Advanced System Administration: Git, PAM, and Syncthing Implementation

Desklib provides past papers and solved assignments for students. This project covers Git, PAM, and Syncthing.

Table of Contents
Solution1....................................................................................................................................2
GIT.........................................................................................................................................2
Process for GIT......................................................................................................................2
Initialising a Git repository for an existing code....................................................................2
Cloning a repository...............................................................................................................3
Adding new files to the repository.........................................................................................3
Recovering from repository...................................................................................................3
Checking status of the local user Git repository....................................................................3
View change history of the local Git repository....................................................................3
Pushing changes back to master Git repository.....................................................................3
Pull changes from master Git repository................................................................................3
Set permissions to the repository...........................................................................................4
Solution 2...................................................................................................................................5
Pluggable Authentication Module (PAM) System................................................................5
Implementation of PAM........................................................................................................6
Solution 3...................................................................................................................................8
Installing Syncthing tool............................................................................................................8
Configuration of Syncthing tool.................................................................................................8
Solution 4.................................................................................................................................10
References................................................................................................................................11

Solution1
GIT
GIT is an Open Source Distributed Virtual Control System which was developed by Linus
Torvalds in 2005 who is also the developer of the Linux operating system kernel. It is the
mostly used control system on the planet. A large number of projects are dependent upon Git
for version control. Software engineers working with Git are widely recognised and it works
with a variety of operating systems and development environments. It is a content tracker
used to store content and supports other features as well in order to make the process of
storing and sharing code efficient and easy. The version control system of Git enables
addition and editing of code to the system in parallel to other developers working with the
same code by maintaining a history of the changes on the code. Git uses a remote repository
and the code is stored on multiple locations i.e. on the server and on a local repository which
is the developer’s machine. The version control system of Git enables many developers to
work on the same code in parallel without any conflicts. All the objects under the Git
repository get secured with an algorithm called SHA1 which is a security hashing algorithm.
Git provides support for tagging and branching as first-class citizens which was not being
provided with SVN and the changes which affect the branches and tags also get stored with
the change history, this is something which is not available with all version control systems.
Git is can be difficult for some users and uses different terminology than SVN or CVS but if
leant once, it can be very useful for increased development speeds. (Blischak et. al, 2016)
Process for GIT
Below are steps which are taken for any basic job with GIT-
 A new repository is created with “git init” after the creation of a new directory
 A working copy of the local repository can be created with “git
clone/path/to/repository” and the command on the remote server would be “git clone
username@host:/path/to/repository”
 The repository can be cloned using “git clone [repo-url]”
 Changes can be proposed with “git add <filename>” “git add”
 To commit these changes “git commit -m "Commit message"” is used
 For pushing and editing in the repository “git pull [branch-name]” & “git push” are
used
Initialising a Git repository for an existing code
 Create a local directory “/home/Planets/ktransit”
 Initialise the code using the “Git init” command
 Commitment to the history can be made with the “git add path/to/file/myfile.ext”
command
 When the commitment is made, the code needs to be uploaded to the main repository
and then initialised which can be done by the command “git commit, git -a commit”

Cloning a repository
The cloning of an existing master Git repository to a local user Git repository is done using
the SSL codes. We can clone a repository using the “git clone <url>” command and to clone
the contents in the local directory use “git clone proto://url//repo.git localdir”
To clone the repository use “git clone /home/Planets/ktransit my-ktransit-clone” command.
Adding new files to the repository
To add new and modified files to the local user Git repository, first of all commitment is
made to make changes to the main files in the repository with the commit command. After
that the files are added with “git add /path /to /file /myfile. ext” command.
Recovering from repository
To recover a particular version of a file from the local Git repository, use “git checkout
{commit -sha | branch} /path /to /file /myfile.ext” command
Checking status of the local user Git repository
The “git status” command can be used to check the state of the repository. It verifies the
status as well as reports if there is anything else left to commit.
View change history of the local Git repository
The changed history of the local Git repository can be viewed using the “git log” command.
–p or –patch can be added to this command to see the patch output. Adding a number such as
-2 to the command would only display the last two changes.
Pushing changes back to master Git repository
To push the commits made on the local repository to master repository use the “git push”
command which takes a maximum of two arguments which are the remote name and the
branch name. The complete command thus would be “git push <remote name> <branch
name>”
Pull changes from master Git repository
Similar to the push command, changes can be pulled from the master repository into the local
repository using the command “git pull <remote name>”

Set permissions to the repository
In order to enable cloning for other developers in the group we can use “chmod -R g+swX
ktransit” command. Here, the write permission is represented by w, inherit state by s and
execute permission by X.

