf : Wejre DNS service are defined

2.12.2 Validating Network Configuration

Commands to review IP config

ip a

Tools

• nmcli : command line tool to manage network
• nmtui : graphical tool to manager network
• ip : to list ip config
• netstat : check ports and process
• tracepath : work similar traceourte tracepath access.redhat.com
• ss : information about process that are opening up listening
  • sample : ss -plunt
  • sample2 : ss -lt : What is listening, t for TCP

2.12.3 Configure Networking from the Command Line

Tool: nmcli an interface to the Network Manager daemon that supoprt tab complition

• Config on : /etc/sysconfig/network-scripts

• High level of how network is configured :

nmcli

#output
Show in New Window
/etc/issue: ASCII text


/bin/bash: ELF 64-bit LSB shared object, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, BuildID[sha1]=a6cb40078351e05121d46daa768e271846d5cc54, for GNU/Linux 3.2.0, stripped
Show in New Window
40652348 -rw-rw-r-- 2 bruno bruno 6556 Aug 12 21:50 LICENSE
Show in New Window
wlp4s0: connected to Auto ALMAX-BRUNO
    "Intel Wireless-AC 3168NGW"
    wifi (iwlwifi), F8:94:C2:74:DC:04, hw, mtu 1500
    ip4 default, ip6 default
    inet4 192.168.15.8/24
    route4 0.0.0.0/0
    route4 169.254.0.0/16
    route4 192.168.15.0/24
    inet6 fe80::19da:6be7:d851:7ea9/64
    route6 fe80::/64
    route6 ::/0

p2p-dev-wlp4s0: disconnected
    "p2p-dev-wlp4s0"
    wifi-p2p, hw

enp3s0f1: unavailable
    "Realtek RTL8111/8168/8411"
    ethernet (r8169), 80:FA:5B:4D:16:6D, hw, mtu 1500

lo: unmanaged
    "lo"
    loopback (unknown), 00:00:00:00:00:00, sw, mtu 65536

DNS configuration:
    servers: 192.168.15.1
    interface: wlp4s0

    servers: fe80::aec6:62ff:fefc:7110
    interface: wlp4s0

Use "nmcli device show" to get complete information about known devices and
"nmcli connection show" to get an overview on active connection profiles.

Consult nmcli(1) and nmcli-examples(7) manual pages for complete usage details.

• Connection show :

nmcli connection show


#output
NAME                UUID                                  TYPE      DEVICE 
Auto ALMAX-BRUNO    72e81729-a7a7-473b-8577-013339a4420b  wifi      wlp4s0 
Wired connection 1  44945eb1-1b88-305b-84c0-c0c13d340a0a  ethernet  --     

• Details about connections : nmcli connection show "Auto ALMAX-BRUNO"

• Sample of modify profile ipv4.dns

# modify the dns 
ncli con mod <connection name> ipv4.dns <New DNS Name>

# or append, in this case we will have two values of dns
ncli con mod <connection name> +ipv4.dns <New DNS Name>

# grep
nmcli con show <name> | grep dns

# reinitialize the profile
ncli con up <connection name>

# check /etc/resolv.conf
cat /etc/resolv.conf

• Status:

nmcli dev status

#output
DEVICE          TYPE      STATE         CONNECTION       
wlp4s0          wifi      connected     Auto ALMAX-BRUNO 
p2p-dev-wlp4s0  wifi-p2p  disconnected  --               
enp3s0f1        ethernet  unavailable   --               
lo              loopback  unmanaged     -- 

• Add conn

sample

nmcli con add con-name <name of connection> type ethernet ifname <name of connection>

• Up and Disconnect

# Activate connection interface
nmcli con up <connection_name>

# disconect interface
nmcli dev dis <device>

• Delete , reload and modify

# Delete
nmcli con del <device>

# reload
nmcli con reload

# modify
nmcli con mod <name>

• List

# show NetworkManager of all Net Interface
nmcli dev status

# show connections
nmcli con show

# show config
nmcli con show <name>

2.12.4 Editing Network Configuration Files

Files :

We can modify the config file and perform reload nmcli con reload and nmcli con up <name> or use the nmcli commands

• Every Device have a config file into /etc/sysconfig/network-scripts/ifcfg-*, where there are the hardware address, ipaddr, gateway, domain, etc

Sample after update ifcf-* file

nmcli con reload
mcli con down "static-ens3"
nmcli con up "static-ens3"

