DevOps Tutorial {Technical Guftgu]

De­vOps Tu­to­ri­­al {Tech­ni­cal Guft­gu]

1- Preq­ui­sites for De­vOps

2- What is De­vOps , De­vOps Stages , Ag­ile

3- Ba­sics of AWS Cloud re­quired for De­vOps

4 Every­thing about Lin­ux from Scratch (ii)

Start­ed in 1964 at Bell Lab­o­ra­to­ry, New JEr­sey for mul­ti-user real time op­er­at­ing sys­tem

With­drawn in 1969
Den­nis Ritchie and Ken Thomp­son start UNICS (Uni­plexed In­for­ma­tion & Com­put­ing Ser­vices)

UNIX V6 was re­leased in 1975 which be­came very pop­u­lar

Dif­fer­ent bod­ies pro­duced dif­fer­ent of their own flavours of UNICS (E.g. IBM-AIX, MAC OS , HP _ UX)

A stu­dent , Li­nus Tor­valds pro­duced LIN­UX in Helsin­ki Uni­ver­si­ty, 1991.Note that LIN­UX is not a UNIX
de­riv­a­tive but mere­ly took in­spi­ra­tion from it
Pro­fes­sor An­drew Tanen­baum pro­duced MIN­UX for teach­ing pur­pos­es

GNU , a prod­uct of the Free Soft­ware move­ment, pro­duced as­so­ci­at­ed soft­ware for Lin­ux ker­nel to com­plete
Lin­ux op­er­at­ing sys­tem
(Lin­ux is orig­in
­ al­ly just ker­nel, not op­er­ar­ing sys­tem : The ker­nel refers to core soft­ware tjhat is re­spon­si­ble for
ac­cess­ing and di­rect­ly in­ter­act­ing with hard­ware com­po­nents of the sys­tem)
Lin­ux source code is free and open source , so any­one is free to pro­duce their own ver­sion of op­er­at­ing sys­tem
Sev­er­al com­pa­nies made their own ver­sion of lin­ux op­er­at­ing sys­tem (aka dis­tri­b­u­tions) - some com­mer­cial
and some open source as well. e.g. RHEL , Fe­do­ra, De­bian <ubun­tu , Cen­tOS, Ama­zon Lin­ux ,etc.
(Op­er­at­ing sys­tems can be op­er­at­ed ei­ther through CLI (Com­mand Line In­ter­face) or GUI (Graph­i­cal User
In­ter­face))

Lin­ux Fea­tures:

1. Open Source - The in­ter­nal source code is open­ly avail­able which al­lows a per­spi­ca­cious in­sight into the in­ner
work­ings and mech­a­nisms of the sys­tem from end user
2. Se­cure - Ad­vanced ac­cess con­trol , au­dit­ing of open source code and de­fence against virus­es' self-repli­cat­ing
mech­a­nism
3. Sim­pli­fied up­dates for all in­stalled soft­ware - Ef­fec­tive pack­age man­age­ment and in­stal­la­tion tools avail­able
which are much
4. Light weight - Dy­nam­ic­ al­ly loads ker­nel mod­ules. Takes less mem­o­ry and disk re­sources. Less bloat­ware.
5. Multi­user and Mul­ti­task­ing - Ad­vanced ac­cess con­trol and ad­vanced process con­trol and re­source shar­ing
mech­a­nisms
**6. Va­ri­ety of mul­ti­ple dis­tri­b­u­tions - ** Due to the open source na­ture, there is a great va­ri­ety of dis­tri­b­u­tions
avail­able. Users can in­stall dis­tri­b­u­tion which most suits them

5 - Every­thing about Lin­ux from Scratch (ii)

Win­dows Lin­ux

Hard­ware in­ter­acts with OS Hard­ware in­ter­acts with ker­nel

Shell is medi­um be­tween OS and user Shell is medi­um be­tween ker­nel and user

Fold­er Di­rec­to­ry

Ad­min­is­tra­tor root user

Soft­ware pack­age

The top lev­el di­rec­to­ry is /, which con­tains all files.