Solution 2
Pluggable Authentication Module (PAM) System
Support for various types of authentication systems is provided by a framework which is
known as the Pluggable Authentication Module (PAM). Sun Microsystems developed these
modules and PAM is supported in 64 and 32 bit systems on Unix, Linux, Solaris as well as
AIX. The PAM system for Linux has been developed from Unix PAM architecture and
provides dynamic authentication support for services and applications. PAM enables the
administrators to use different authentication modes as per the requirement and multiple
modules can be used one after another to add multiple layers of security to increase the level
of authentication before the access is granted. Sessions, passwords, accounts etc. can be
managed by the use of these modules. A documented library is provided by PAM and this
makes it possible for the developers to develop applications without them having to create
authentication schemes of their own. Earlier, super user – su, password – passwd, xlock and
login were used for authentication purposes. The Pluggable Authentication Module has a
variety of uses and provides functionalities of administration and control in an organized
manner.
The Linux PAM divides the jobs of verification and authentication into independent groups-
 Validation of the account is processed and checked by the account module. It also
covers other conditions such as if the user has service access, time, expiry of account
etc.
 Identity of the user is checked by the authentication module
 All the passwords are managed by password management module. This includes
password updating and working with the authentication module to generate better
passwords
 After the user is authenticated, a session is started which is managed under the session
management module
PAM can also be used to set up an environment which can incorporate validation of
customers and this way, the clients can also work with the system. The framework directory
can be updated and improved if PAM is learnt and understood by the developers. (O’Reilly |
Safari, 2019)

Figure 1: PAM Framework
The above figure shows the basic PAM framework.
With the use of PAM, sessions can be managed efficiently and also provides functions like
setting and alert for the day or setting up other basic things in the environment.
PAM can be incorporated easily with the boot loaders and can also work well in cases where
authentication of the customer is required along with other customer related functions.
Clients can work along in the environment and when the user signs out the PAM cuts services
off. PAM can be implemented at many other places to develop a similar database for client
log in system and can be used with all programming interfaces.
Implementation of PAM
The user can check the login service by “ldd / bin / login | grep libpam” command. This
command must be executed properly to check for the login service.
This command outputs the shared libraries which were used in the PAM system and the
output for the same is shown:
Figure 2: PAM implementation
By confirming the login entity for the main authentication process, the outputs describe the
shared libraries. PAM is also a good way to perform certain sessions primarily based on
exercises, such as setting unique session parameters and displaying the day's message for an
instance, asset usage features and in addition, installing the working register set when client

responds to the framework or the software. After the user verification process is complete, the
job is carried out further.
There is a strong existence of main content that can be associated with the development of
PAM. The PAM is able to use the baselines of the module that are effectively the
consequence of login service. There are optional and authentic services that are very effective
in completion of the login service, which later on helps the config file. (Tldp.org, 2019)
Below are some important statements in the config file
auth—optional—painfaildelay.so—delay=3000000
session[success=ok new_autlitpk_recid=ok ignore user_unknown=bad default=die] pam_sectuity.so
auth—requisite—pam_nologin.so
session[success=ok ignore=ignore module_unknown=ignore default=bad]-
pam_selinux.so—close
session—required—pam_env.so—readenr=1
auth—optional—pam aroup.so
session—required—pam_limits.so
session—required—pam_lastiog.so auth—optional—pam_exec.so—type=open_sessionsdoubt
/bin/uname -snryin session—required—pam_motd.so
session—required—pam_mail.so standard
session—required—pam_loginuid.so
session[success=ok ignore=ignore module_unknown=ignore
default=bad]pam_selinux.so—open
Control flags help to check status and after the flag is checked, no root instance is used to
consider it later.
The essential contents related with PAM's growth are the specifications of the module that
are essentially used in the login services. The effective optional and authentic services are
generally considered while completing the login services in order to support the config file.
In session organization, the use of module parameters is effective in combination with the
control flags. These control flags are the ones that are taken into account in the status of
different activities using different no root instances for future purposes.

