Chapter 2: Git, GitHub and Git Bash

2.3 Repository, remote and origin

By Ravindra Bagale · Last updated: 25 October 2026

Why: the problem

When you join a team at TCS or Zomato, your lead will say: "clone the repo and push to origin". Three new words in one line. If you do not know them, you will copy commands without understanding, and one day you will push to the wrong place. A repository, or repo, is just a project folder that Git watches. Inside it there is a hidden folder called .git, which holds every commit. A remote is another copy of the same repo, usually on a server such as gitlab.com or the company server git.example.com. origin is only a nickname. It is the default name Git gives to the remote you cloned from. It is not magic. You can rename it, add more remotes, or remove them. In this lesson you will look inside the .git folder, make a bare remote on your own laptop so you can see what a server really stores, connect to it as origin, and then rename and change remotes safely.

What we will build

You will build a tiny copy of a real company setup on one machine: a bare repo that acts as the server, and two laptop folders, one for Asha and one for Ravi. Asha pushes, Ravi fetches. You will see that origin is only a saved name for a URL.

What: every term in plain English

• Repository — A folder plus its .git history.
• Working tree — The normal files you see and edit.
• .git folder — The hidden database of commits, branches and settings.
• Bare repo — A repo with only history and no working files. Servers keep repos this way.
• Remote — A saved name plus URL for another copy of the repo.
• origin — The default name of the remote you cloned from.
• upstream branch — The remote branch your local branch tracks, like origin/main.
• Remote-tracking branch — Your local memory of a remote branch, such as origin/main.

How: step by step

1. Make the fake server

   mkdir -p ~/devops-lab/remote-lab && cd ~/devops-lab/remote-lab && git init --bare -b main server.git
   

   What it does: A bare repo is what gitlab.com keeps for you.

   Output

   Initialized empty Git repository in /home/ravi/devops-lab/remote-lab/server.git/
   

2. Clone it as Asha

   git clone server.git asha
   

   What it does: Clone sets origin to the URL automatically.

   Output

   Cloning into 'asha'...
   warning: You appear to have cloned an empty repository.
   

3. See origin

   cd asha && git remote -v
   

   What it does: origin points to the server folder.

   Output

   origin  /home/ravi/devops-lab/remote-lab/server.git (fetch)
   origin  /home/ravi/devops-lab/remote-lab/server.git (push)
   

4. Commit and push

   echo "v1" > app.txt && git add . && git commit -qm "Add app v1" && git push -u origin main
   

   What it does: Asha sends her commit to the server.

   Output

   To /home/ravi/devops-lab/remote-lab/server.git
    * [new branch]      main -> main
   

5. Clone it as Ravi

   cd .. && git clone -q server.git ravi && cat ravi/app.txt
   

   What it does: Ravi gets Asha's work from the server.

   Output

   v1
   

6. Look inside .git

   ls asha/.git
   

   What it does: HEAD, config, objects and refs hold the history.

   Output

   HEAD  branches  config  description  hooks  index  info  logs  objects  refs
   

7. Read the remote config

   git -C asha config --get remote.origin.url
   

   What it does: origin is just this line in .git/config.

   Output

   /home/ravi/devops-lab/remote-lab/server.git
   

Example 1: See that a bare repo has no working files

Stack. Git

Goal. See that a bare repo has no working files.

bare-check.sh

cd ~/devops-lab/remote-lab
ls server.git
git --git-dir=server.git log --oneline
ls asha

What each line does

• ls server.git — No app.txt, only Git data.
• git --git-dir=server.git log — Read history straight from the bare repo.
• ls asha — Asha has normal files to edit.

Green output

HEAD  config  description  hooks  info  objects  refs
1c2b3a4 Add app v1
app.txt

Expected result. A server keeps only history. People edit in working trees on their laptops.

Example 2: Rename origin and add the real company remote

Stack. Git

Goal. Rename origin and add the real company remote.

rename.sh

cd ~/devops-lab/remote-lab/asha
git remote rename origin local
git remote add origin https://git.example.com/shop/app.git
git remote -v
git remote set-url origin git@git.example.com:shop/app.git
git remote get-url origin

What each line does

• git remote rename origin local — origin is only a name. Rename it.
• git remote add origin URL — Add a new origin that points to the company server.
• git remote set-url origin — Change the URL, for example from HTTPS to SSH.
• git remote get-url origin — Print just the URL to be sure.