C:\ con­tains all sys­tem files - De­fault fold­ers in­clude
Root di­rec­to­ries in­clude /root , /home , /boot , /etc ,
'Pro­gram Files, Users , Pro­grams(x86) and Per­flogs
/usr and /bin

CLI tends to be faster and more ac­cu­rate than do­ing the same thing in GUI mode
A 'fold­er' in Lin­uc is re­ferred to as di­rec­to­ry

File sys­tem hi­er­ar­chy in Lin­ux

1. /home - home di­rec­to­ry of oth­er users

2. /root - home di­rec­to­ry for root user

3. /boot -Con­tains bootable files for lin­ux
4. /etc - con­tains all con­fig­u­ra­tion
5. /usr - soft­ware is in­stalled in this di­rec­to­ry by de­fault
6. /bin - con­tains com­mands used by all users
7. /sbin - con­tains com­mands used by only root user
8. /opt - Op­tion­al ap­pli­ca­tion soft­ware pack­ages
9. /dev - De­vice files . In­cludes ter­mi­nal de­vices, USBs

POST (Pow­er On Self Test) :

6 - Every­thing about Lin­ux from Scratch (iii) - Lin­ux com­mands and its
use cas­es

**How to cre­ate a file: **

1. cat
E.g.

cat > file1

1
Copy Copied

Here, file1 is the new file cre­at­ed. Any­thing en­tered into stdin is writ­ten into this file. To ter­mi­nate in­put , the
key­board short­cut Ctrl+D can be used which sends EOF (End-of-File char­ac­ter)
2. touch cre­ates an emp­ty file
3. vi / vim - one of the most pop­u­lar CLI ed­it­ors
4. nano - an­oth­er pop­u­lar CLI ed­it­or

sudo su - (su­pe­ruser-do switch-user) changes from

NOTE : Every­thing in lin­ux bash is case-sen­si­tive, stick to small­er let­ters most­ly

The cat com­mand

One of the most uni­ver­sal tools . It es­sen­tial­ly copies stan­dard in­put to stan­dard out­putThere are three main use
cas­es :

1. Cre­ate file: Cre­ate sin­gle file

cat > file1

Hello

1
2

Copy Copied

2. Con­cate­nate file: Con­cate­nate the con­tent of files

3. Copy files Copy­ing con­tent of one file to an­oth­er

The touch com­mand

Up­dates time­stampUse cas­es of touch :

1. Cre­ate an emp­ty file

touch file1

Copy Copied

(cre­ates emp­ty file file1 )

2. Cre­ate mul­ti­ple emp­ty files

touch file1

Copy Copied

(cre­ates emp­ty file file1 )

3. Change all time­stamps of a file
touch file2

Copy Copied

4. Up­date only ac­cess time of file, mod­if­y time of file

touch -a file1
touch -m file2

1
2

Copy Copied

(cre­ates emp­ty file file1 )Types of time­stamps:1.**Ac­cess time: ** ( Last time when user ac­cesed)

2. **Change time: ** ( Last time when file meta­da­ta was changed)

3. **Mod­if­ied time: ** (Last time when a file was mod­if­ied)

The vi ed­i­tor

A pro­gram­mer text ed­it­or

Can be used to edit any plain text. Es­pe­cial­ly use­ful for edit­ing pro­grams;
:w Saves file
:wq / :x Saves file and quits ed­it­or
:q Quits file
:q: Force­ful­ly quits, even if file is un­saved

The nano ed­i­tor

nano file_a

Copy Copied

cre­ates file_a if it doesn't al­ready ex­ist and opens it.

Con­tains in­struc­tions on key­baord short­cuts for sav­ing /ex­it­ing/ etc.

Demon­stra­tion

7 - Every­thing about Lin­ux from Scratch (iv)

mkdir com­mand

Cre­ates new di­rec­to­ry

mkdir dir1 dir2

 1

Copy Copied

cre­ates new di­rec­to­ries dir1 and dir2

mkdir dir1/dir2/dir3

Copy Copied

cre­ates dir1 di­rec­to­ry , a dir2 in­side it and an­oth­er dir2 in­side it.

cd di­rec­to­ry

Change di­rec­to­ry com­mand

cd dir2

Copy Copied

switch­es to di­rec­to­ry dir2

cd ..