2.12.5 Configuring Hot Names and Name Resolution

• Command

  • hostname
  • hostnamectl
  • hostnamectl status

hostnamectl status


#output
   Static hostname: turing
         Icon name: computer-laptop
           Chassis: laptop
        Machine ID: bbada5cfcc9a4a80ac8dc5ce6d9f53aa
           Boot ID: cf3c63ed65874b238e2f825055abbf1f
  Operating System: Linux Mint 20.2
            Kernel: Linux 5.4.0-81-generic
      Architecture: x86-64

• To update hostname

hostnamectl set-hostname <new_name>

# check
cat /etc/hostname

Name resolution are om /etc/hosts , the order of resolution are on /etc/nsswitch.conf

• Adding additional DNS search and IP Address for IPv4 similar we can perform for IPv6

sample

# add 
nmcli con mod <connection name> +ipv4.dns <IP> +ipv4.dns-search <name>

# up the connection
nmcli con up <connection name>

# check
cat /etc/resolv.conf

2.13 Archiving and transferring files

2.13.1 TAR

• Create tar file

tar -cf etc.tar /etc

• Test

tar -tf etc.tar

• Extract

tar -xf etc.tar

# Extract one file
tar -xf etc.tar etc/hosts

compression and tar

• Create tar file and compress with gzip2

tar -czf etc-backup-$(date +%F).tar.gz /etc

• Create a tar file and compress with bzip2

tar -cjf etc-backup-$(date +%F).tar.bz /etc

• Create a tar file and compress with Xzip

tar -cJf etc-backup-$(date +%F).tar.xz /etc

2.13.2 SCP

Securely transfer files between systems

• Copy dir and files from another system

copy xf dir from servera to current dir ‘.’

scp -r user@servera:/xf . 

• Copy without password

# create keygen
ssh-keygen -N ''

# copy id
ssh-copy-id servera

# run scp command
scp -r user@servera:/xf . 

2.13.3 SFTP

# connect
sftp user@servera

# list
sftp> ls

# create dir
mkdir backup
cd backup

# local change dir
sftp> lcd /etc

# upload hosts file
sftp> put hosts

2.13.4 Synchronizing Files Between System

• Transfer using rsync

rsync -Par servera:/xf . 

If we update the data into servera on xf dir next time we perform rsync it will perform incremental receiving and download only the difference

2.14 Intalling and updating software packages

To have the Red Hat software update and donwload the bins from RedHat vendor we need a subscription, there are 4 basics elements :

1. You have to register the server with Red Hat or in fact a satellite server
2. After that need to subscribe the server to entitle it to update
3. Enable repositories
4. Go to lifecycle manamgent to track and manage entitlements via portal or subscription asset manager tool

Commands

• Status : subscription-manager status
• Register : subscription-manager register
• Attach to subscription : subscription-manager attach --auto
• Enable repos : subscription-manager repos --disable='*'--enable='repos_name'

2.14.1 Package manager

rpm : Red Hat Package Manager is a popular format for installing software

• rpm database : keep track of what software is installed and versions

  • Query the rpm database to list all packages installed

rpm -qa

• Query a particular package

rpm -q <package>

#or 

rpm -qi <package>

• List files associate with packages

rpm -ql <package>

• Show the config files

rpm -qc <package>

• Show de documentation

rpm -qd <package>

• Show the script that going to be executed

rpm -q -p --scripts <package>.rpm

• Install Limitations : Cannot manage dependency
  

rpm -i <package>

• Which package provide a particular file

rpm -qf <file>

• Download using yumdownloader

yumdownloader <package>

• Query on rpm file, listing a list of files provided by this package

rpm -qpl <file>.rpm

• Query the config files into the rpm file

rpm -qpc file.rpm

• Extract all the files and dirs that are inside a RPM file using rpm2cio

rpm2cpio <file>.rpm | cpio -duim

2.14.2 Inslalling and Update Software Packages with Yum

• Search or get info on current repos

yum search <pacakge>

yum info <pacakge>

• Query RPM files without have to download using repoquery

repoquery -l <package>

• Find out what package provide a particular file

yum provides <file like /etc/fstab>

yum profides *bin/authconfig

• Install yum resolve all dependencies

sudo yum install <package>

#or

sudo yum localinstall <package>.rpm

• Update

sudo yum update <package>

• Remove Tip : do not use -y to review what going to be removed

sudo yum remove <package>

• Check groups

yum group list

• Check packages associate with a group for example Development Tools group

yum group info "Development Tool"

