Reflog for finding lost commits in Git - Step-by-Step Execution

Process Flow - Reflog for finding lost commits

Commit made

↓

HEAD moves to new commit

↓

Commit lost by reset/rebase

↓

Use git reflog

↓

Find lost commit SHA

↓

Recover commit with git checkout or git reset

This flow shows how commits move HEAD, how commits can be lost by reset or rebase, and how reflog helps find and recover them.

Execution Sample

Git

git commit -m "Add feature"
git reset --hard HEAD~1
git reflog
# find lost commit SHA
git checkout <SHA>

This sequence creates a commit, resets HEAD to previous commit losing the new one, then uses reflog to find and recover the lost commit.

Process Table

Step	Command	HEAD Position	Action	Result
1	git commit -m "Add feature"	commit1	Create new commit	HEAD points to commit1
2	git reset --hard HEAD~1	commit0	Move HEAD back one commit	commit1 becomes unreachable (lost)
3	git reflog	commit0	Show HEAD history	Lists commit1 as recent HEAD position
4	git checkout <commit1 SHA>	commit1	Checkout lost commit	HEAD points to recovered commit1

💡 Lost commit found in reflog and recovered by checkout

Status Tracker

Variable	Start	After Step 1	After Step 2	After Step 4
HEAD	commit0	commit1	commit0	commit1
commit1	does not exist	exists and pointed by HEAD	unreachable (lost)	checked out and reachable

Key Moments - 3 Insights

Why does the lost commit still appear in reflog after reset?

Because reflog records all HEAD movements including resets, so even unreachable commits are listed (see execution_table step 3).

Can I recover a lost commit if it is not in reflog?

No, reflog only tracks recent HEAD changes. If the commit is too old and reflog expired, recovery is not possible.

What does git reset --hard do to the commit?

It moves HEAD and updates working files, making the newer commit unreachable but still in reflog temporarily (see step 2).

Visual Quiz - 3 Questions

Look at the execution table, what is HEAD pointing to after step 2?

A commit0

B commit1

C commit2

D No commit

💡 Hint

Check the 'HEAD Position' column for step 2 in the execution_table.

At which step does the lost commit become reachable again?

A Step 1

B Step 4

C Step 2

D Step 3

💡 Hint

Look for when HEAD points back to commit1 in the execution_table.

If you did not run git reflog, how would the recovery process change?

A The commit would be deleted immediately

B You could still recover the commit easily

C You would not know the lost commit SHA to recover

D HEAD would automatically point to the lost commit

💡 Hint

Refer to the importance of reflog in finding lost commit SHA in the concept_flow.

Concept Snapshot

git reflog records all HEAD movements including resets.
Lost commits become unreachable but are listed in reflog.
Use 'git reflog' to find lost commit SHA.
Recover lost commits with 'git checkout <SHA>' or 'git reset <SHA>'.
Reflog entries expire after some time, so recover quickly.

Full Transcript

This lesson shows how git commits move HEAD pointer and how commands like reset can make commits unreachable or lost. The git reflog command records all HEAD movements, including resets and rebases, so it helps find lost commits by showing their SHA hashes. By running git reflog, you can see the history of HEAD positions and find the SHA of a lost commit. Then you can recover it by checking it out or resetting HEAD to it. The execution table traces these steps: making a commit, resetting HEAD back, listing reflog, and recovering the commit. Variables like HEAD and commit1 change state accordingly. Key moments clarify why reflog still shows lost commits and the limits of recovery. The quiz tests understanding of HEAD positions and recovery steps. Remember, reflog is a safety net for recent lost commits but entries expire over time.