Copy Copied

changes di­rec­to­ry to .. (i.e. par­ent di­rec­to­ry)

pwd di­rec­to­ry

print work­ing di­rec­to­ry/ Shows cur­rent work­ing di­rec­to­ry.

pwd

Copy Copied

8 - Every­thing about Lin­ux from Scratch (v)

Any file/di­rec­to­ry which be­gins with . is hid­den, they can be shown us­ing ls only with -a flag

ls -a
 1

Copy Copied

cp com­mand

Used for copy­ing files

cp file1 file2

Copy Copied

Copies file1 to file2

mv com­mand

Move files/di­rec­to­ries

mv file1 dir1

Copy Copied

moves file1 to dir1

rmdir com­mand

Com­mand for re­mov­ing spec­if­ied di­rec­to­ry (if emp­ty)

rmdir -p

Copy Copied

Re­move both men­tioned di­rec­to­ry and its par­ent di­rec­to­ry

rm com­mand

re­move files/di­rec­to­ries

rm -rf

Copy Copied

force­ful­ly re­moves all con­tents of di­rec­to­ry and its con­stituent sub­di­rec­to­ries and files re­cur­sive­ly

rm -rp
 1

Copy Copied

re­cur­sive­ly re­move both di­rec­to­ry and its par­ent di­rec­to­ry.

Demon­stra­tion

9 - Every­thing about Lin­ux from Scratch (vi)

hostname com­mand

Re­veals de­tails about the cur­rent­ly run­ning ma­chine

hostname -i

1
2

Copy Copied

Demon­strates ver­sion name of OS

ifconfig com­mand

Shows IP ad­dress of the ma­chine.

yum pack­age man­ag­er

Yel­low­Dog Up­date Mod­i­fi­er (YUM)

yum install httpd -y

Copy Copied

In­stalls httpd pack­age to sys­tem. The -y flag auto-con­firms every prompt

yum remove httpd

Copy Copied

Re­moves httpd pack­a­gr

yum update httpd

Copy Copied
Up­dates httpd pack­age

yum list

Copy Copied

Show list of in­stalled pack­ages

Sys­tem ser­vices

service httpd start

Copy Copied

Ini­ti­ates httpd ser­vice

service httpd status

Copy Copied

Checks sta­tus of httpd ser­vice

chkconfig httpd on
chkconfig
chkconfig httpd off

echo com­mand

re­turns what­ev­er is sent as in­put as out­put.

echo "Hello"

Copy Copied

re­turns Hello to out­put.

echo "Welcome" >file1

Copy Copied

redi­rects in­put Welcome to file1 which is cre­at­ed if it doesn't al­ready ex­ist.

echo "Wilkommen" >> file1

1
Copy Copied

ap­pends in­put Wilkommne to file

grep com­mand

used for search­ing strings any­where , ap­plies reg­u­lar ex­pres­sion match­ing

sort com­mand

sorts out­put in al­pha­bet­ic­ al­ly as­cend­ing or­der

Demon­stra­tion

10 - Every­thing about Lin­ux from Scratch (vii)

User con­trol com­mands

useradd com­mand - Cre­ates user

useradd bhupinder

Copy Copied

cre­ates a new user named bhupinder . The users/ user­groups are saved in /etc/passwd
groupadd com­mand - Cre­ates user group

groupadd techguftgu

Copy Copied

cre­ates a new user­group called techguftgu . The user­groups are saved in /etc/group
gpasswd com­mand - Adds user to user­group

