at command for one-time jobs in Linux CLI - Commands & Configuration

Introduction

Sometimes you want your computer to do a task just once at a specific time later. The 'at' command lets you schedule these one-time jobs easily without needing to keep the terminal open.

When you want to run a backup script at 2 AM tonight without staying logged in.

When you need to send an email reminder automatically at a future time.

When you want to shut down your computer after finishing a long download.

When you want to run a cleanup script once after office hours.

When you want to test a command later without running it immediately.

Commands

This schedules a simple command to run at 2 AM today. It writes 'Backup started' into a file named backup.log. We use echo and pipe it to 'at' to schedule the job.

Terminal

echo "echo 'Backup started' > backup.log" | at 02:00

Expected Output

job 1 at Thu Apr 27 02:00:00 2024

This command lists all pending jobs scheduled with 'at'. It helps you check what jobs are waiting to run.

Terminal

atq

Expected Output

1	Thu Apr 27 02:00:00 2024 a user

This removes the scheduled job with ID 1. Use this if you want to cancel a job before it runs.

Terminal

atrm 1

Expected Output

No output (command runs silently)

This starts an interactive prompt to schedule commands at 2:30 PM. You type the commands you want to run, then press Ctrl+D to save and exit.

Terminal

at 14:30

Expected Output

warning: commands will be executed using /bin/sh

at> echo 'Meeting reminder' > reminder.txt

at> <Ctrl+D>

job 2 at Thu Apr 27 14:30:00 2024

Key Concept

If you remember nothing else from this pattern, remember: the 'at' command schedules one-time tasks to run later without needing a running terminal.

Common Mistakes

Typing the time in a wrong format like '25:00' or '2pm' without spaces.

The 'at' command expects valid time formats; wrong formats cause errors or unexpected scheduling.

Use 24-hour format like '14:00' or natural language like '2 pm' with a space.

Not pressing Ctrl+D after typing commands in the interactive 'at' prompt.

Without Ctrl+D, the job is not saved and scheduled, so the commands never run.

Always press Ctrl+D to signal the end of input and save the job.

Trying to schedule a job without having the 'at' daemon running or installed.

The 'at' command depends on a background service; if it's missing, scheduling fails silently or with errors.

Ensure 'atd' service is installed and running before scheduling jobs.

Summary

Use 'echo "command" | at time' to schedule a one-time job quickly.

Check scheduled jobs with 'atq' and remove them with 'atrm job_id'.

Use 'at time' to enter interactive mode for multiple commands, ending with Ctrl+D.

Practice

1. What is the primary purpose of the at command in Linux?

easy

A. To list all running processes

B. To schedule a one-time job to run at a specific time

C. To monitor system resource usage continuously

D. To create a recurring job every day

Solution

1. Step 1: Understand the at command function

   The at command is designed to schedule commands or scripts to run once at a specified future time.

2. Step 2: Compare with other options

   Options A, C, and D describe other commands or functionalities unrelated to at. For example, listing processes is done by ps, monitoring resources by top, and recurring jobs by cron.

3. Final Answer:

   To schedule a one-time job to run at a specific time -> Option B

4. Quick Check:

   at schedules one-time jobs = B [OK]

Hint: Remember: at is for one-time future tasks [OK]

Common Mistakes:

• Confusing at with cron for recurring jobs
• Thinking at lists running processes
• Assuming at monitors system resources

2. Which of the following is the correct syntax to schedule a job at 3:30 PM using the at command?

easy

A. at 3-30

B. at 3.30pm

C. at 15:30pm

D. at 3:30pm

Solution

1. Step 1: Identify valid time format for at

   The at command accepts time in formats like HH:MMam or HH:MMpm. So 3:30pm is valid.

2. Step 2: Check other options for syntax errors

   B: at 3.30pm uses dot instead of colon, invalid. C: at 15:30pm mixes 24-hour format with pm, invalid. D: at 3-30 uses dash instead of colon, invalid.

3. Final Answer:

   at 3:30pm -> Option D

4. Quick Check:

   Correct at time format = A [OK]

Hint: Use colon and am/pm for time with at [OK]

Common Mistakes:

• Using dot or dash instead of colon in time
• Omitting am/pm in 12-hour format
• Confusing 24-hour and 12-hour formats

3. What will be the output of the following commands sequence?

echo "echo Hello World" | at now + 1 minute
sleep 70
atq

medium

A. No jobs listed because the job already ran

B. Shows an error because at cannot read from pipe

C. Lists the scheduled job with its job number

D. Prints 'Hello World' immediately

Solution

1. Step 1: Understand the job scheduling and timing

   The command schedules a job to run 1 minute from now that prints 'Hello World'. Then the script waits 70 seconds (more than 1 minute), so the job runs during sleep.

2. Step 2: Check the atq output after sleep

   Since the job already executed, atq will show no pending jobs.

3. Final Answer:

   No jobs listed because the job already ran -> Option A

4. Quick Check:

   Job runs after 1 min, sleep 70s, then atq empty = C [OK]

Hint: Wait past scheduled time, then atq shows no jobs [OK]

Common Mistakes:

• Expecting atq to list completed jobs
• Thinking at cannot read from pipe
• Assuming immediate output from scheduled job

4. You tried to schedule a job with at 25:00 but got an error. What is the likely cause and how to fix it?

medium

A. You must run at as root user

B. Missing command after at; add a command

C. Invalid hour; use a valid 24-hour time like 23:00

D. The at daemon is not installed

Solution

1. Step 1: Identify the error cause

   The hour 25 is invalid because valid 24-hour times range from 00 to 23.

2. Step 2: Correct the time format

   Use a valid time like 23:00 or a 12-hour format with am/pm.

3. Final Answer:

   Invalid hour; use a valid 24-hour time like 23:00 -> Option C

4. Quick Check:

   Hour must be 0-23 in at time = A [OK]

Hint: Hours must be 0-23 in 24-hour format for at [OK]

Common Mistakes:

• Using invalid hour values above 23
• Assuming at needs root for scheduling
• Confusing missing command error with time format error

5. You want to schedule a script /home/user/backup.sh to run once at 11:45 PM today and then remove it from the queue after scheduling. Which sequence of commands achieves this correctly?

hard

A. echo "/home/user/backup.sh" | at 11:45pm && atrm $(atq | awk '{print $1}')

B. at 11:45pm < /home/user/backup.sh && atrm $(atq | awk '{print $1}')

C. at 11:45pm <<< "/home/user/backup.sh" && atrm $(atq | awk '{print $1}')

D. echo "/home/user/backup.sh" | at 11:45pm && atq && atrm $(atq | awk '{print $1}')

Solution

1. Step 1: Schedule the script using echo piped to at

   This schedules the command to run at 11:45pm by piping the script path as a command.

2. Step 2: Remove the job from the queue immediately after scheduling

   Use atrm with the job number obtained from atq. echo "/home/user/backup.sh" | at 11:45pm && atrm $(atq | awk '{print $1}') correctly schedules and then removes the job.

3. Final Answer:

   echo "/home/user/backup.sh" | at 11:45pm && atrm $(atq | awk '{print $1}') -> Option A

4. Quick Check:

   Pipe command to at, then remove job = D [OK]

Hint: Pipe command to at, then remove job with atrm [OK]

Common Mistakes:

• Using input redirection with script file instead of command
• Removing job before it is scheduled
• Confusing atq output format for job ID