Green output

local   /home/ravi/devops-lab/remote-lab/server.git (fetch)
local   /home/ravi/devops-lab/remote-lab/server.git (push)
origin  https://git.example.com/shop/app.git (fetch)
origin  https://git.example.com/shop/app.git (push)
git@git.example.com:shop/app.git

Expected result. You can change remotes any time. Nothing is changed on any server by these commands.

Example 3: Fetch a teammate's work and see origin/main move

Stack. Git

Goal. Fetch a teammate's work and see origin/main move.

fetch.sh

cd ~/devops-lab/remote-lab
echo "v2" > ravi/app.txt
git -C ravi commit -qam "Bump to v2" && git -C ravi push -q
cd asha
git fetch local
git status -sb
git log --oneline main..local/main
git merge --ff-only local/main
cat app.txt

What each line does

• git -C ravi push — Ravi pushes v2 to the server.
• git fetch local — Download new commits but do not change files.
• git status -sb — Shows Asha is behind by 1.
• main..local/main — Commits on the server that Asha does not have yet.
• git merge --ff-only — Move main forward only if it is a simple fast-forward.
• cat app.txt — Asha now has v2.

Green output

From /home/ravi/devops-lab/remote-lab/server
   1c2b3a4..7e8f9a0  main       -> local/main
## main...local/main [behind 1]
7e8f9a0 Bump to v2
Updating 1c2b3a4..7e8f9a0
Fast-forward
v2

Expected result. This is exactly how two teammates share work through gitlab.com, only on one laptop.

Verify: did it work?

1. Check both clones match

   git -C ~/devops-lab/remote-lab/asha rev-parse HEAD; git -C ~/devops-lab/remote-lab/ravi rev-parse HEAD
   

   What it does: Same commit ID means same state.

   Output

   7e8f9a0c...
   7e8f9a0c...
   

2. Check tracking

   git -C ~/devops-lab/remote-lab/ravi branch -vv
   

   What it does: Shows main tracks origin/main.

   Output

   * main 7e8f9a0 [origin/main] Bump to v2
   

3. List remotes

   git -C ~/devops-lab/remote-lab/asha remote
   

   What it does: Names only.

   Output

   local
   origin
   

When it is red 1

! [rejected]        main -> main (fetch first)
error: failed to push some refs

Cause. The server has commits you do not have.

Fix, step by step

1. git pull --rebase origin main
2. Fix any conflict.
3. git push again.

When it is red 2

fatal: 'origin' does not appear to be a git repository
fatal: Could not read from remote repository.

Cause. No remote named origin, or the URL is wrong.

Fix, step by step

1. git remote -v
2. git remote add origin URL, or git remote set-url origin URL.

When it is red 3

warning: You appear to have cloned an empty repository.

Cause. Not an error. The server has no commits yet.

Fix, step by step

1. Make a first commit.
2. git push -u origin main.

Cleanup

1. Delete the lab

   rm -rf ~/devops-lab/remote-lab
   

   What it does: Removes the fake server and both clones.

   Output

   (no output)
   

Practice tasks

Homework: try at home

Recap

• A repo is files plus the .git history.
• A server keeps a bare repo.
• A remote is a name plus a URL.
• origin is only the default name.
• git fetch updates origin/main without touching your files.

FAQ

Can I have no origin?

Yes. A local repo works fine without remotes.

Is origin/main a real branch?

It is your local memory of the server branch. It moves on fetch.

Why use a bare repo for servers?

Nobody edits files on the server, so working files are not needed.

Can a remote be a folder?

Yes. A path works as a URL, as in this lab.

Does remote remove delete the server repo?

No. It only removes the name on your laptop.

Interview questions

1. What is the difference between origin and origin/main?

   origin is the remote name. origin/main is the remote-tracking branch for main on that remote.

2. What is a bare repository?

   A repository with no working tree, used as a central shared copy.

3. How do you add a second remote?

   git remote add NAME URL, then push or fetch using that name.

4. What does git fetch do?

   Downloads new commits and updates remote-tracking branches without changing local branches or files.

5. How do you check where you will push?

   git remote -v and git branch -vv.

Samajla ka?

Ravi Sir's Cards for this lesson

Revision cardInterview card

मराठी cards