gpasswd -a aman techguftgu```

## Shortcut(symbolic link) commands
`ln` command - creates link
```bash
ln file2 backupfile2

2
3

4
 5
6

Copy Copied

cre­ates a hard link backupfile2 for file2 while

ln -s file1 softfile1

Copy Copied

cre­ates a soft link softfile1 for file1

A soft link is iden­ti­cal to a short­cut in win­dows. Even if it is delet­ed, the orig­in
­ al file is un­af­fect­ed. A hard link cre­ates
a com­plete back­up copy of the giv­en file. It,how­ev­er, doesn't re­flect changes made to orig­in
­ al file.

File archiv­ing com­mands

tar com­mand - an archiv­er tool used to com­bine mul­ti­ple files into one.
gzip com­mand

tar -cvf dirx.tar dirx

Copy Copied

the flag -c , -v and -f stand for creak , ver­bose and force­ful­ly re­spec­tive­ly. The com­mand cre­ates a tar file
dirx.tar from the di­rec­to­ry dirx .

gzip dirx.tar

Copy Copied

Com­press­es the con­tents of dirx.tar into dirx.tar.gz

gunzip dirx.tar.gz

Copy Copied

tar -xvf dirx.tar

Copy Copied

The -x flag stands for ex­trac­tion

wget com­mand

Non-in­ter­ac­tive down­loader tool

Demon­stra­tion

Every­thing about Lin­ux from Scratch

Ac­cess con­trols in Lin­ux

Ac­cess mode File Di­rec­to­ry

r (4) To dis­play con­tent To list con­tent

w (2) To mod­if­y To cre­ate/re­move

x (1) To ex­e­cute the file To en­ter into di­rec­to­ry

drwxr_xr___ 1 root root 0 July 01 04:00PM dirx

The first char­ac­ter d sig­ni­fies that the giv­en file is a di­rec­to­ry. If it was a sym­bol­ic link, it would be l ; _ if it just a
file.
The next three char­ac­ters de­fine per­mis­sions for own­er/root user , the next three for user group and the last three
for oth­ers.
The per­mis­sions are de­fined by three char­ac­ters : r stands for read­able, w stands for writable and x stands for
ex­e­cutable.
Then fol­lows the own­er of the file/di­rec­to­ry and the user group of file/di­rec­to­ry and then the file size in bytes.
chmod com­mand - change ac­cess per­mis­sions of file/di­rec­to­ry

1 chmod 536 dirx

changes per­mis­sions od dirx to 536 ( equiv­a­lent of r_x_wxr_x)

An­oth­er way to al­ter these per­mis­sions :-

1 chmod u+x,g-w,o=rwx file1

This will add ex­e­cute per­mis­sions to user (if they al­ready dont have it), re­move write per­mis­sions from group (if
they have it) and set oth­ers' ac­cess con­trol to rwx.
chown com­mand - change own­er of file/di­rec­to­ry

1 chown bhupinder file1

chgrp com­mand - change group of file/di­rec­to­ry

1 chgrp main file1

11 - What is GIT /GitHub

Source Code Man­age­ment is nec­es­sary to en­sure ef­fi­cient col­lab­o­ra­tion for soft­ware pro­jects.
Be­fore the era of ver­sion con­trol, a man­ag­er had to in­di­vid­u­al­ly mon­it­or code as­signed to sev­er­al dif­fer­ent
per­son­nel.
There are two forms of code man­age­ment :

1. Cen­teal­ized Ver­sion Con­trol Sys­tem

2. Dis­trib­uted Ver­sion Con­trol Sys­tem

Note: Git is just a soft­ware/tool . Github and Git­lab are tools built on top of

12 - Three stages of Git - Work­ing di­rec­to­ry , stag­ing area and lo­cal

repos­i­to­ry

Git work­flow

1. Cre­ate your pro­ject di­rec­to­ry

2. Run git init in­side pro­ject repos­it­o­ry. This makes the di­rec­to­ry a lo­cal repos­it­o­ry and makes a .git file
3. This log­ic­ al­ly (as op­posed to phys­ic­ al­ly) con­verts your di­rec­to­ry into three di­vi­sions
4. Work­ing di­rec­to­ry

5. Stag­ing area
6. Lo­cal repo
7. The code you write first gets saved in work­ing di­rec­to­ry.
8. Stag­ing area is where pre­ma­ture code can be saved. It is not nec­es­sary but rec­om­mend­ed for or­gan­ised
work­flow. This can be done with git add .
9. git commit will save a snap­shot of code as it was and changes from last com­mit to lo­cal repo.

What is Github? - Not to be con­fused with Git, Github pro­vides a cen­tral repos­it­o­ry. It is owned by Mi­crosoft with