• Install in a particular group

yum group install "Development Tool

• Log can be checked on /var/log/dnf.rpm.log and we can see the history

yum history

• We can undo a transaction on history

sudo yum history undo <number_of_line>

Tip

1. List if the package are installed

sudo yum list httpd mod_ssl

2. Install

^list^-y install

2.14.3 Enabling Yum Software Repositories

• List the repositories that we are connected and repo definition path are on /etc/yum.repos.d/

yum repolist all

• Enable and disable repos using yum-config-manager

yum-config-manager --disable <repo_name>

yum-config-manager --enable <repo_name>

• List subscription manager repos list

subscription-manager-repos --list

2.14.4 Managing Packages Module Streams

Module is a set of RPM packages

• List modules and get info

yum module list <package/module like perl>


yum module info <package/module like perl>


# List installed modules
yum module list --installed

• Install module stream

yum module install <package/module like perl>

• Remove stream

yum module remove <package/module like perl>

• Disable stream

yum module disable perl

• To add we can use the structure

[BASE]
name=Base
baseurl=httpd://......
gpgcheck=0

• Also if we have config-manager

yum config-manager --add-repo <URL>

2.15 Accessing Linux File Systems

• df -h : overview of various file systems
• blkid /dev/vda1 : show the UUID (universally-unique identifier) of file system
• findmnt : tree overview of file system starting at roots
• lsblk : show overview of the various block devices like /dev/vda or /dev/sda
• lsblk -fp /dev/vdb check the UUID

lsblk

#Output

NAME   MAJ:MIN RM   SIZE RO TYPE MOUNTPOINT
sda      8:0    0 931,5G  0 disk 
├─sda1   8:1    0   512M  0 part /boot/efi
├─sda2   8:2    0 915,1G  0 part /run/timeshift/backup
└─sda3   8:3    0  15,9G  0 part [SWAP]

• Check the entire disk

$ls -l /dev/sda
brw-rw---- 1 root disk 8, 0 Aug 22 08:50 /dev/sda
[SWAP]

• b : Indicate that this is a block based device

• du -sh : check space that has being occupied by a directory

  • s : summary
  • h : human read format

$du -sh /var/log
1,3G    /var/log

2.15.1 Mount and Unmounting File systems

To Mount

1. Check the UUID of block device using blkid /dev/vb1 for instance

2. Create directory : mkdir -p <path>

3. Mount : mount UUID="XXX" /<path>

4. Check the return code echo $? (0 means successfully)

5. Check with df -h or mount command

To Unmounting

• umount </dev/vdb1>

2.15.2 Locating Files on the System

find

Basic find syntax :

• find <where> <how> <what is lokking for>

Some examples :

• find / -name ssh_config : looking for particular name
• find / -iname ssh_config: using i going to use case insensitive search
• find /usr -iname "*.pdf" : looking all files that end with pdf
• find / -user <user_name> : find files for a particular user
• find / -user <user_name> -delete : find and delete
• find / -type f -user rick -size 10M : find files from user rick with 10MB (+10M more than 10MB , -10M less than 10MB)
• find /home -size +10M -exec ls -lh {} \; : find on home files more than 10MB and list with ls -lh
• find /home -size +10M iname "*.mkv" -exec rm -f {} \; : find files on home with more than 10MB ending with .mkv and execute rm -f on each file, similar find /home -size +10M iname "*.mkv" -delete`
• find /home -type f -perm /111 : find files on home that have execute permission (111)
• find /home -min -60 : find files modified in last 60min
• find / -user <user_name> -exec -cp {} /<path_copy>/ \;

locate

pre-req : run the command updatedb to create an index of the files on file system

• locate <file> : basic locate a file

2.16 Analysing Server and Getting Support

cockpit service used to analysing and managing remote servers, this is a web user interface that use TCP 9090

• Enable : systemctl enable --now cokpit.socket
• Reload firewall daemon and set permanent : firewall-cm --add-service cockpit --permanent and firewall-cmd --reload
• link : https://:9090/system

2.16.1 Deteting and Resolving issues with Red Hat Insights

It is a form of AI which is available as a software as a service

Install Insights

1. Register and attach the server on subscription-manager

sudo subscription-manager register --auto-attach

2. Install client Insights and register

sudo yum install -y insights-client 

sudo insights-client --register

3. Go to insights dashboard cloud.readhat.com/insight

2.17 Extra

2.17.1 Create 8G file disk

1. Use dd command to create 8G disk file

dd if=/dev/zero of=</path/diskfile.img> bs=1M count=8192

2. create loop device

loseup -fP diskfile.img

3. list the device

losetup -a

4.Can use new diskfile as a device

fdisk /dev/loop0

2.17.2 Find users wit Set UID

find / -perm /4000 -exec ls -l {} \ 2> /dev/null

2.17.3 Show dump device

xxd -l 512 /dev/sda | less

2.17.4 Configure labels of FS

tune2fs -L <label> /dev/sda<number>