Solution 3
Data synchronization over each and every device has become of vital importance with the
advancement of a variety of platforms being used and significantly increased use of mobile
devices. In these kinds of situations, the best solution is to use disbursed file synchronization
programs. An open source program popularly referred to as Syncthing can be quite helpful in
offering document synchronization between a range of various devices installed with
windows or any other operating system.
Syncthing allows synchronization of all files across various device platforms including
servers and is free for the users. Syncthing is an open source synchronization tool that is an
alternative to the most popular BitTorrent Sync application which is also known as bitsync.
Syncthing has replaced the techniques of conventional sync and cloud services. Syncthing is
a simple tool used to synchronize your internet transmitted data. (Singh et. al, 2013)
To install Syncthing tool on the Ubuntu system, there are two methods:
 Installing it using the snapd tool
 Using Ubuntu commands such as apt – get tool which involves auto – installation of
Syncthing
There are a number of choices in today's world when it is about packages for open source
synchronization. Syncthing has certain distinct capabilities that make it suitable for a variety
of applications. Syncthing can be tailored according to the user's requirement and is
completely programmable. It is Mac OS X, Linux, Free BSD, Solaris, OpenBSD and
Windows compatible, making it a multipurpose solution. Due to Web GUI, which allows
tracking and configuration, it is simpler to use. Syncthing is easy to set up and configure
because it does not involve advanced configuration options such as IP address and identifies
each machine with its unique ID which is an ID that is required to be shared in order to
enable folder sharing.
Installing Syncthing tool
<INSTALLATION SCREENSHOT GOES HERE>
Configuration of Syncthing tool
When the installation is complete, it is needed to enable the syncthing service which can be
done using the command “sudo systemctl enable [e-mail]”
<SCREENSHOTS FOR CONFIGURATION GO HERE>
To access the package on a web browser, the config file is needed to be saved after editing
and the service needs to be executed. The command to do this is “sudo systemctl start
[e-mail]”
The status can be checked with “sudo systemctl status [e-mail]”

<SCREENSHOT>
The device details and other relevant information is shown and the remote devices connected
are visible.

Solution 4
Timekeeping is essential for surveillance and resolving technical issues for each organization.
Devices need to be in the network. By assessing the sequence of events, accurate timestamps
generated by devices aid to analyse the source of the problem. Regulatory bodies also have a
clear mandate to maintain proper time. The system clock is a tiny quartz crystal circuit that is
responsible for the control of all timing operations in a system. The time is maintained for
UTC (Coordinated Universal Time) according to the World Standard. There is also a
hardware clock inside the desktops better known as Real-Time-Clock (RTC) which is
responsible for keeping track of the right time when the system is turned off. It runs on a
battery when the system is not running on a power source and this battery is called CMOS
battery.
The Time Stamp Counter is 64 bit register which is responsible for measuring the no. of
cycles since the last reset. TCS works on the clock cycles and enables the user to calculate the
precise time due to the high reading speed. TCS may sometimes produce inaccurate data
because of the recurrent changes made in accordance to the CPU frequency. This can create
problems because if one system takes time from TCS and later on takes time from another
CPU it may not match. The High Precision Event Timer i.e. HPET is a hardware timer which
can handle time keeping operations. When the time generated by HPET is compared with that
of RTC, we may find that the one that of HPET is of high resolution. The hardware in a
Linux machine also has a clock but even that cannot be taken as precise due to the fact that it
runs on periodic time intervals. A centralized clock is used to overcome the larger running
times of processes when the system is engaged in running too many of them which may result
in the Linux system’s clock to lose time. A public internet time server or an internal clock
may be used to keep the time synchronization under control.
In a computer network, the time is supplied by an internal clock source which is in from of an
NTP server. The network time protocol or the NTP server is used by networking devices and
servers to get their IP addresses configured which are then used for time syncing in
accordance with UTC for a packet-switched network. A semi-layered system of time sources
is used by NTP in a hierarchical order and the levels under this hierarchical system are called
stratum which are allocated numbers starting from zero. The number zero is allocated to the
main reference clock. Server running at stratum n+1 is in sync to stratum n and these
numbers depict the distance between the reference clock and the server.
The UTC time is a more accurate and tracked time in accordance with the true Earth’s
rotation time and thus it can give second-to-second metrics for precise calculations which is
the reason that it is used instead of GMT which is also similar but does not provide as much
accuracy as UTC gives.

References
Blischak, J.D., Davenport, E.R. and Wilson, G., 2016. A quick introduction to version control
with Git and GitHub. PLoS computational biology, 12(1), p.e1004668.
O’Reilly | Safari. 2019. Practical UNIX and Internet Security, 3rd Edition. [online] Available
at: https://www.oreilly.com/library/view/practical-unix-and/0596003234/ch04s05.html
Singh, S., Raval, G. and Valiveti, S., Reliable and Efficient Storage Solution using Multiple
Cloud Services.
Tldp.org. 2019. PAM (Pluggable Authentication Modules). [online] Available at:
http://tldp.org/HOWTO/User-Authentication-HOWTO/x115.html