free pub­lic repos­it­o­ry stor­age.
Since GitHub was ac­quired by a ma­jor tech­nol­o­gy gi­ant, Git­Lab was launched as a com­plete­ly free al­ter­na­tive.
git push is used to push changes to lo­cal repos­it­o­ry into re­mote repos­it­o­ries (like those on GitHub).
git pull is used to pull changes on re­mote repos­it­o­ry to lo­cal repos­it­o­ry.
A code repos­it­o­ry may have mul­ti­ple branch­es, the de­fault branch is known as mas­ter branch.
Every com­mit cre­at­ed pro­duces a Com­mit ID and record of changes made then. Every­thing is kept trans­par­ent for
all col­lab­o­ra­tors to see.

Repos­it­o­ry

Repos­it­o­ry is where all code is stored as a di­rec­to­ry in serv­er.

The fold­er is par­tic­u­lar to the prod­uct

The Serv­er stores all repos­it­o­ries and con­tains all meta­da­ta. E.g. Mi­crosoft stores all GitHub data
All files can be phys­ic­ al­ly seen and worked upon in the work­ing di­rec­to­ry/ work­space.
 l

In oth­er CVCS , there is no stag­ing area but di­rect com­mit-ing only from

Com­mit ID -

Ref­er­ence to iden­ti­fy each change.

Iden­ti­fies who changed file.
Tags:
These as­sign a mean­in­ful name with a spe­cif­ic ver­sion in the repos­it­o­ry. Once a tag is cre­at­ed for a par­tic­u­lar
save
Snap­shots

Rep­re­sens data state at par­tic­u­lar time.

It is in­cre­men­tal: in­stead of stor­ing en­tire data, only changes are stored
Push:
Push op­er­a­tions copies changes from lo­cal repos­it­o­ry to a re­mote / cen­tral repo. Changes are stored
per­ma­nent­ly into git repos­it­o­ry.
Pull:
Pull op­er­a­tion copies changes from a re­mote repos­it­o­ry to lo­cal ma­chine
Branch
Branch­es can be used to di­vide the work­flow by dif­fer­ent tasks. E.g. some­one is do­ing CSS styling and
back­end dvel­op­ment is han­dled sep­a­rate­ly. They can han­dle dif­fer­ent branch­es, which can be merged lat­er
when work is com­plete on both ends.
Each task can be giv­en one sep­a­rate branch

Fi­nal­ly all branch­es can be merged

Use­ful for par­al­lel work.
Can cre­ate one branch through an­oth­er branch.

Changes are per­son­al to the par­tic­u­lar branch.

De­fault branch is known as Mas­ter branch.
File cre­at­ed in work­space will be vis­ib
­ le in any of the branch­es.
Once com­mit­ed, changes will be­long to the par­tic­u­lar branch.
Ad­van­tages of Git
Free and Open ource

Fast and small in size

Ef­fec­tive Se­cu­ri­ty
Pow­er­ful hard­ware is not nec­es­sary
Easy branch­ing of code al­lows good com­part­men­tal­is­ a­tion
**Types of branch­es: **

1. Bare repos­it­o­ries
2. Non-bare repos­it­o­ries

Store and share only

 A

Com­mit

Store changes in repos­it­o­ry : This is asigned a Com­mit ID.

It con­sists of 40 al­phanu­mer­ic char­ac­ters

It uses SHA-1 check­sum ver­if­i­ca­tion

Even if a sin­gle char­ac­ter is changed , Com­mit ID is al­tered as well.
Helps you track changes
Com­mit is also known as SHA-1 hash.

13 - How to in­stall git on AWS Lin­ux ma­chine

14 - How to add, com­mit in git

There

15 - How to cre­ate branch in git

Branch­es en­able par­al­lel work sep­a­rate from mas­ter branch.

A pro of paal­lel work is that mas­ter branch can be kept free from fa­tal er­rors which may oc­cur whilst work­ing on
oth­er branch­es.
While we make a branch, all data is copied as it is on the main branch it split from.
When want­ed, work done in the branch can be merged back into mas­ter/par­ent branch and work can fur­ther
con­tin­ue.
We can make as many branch­es as per task re­quire­ments

File cre­at­ed in work­space is vis­ib

­ le from all branch­es.

1 git branch

Shows list of all branch­es, and cur­rent branch

1 git log ---oneline

Shows log of com­mits in one line

1 git branch <branch-name>

Cre­ates new branch with branch-name

