Log rotation script in Bash Scripting - Time & Space Complexity

Time Complexity: Log rotation script

O(n)

Understanding Time Complexity

We want to understand how the time needed to run a log rotation script changes as the number of log files grows.

Specifically, how does the script's work increase when there are more logs to rotate?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

#!/bin/bash
LOG_DIR="/var/log/myapp"
MAX_FILES=5

files=($(ls -t "$LOG_DIR"/*.log))

for (( i=$MAX_FILES; i<${#files[@]}; i++ )); do
  rm "${files[$i]}"
done

This script lists log files sorted by newest first, then deletes older logs beyond a set limit.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Loop over log files starting from the (MAX_FILES)th file to delete old logs.
• How many times: The loop runs once for each log file beyond the maximum allowed, which depends on total log count minus MAX_FILES.

How Execution Grows With Input

As the number of log files grows, the script deletes more files beyond the limit.

Input Size (n)	Approx. Operations
10	5 deletions (10 - 5)
100	95 deletions (100 - 5)
1000	995 deletions (1000 - 5)

Pattern observation: The number of deletions grows linearly as the number of log files increases beyond the limit.

Final Time Complexity

Time Complexity: O(n)

This means the script's work grows in direct proportion to how many extra log files there are to delete.

Common Mistake

[X] Wrong: "The script always takes the same time because MAX_FILES is fixed."

[OK] Correct: Even though MAX_FILES is fixed, the script deletes all files beyond that number, so if there are many extra logs, it takes longer.

Interview Connect

Understanding how scripts scale with input size helps you write efficient automation that works well as data grows.

Self-Check

"What if the script compressed old logs instead of deleting them? How would the time complexity change?"

Practice

1. What is the main purpose of a log rotation script in bash?

easy

A. To archive old log files and start new ones to prevent disk space issues

B. To delete all log files permanently

C. To create new log files without removing old ones

D. To convert log files into executable scripts

Solution

1. Step 1: Understand log rotation purpose

   Log rotation scripts archive old logs and create fresh logs to avoid disk space overflow.

2. Step 2: Compare options

   Only To archive old log files and start new ones to prevent disk space issues describes archiving old logs and starting new ones, which matches the purpose.

3. Final Answer:

   To archive old log files and start new ones to prevent disk space issues -> Option A

4. Quick Check:

   Log rotation = archive old + new logs [OK]

Hint: Log rotation means archiving old logs and starting fresh ones [OK]

Common Mistakes:

• Thinking log rotation deletes all logs
• Believing it only creates new logs without archiving
• Confusing log rotation with script conversion

2. Which of the following bash commands correctly appends the current date to a log filename for rotation?

easy

A. mv logfile.log logfile-date.log

B. mv logfile.log logfile-$(date +%Y-%m-%d).log

C. mv logfile.log logfile-$(date +%H:%M:%S).log

D. mv logfile.log logfile-$(date +%Y/%m/%d).log

Solution

1. Step 1: Understand date format for filenames

   Using $(date +%Y-%m-%d) produces a safe filename format like 2024-06-01.

2. Step 2: Check each option

   mv logfile.log logfile-$(date +%Y-%m-%d).log uses dashes which are safe in filenames. mv logfile.log logfile-date.log uses literal 'date', not dynamic. mv logfile.log logfile-$(date +%H:%M:%S).log uses colons which are invalid in filenames on many systems. mv logfile.log logfile-$(date +%Y/%m/%d).log uses slashes which are directory separators, invalid in filenames.

3. Final Answer:

   mv logfile.log logfile-$(date +%Y-%m-%d).log -> Option B

4. Quick Check:

   Date format with dashes = safe filename [OK]

Hint: Use date with dashes, avoid colons or slashes in filenames [OK]

Common Mistakes:

• Using colons or slashes in filenames
• Not using command substitution for date
• Using literal text instead of date command

3. Given this snippet in a log rotation script:

gzip logfile-$(date +%Y-%m-%d).log
ls logfile-*.log.gz | wc -l

What will the wc -l command output if there are 3 compressed log files matching the pattern?

medium

A. Error: No such file or directory

B. 0

C. 1

D. 3

Solution

1. Step 1: Understand the commands

   The gzip command compresses a log file, creating a file ending with .gz. The ls logfile-*.log.gz lists all compressed log files matching the pattern.

2. Step 2: Count matching files

   If there are 3 compressed log files, ls outputs 3 lines. Piping to wc -l counts these lines, so output is 3.

3. Final Answer:

   3 -> Option D

4. Quick Check:

   Count of compressed logs = 3 [OK]

Hint: Count lines from ls output to get number of files [OK]

Common Mistakes:

• Assuming gzip outputs count
• Thinking wc -l counts characters
• Expecting error if files exist

4. Identify the error in this log rotation snippet:

mv logfile.log logfile-$(date +%Y-%m-%d).log
gzip logfile-$(date +%Y-%m-%d).log
rm logfile.log

medium

A. The date format uses slashes which are invalid

B. The gzip command is missing the -d option

C. The rm logfile.log command deletes the original log after moving, causing error

D. The mv command should use cp instead

Solution

1. Step 1: Analyze the commands order

   The mv command renames logfile.log to logfile-date.log, so logfile.log no longer exists after this.

2. Step 2: Check the rm command

   After moving, rm logfile.log tries to delete a file that was already moved, causing an error or no effect.

3. Final Answer:

   The rm logfile.log command deletes the original log after moving, causing error -> Option C

4. Quick Check:

   Deleting moved file causes error [OK]

Hint: Don't delete a file after moving it; it's already renamed [OK]

Common Mistakes:

• Deleting a file after moving it
• Confusing gzip options
• Using invalid date formats

5. You want to write a log rotation script that keeps only the last 5 compressed log backups and deletes older ones. Which bash command snippet correctly deletes older backups?

hard

A. ls -1t logfile-*.log.gz | tail -n +6 | xargs rm -f

B. ls -1t logfile-*.log.gz | head -n +6 | xargs rm -f

C. ls -1 logfile-*.log.gz | tail -n 5 | xargs rm -f

D. ls logfile-*.log.gz | grep -v 'logfile-*.log.gz' | xargs rm -f

Solution

1. Step 1: Understand the command parts

   ls -1t lists files sorted by modification time, newest first. tail -n +6 outputs lines starting from line 6, i.e., files older than the 5 newest. xargs rm -f deletes those files.

2. Step 2: Check other options

   ls -1t logfile-*.log.gz | head -n +6 | xargs rm -f deletes the first 6 newest files (wrong). ls -1 logfile-*.log.gz | tail -n 5 | xargs rm -f deletes last 5 files by name order, not time. ls logfile-*.log.gz | grep -v 'logfile-*.log.gz' | xargs rm -f uses grep incorrectly and does not select older files.

3. Final Answer:

   ls -1t logfile-*.log.gz | tail -n +6 | xargs rm -f -> Option A

4. Quick Check:

   Keep 5 newest, delete older = tail from 6th [OK]

Hint: Use ls -t and tail +6 to delete older files beyond 5 [OK]

Common Mistakes:

• Using head instead of tail to select files
• Deleting newest files instead of oldest
• Not sorting files by modification time