1 git checkout <branch-name>

switch to branch-name

1 git branch -d <branch-name>

Delete branch-name branch

1 git merge <branch-name

Merge branch-name with master branch

1 git log

Ver­if­y mes­sage of git

1 git push origin master

Push to cen­tral repo like GitHub

Git con­flict

When same files hav­ing dif­fer­ent con­tent in dif­fer­ent branch­es are merged, a con­flict emerges.

Git stash

Set aside files/changes for work­ing upon lat­er

1 git stash

stash­es an item

1 git stash list

See stashed items

1 git stash apply stash @[0]

Ap­ply stashed items

1 git stash clear

Clear stashed items

 1 git reset

Demon­stra­tion

16 - What is git re­vert/ git­lab

17 - What is Chef in De­vOps(i)

Used to be han­dled by sys­tem en­gi­neer ear­li­er. Chef is used for au­toma­tion of sysad­min tasks be­fore,
eThere are two types of con­gi­fure man­age­ment tools :-

1. Oush vased tools that use push con­fig­u­ra­tion servers that push­es con­fig­u­ra­tion o the nodes. In­fra­struc­ture as
code ser­vices. Serv­er no­ti­fies about up­dates be­ing rolled out.
An­si­ble and are push based tools
2. Pull based tools - Pull con­fig­u­raion nodes. Reuces fault tore­ance.
E.g. Chef. Pup­pet

Too­lused for de­vops en­gi­neer de­vel­oped in ruby and er­lang.

Used to be called Mar­io
­ nette.
Made open source in 2019

Au­to­mates all tasks done by sysad­mins ear­li­er

Con­fig­u­ra­tion man­age­ment tools au­to­mate all sysad­min tasks
Code be­comes re­peat­able, testable and ver­sion­able.
-Ad­v­nat­ges of such tools

1. Com­plete au­toma­tion
2. Re­duced down­times
3. Imorive per­for­mance

4. En­sure com­pli­ance
5. Pre­vent er­rors
6. Re­duce costs

Chef ar­chi­tec­ture

Con­sists of:

1. Work­sta­tion - Code is worked on here. Code grouped to­geth­er may be termed as recipes. Cook­book is a
col­lec­tion of recipes.
2. Chef serv­er - Cook­books are stored here. It id con­nect­ed to chef serv­er by a CLI tool known as knife
3. Mode serv­er -
**chef suo­er­mar­ket ** con­tains sev­er­al chef cook­books.

0 Work­sta­tions are per­son­al compyters

De­vops en­gi­neers op­er­ate from chef serv­er.
Mi­dle­man be­tween node serv­er and work­sta­tion.
Node are where codes are ac­tu­al­ly ap­plied.
Ohai fetch­es cur­rent state of node.

18 - What is Chef in De­vOps(ii)

How to cre­ate cook­book and Recipes

Cre­ate a Lin­ux Ma­chine in AWS

Use PuT­Ty

1 which chef

1 chef --version

check which ver­sion of chef you have

Cook­books

It is a col­lec­tion of recipes and some oth­er files and fold­ers

Con­tents of a cook­book:-

1. .chefignore - Sim­il­ar to func­tions of .gitignore

2. Kitchenyml

3. Metadata.rb
4. Readme.md
5. Recipes

6. Specs
7. Tests -

1 chef generate cookbook

1 generate-recipe <name>

Ruby recipe code struc­ture:

1 File do
2 content
3 action Create
4 end
Ruby recipe for soft­ware in­stal­la­tion could be

1 package 'tree' do
2 action install
3 end

Chef does not over­write any­thing if a recipe is run mul­ti­ple times.

1 chef-client -zr "recipe[test-cookbook "recipe-name"]"

1 chef exec ruby -c <path of chef recipe>

Re­source

Ba­sic Com­po­nent of a recipe used to man­age the in­fra­struc­ture with dif­fer­ent kind of states. There can be mul­ti­ple
re­sources in a recipe, which will help in con­fig­ur­ing and man­ag­ing in­fra­struc­ture.
Some re­sources are:

1. Pack­age - Man­ages pack­ages on a node

2. Ser­vice - Man­ages ser­vices on a node
3. User - Man­ages users on the node
4. Group - Man­ages groups

5. Tem­plate - Man­ages file with em­bed­ded ruby tem­plate

6. Cokkbook.file - Trans­fers the files from the files sub­di­rec­to­ry in the cook­book to a lo­ca­tion on the node
7. File - Man­ages the con­tent of a file on the node
8. ex­e­cute - ex­e­cutes a com­mand on the node

9. cron - Edit an ex­ist­ing cron file on the node

10. di­rec­to­ry - man­ages di­rec­to­ry on a node

Demon­stra­tion

57c5945428bb341b0946f2c7ad5837f2.png
5412f89fcac8e1aa16eb394dd36536ba.png

19 - What is at­tribute in CHEF

Helps dis­tin­guish fea­tures of sev­er­al dif­fer­ent node servers run­ning.

At­tribute is a key-val­ue pair whch rep­re­sents a spe­cif­ic de­tail about a node

An at­tribute can help us find out

Cur­rent state of node

 What the state of a node was af­ter the end of pre­vi­ous chef-client run
what state of a node should be at the end of the cur­rent chef-client run

Types of at­tribute (n ncas­ing pri­orty)

de­fault
force=de­fault

nor­mal
Over­ride
force-over­ride
au­to­mat­ic

At­trib­ut­es de­fined by Ohai have the high­est pri­orty, fol­lowed by at­trib­ut­es , fol­lowed by those de­fined by at­tribute
files

20 - Chef recipes

21 - Boot­strap­ping in CHEF

22 - Roles in CHEF

23 - In­tro­duc­tion to Dock­er

24 - OS Lev­el vir­tu­al­i­sa­tion in Dock­er

Dock­er is an open source cen­tralised plat­form

Dock­er per­ofmrs OS-lev­el vir­tu­al­is­ a­tion while VMware per­forms hard­ware-lev­el vir­tu­al­is­ a­tion

Vir­tu­al­i­sa­tion in VMWare schema : Hy­per­vi­sor is built upon Hard­ware. The VM takes all re­sources from
hard­ware.

Con­tainer­iza­tion in Dock­er : Dock­er is in­stalled on top of host OS. Based upon con­tain­ers rather than vir­tu­al
ma­chines which con­tain UNIX based op­er­at­ing sys­tems. Con­tain­er im­ages can be pulled from an on­line repos­it­o­ry
called Dock­er Hub. Most code of UNIX like dis­tros are sim­il­ar, the con­tain­er-spe­cif­ic code is ob­tained from Dock­er
Hub. A con­tain­er is very light weight com­pared to a vir­tu­al ma­chine as only mar­gin­al op­er­at­ing sys­tem files
nec­es­sary for boot­ing and run­ning the sys­tem is re­quired.

25 - Dock­er - ar­chi­tec­ture, pros and cons

Ad­van­tages of Dock­er:

No pre-al­lo­ca­tion of RAM - Un­like vir­tu­al ma­chines. When n im­age is run, it is re­ferred to as a con­tain­er.
In­creased ef­fi­cien­cy of Con­tin­u­ous In­te­gra­tion - Dock­er re­moves de­pen­den­cy on plat­form and hard­ware

Low­er costs - Since there is no pe-al­lo­ca­tion of RAM

 Light weight
Can

In or­der to run Dock­er on win­dows , you will need to en­able vir­tu­al­is­ a­tion and hy­per­vi­sor. Dock­er is main­ly
de­signed for Lin­ux sys­tems.

Changes to a con­tain­er can be saved to im­ages.

Dis­ad­van­tages of Dock­er

Not great for cross-plat­form com­pat­ib

­ il­it­y.
Not a good so­lu­tion for ap­pli­ca­tions re­quir­ing GUI

Suit­able when dvelop­ing OS and test­ing OS is same

Dock­er ar­chi­tec­ture

A dock­er­file plays the same role which is played by cook­book in Chef. The dock­er­file makes an im­age and in­stalls
de­pen­den­cies and re­quire­ments as men­tioned. A con­tain­er is pro­duced from the im­age. The con­tain­er may be
pub­lished to Dock­er Hub (a.k.a. reg­istry) as an im­age. It can be pulled by oth­er users as re­quired.

A con­tain­er has a lay­ered file sys­tem, which has its process­es run lay­er-by-lay­er.

Dock­er ecosys­tem

1. Dock­er Client - The end user of Dock­er

2. Dock­er dae­mon - ex­e­cutes com­mands and pro­duces im­ages

3. Dock­er Hub - Re­mote stor­age for keep­ing im­ages

4. Dock­er Im­ages - tem­plates for run­ning con­tain­ers
5. Dock­er com­pose - Man­ag­ing and run­ning sev­er­al im­ages at once

Dock­er Host - used to pro­vide an en­vi­ron­ment to ex­e­cute and run ap­pli­ca­tions . Con­tains the whole dock­er
ecosys­tem.

Dock­er reg­istry man­ages and stores dock­er im­ages.There are two types of reg­istry in Dock­er:

1. Pub­lic Reg­istry
2. Pri­vate reg­istry - For use with­in en­ter­prise

Dock­er Im­ages are bi­na­ry read-only tem­plates re­quired to run con­tain­ers.

There are three ways to cre­ate im­ages:

1. Take im­age from Dock­er Hub

2. Cre­ate your own im­age from Dock­er­file
3. Cre­ate im­age from ex­ist­ing dock­er con­tain­er

Dock­er Con­tain­er - a copy of the im­age it was cre­at­ed from. Hold all es­sen­tial pack­ages in or­der to run an
ap­pli­ca­tion.
26 - Us­ing dock­er - run­ning/stop­ping con­tain­ers

27 - Dock­er­file cre­ation

docker diff com­mand

1 docker diff <Container1> <Container2>

Find the dif­fer­ence be­tween <Con­tain­er1> and <Con­tain­er2>

Out­op­ut shown shows three types of changes:-

1. C (Change)
2. A (Ap­pend)
3. D (Delete)

1 docker commit <ContainerName> <ImageName>

Cre­ates new im­age of con­tain­er.

1 docker run -it --name <ContainerName> <UpdatedContainer> /bin/bash

Dock­er­file

A text file con­tain­ing a set of in­struc­tions

Au­to­mates dock­er im­age cre­ation
Dock­er com­ponenets

1. FROM
Base im­age on top of which im­age is be­ing cre­at­ed. A dock­er­file al­ways starts with this
2. RUN
Ex­e­cutes com­mands. This also pro­duces lay­ers in im­age
3. MAIN­TAIN­ER
Pro­vides in­for­ma­tion about au­thor and con­trib­u­tors
4. COPY
Copy files from lo­cal sys­tem to con­tain­er
5. ADD
Sim­il­ar func­tion to COPY , but also ca­pa­ble of down­load­ing off in­ter­net and un­pack­ing/ex­tract­ing archived
di­rec­to­ries
6. EX­POSE
Ex­pos­es ports for net­work­ing
7. WORKDIR
Set work di­rec­to­ry for con­tain­er
 8. CMD
Ex­e­cute com­mands dur­ing con­tain­er cre­ation
9. EN­TRY­POINT
Like CMD, but with high­er pri­orty. En­sures se­quence of com­mands be­ing ex­e­cut­ed
10. ENV
set en­vi­ron­ment vari­ables

1 docker build

Used for build­ing im­ages from dock­er­file

Demo

28 - Dock­er Vol­ume

Vol­ume is just a di­rec­to­ry where con­tain­er run­time in­for­ma­tion may be per­sis­tent­ly stored. It en­ables shar­ing of
in­for­ma­tion among mul­ti­ple con­tain­ers.

A di­rec­to­ry is just de­clared as a vol­ume, then the vol­ume can be shared

Even if con­tain­er is stopped, vol­ume is ac­ces­si­ble.
In­for­ma­tion from a con­tain­er stored in a vol­ume can be shared with any con­tain­er
Vol­umes can't be cre­at­ed from ex­ist­ing con­tain­ers
Vol­ume is not in­clud­ed when an im­age is up­dat­ed.
Vol­umes can be mapped in two ways: Con­tain­er <-> Con­tain­er, Host <-> Con­tain­er

Vol­ume will act as sim­ply just an­oth­er di­rec­to­ry if it is not linked up with con­cerned con­tain­er.

Ben­e­fits of us­ing vol­umes

De­cou­pling con­tain­ers from stor­age

Share vol­ume among dif­fer­ent con­tain­ers
At­tach vol­umes to con­tain­er
Upon delet­ing con­tain­ers, vol­umes are not delet­ed

1 docker

29 - Dock­er port map­ping, Dock­er ex­pose

IP ad­dress­ing ba­sics

Every net­work­ing process has a port as­signed to it to unique­ly iden­ti­fy it.

Dock­er IP net­work­ing

Ports can be mapped from host to con­taine

 To ac­com­plish this, cer­tain ports of a con­tain­er are set as ex­posed

1 docker run -td --name techserver -p 80:80 Ubuntu

would map port num­ber 80 of host to the tech­serv­er ubun­tu con­tain­er.

1 docker port techserver - 80

2
3
4 # 30 - What is dock­er hub and how it works
5
6 # 31 - An­si­ble in­tro­duc­tion and work­
ing
7
8 # 32 -
9
10 # 33 -
11
12 # 34 -
13
14 # 35 -
15
16 # 36 -
17
18 # 37 -
19
20 # 38 - Jenkins and Maven basics
21
22 # 39 -
23
24 # 40 -
25
26 # 41 -
27
28 # 42 -
29
30 # 43 - Nagios Continuous Monitoring Tool
31
32 # 44 - Nagios setup on Linux
33
34 # 45 - Ku­ber­netes in­tro­duc­tion
35
36 # 46 - Ku­ber­netes ar­chi­tec­ture
37
38 # 47 -
39
40 # 48 -
41
42 # 49 -
43
44 # 50
45
46 # 51 - Kubernetes Networking
47
48 # 52 -
49
50 # 53 -
51
52 # 54 -
53
54 # 55 -
55
56 # 56 -
57
58 # 57 - What is Helm | HELM chart
59
60 # 58 - Helm in­stal­la­tion and com­
mands