📘

Crontab Explainer

Crontab lets you schedule commands and scripts to run automatically on your Ultra service. Each scheduled task is called a cron job.

The crontab command manages your personal crontab file. The cron service reads this file and runs its entries at their scheduled times.

Test scripts before scheduling them: A script that does not exit correctly, or a schedule that starts a second copy before the first copy finishes, can consume excessive resources and may make your service inaccessible. Use a locking mechanism such as flock when a job must not overlap.

Understand the Crontab Format

A cron entry contains five schedule fields followed by the command:

* * * * * command_to_execute
│ │ │ │ │
│ │ │ │ └─ Day of the week
│ │ │ └─── Month
│ │ └───── Day of the month
│ └─────── Hour
└───────── Minute

Field	Accepted values
Minute	`0-59`
Hour	`0-23`
Day of the month	`1-31`
Month	`1-12`
Day of the week	`0-7`, where `0` and `7` represent Sunday
Command	The command or script to execute

For example, the following entry runs a command every day at 4:30 AM:

30 4 * * * command_to_execute

Confirm the service time zone: Cron uses the service's system time zone, not the time zone of your local computer. Before creating a time-sensitive schedule, execute date '+%F %T %Z %z' over SSH to confirm the current time, time-zone abbreviation, and UTC offset.

Check both day fields: When Day of the month and Day of the week contain restricted values, cron runs the command when either field matches. For example, 30 4 1 * 5 runs at 4:30 AM on the first day of each month and every Friday.

Use crontab.guru to check the meaning of a five-field cron expression.

Use Special Characters

Cron supports the following schedule operators:

matches every possible value.
, separates multiple values. For example, 1,15,30 matches minutes 1, 15, and 30.
specifies an inclusive range. For example, 1-5 in the day-of-week field represents Monday through Friday.
/ specifies an interval within a range. For example, /15 in the minute field runs every 15 minutes.
Comments must begin with # on their own lines:

# Run the cleanup script every day at 03:00
0 3 * * * /home/username/scripts/cleanup.sh

An unescaped % in the command section has a special meaning in cron. Escape a literal percent sign as \%, particularly when using a command such as date inside a cron entry.

Prepare Commands and Scripts

Cron jobs run with a smaller environment than commands entered in an interactive SSH session. Shell startup files may not be loaded, and commands available in your normal shell may not be found automatically.

Use the full path to scripts and important commands.
Use absolute paths for input files, output files, lock files, and logs.
Add an appropriate shebang, such as #!/bin/bash, to the beginning of each script.
Ensure the script is executable.
Create log directories before scheduling the job.
Test the complete command over SSH before adding it to the crontab.
Execute command -v <command> over SSH to locate a command when necessary.
For example, prepare and test a script with the following commands:

mkdir -p /home/username/scripts/logs
chmod +x /home/username/scripts/myscript.sh
/home/username/scripts/myscript.sh

Replace username with your Ultra service username.

Configure Cron Jobs

Connect to your Ultra service via SSH.
Open your personal crontab:

crontab -e

Select nano if you are prompted to choose an editor.
Add each cron job on a separate line.
Press CTRL+X, then Y, and then ENTER to save and exit nano.
Display the installed entries to confirm that the job was saved:

crontab -l

Before making extensive changes, save a backup of your crontab:

crontab -l > /home/username/crontab-backup.txt

Do not remove the entire crontab accidentally: The crontab -r command removes your complete personal crontab without opening the editor. Use crontab -e to remove an individual entry.

Add Cron Jobs

The following entry runs myscript.sh every day at noon and appends standard output and errors to a log:

0 12 * * * /home/username/scripts/myscript.sh >> /home/username/scripts/logs/myscript.log 2>&1

The log directory must exist before the job runs.
To prevent two copies of the script from running simultaneously, use flock:

0 12 * * * /usr/bin/flock -n /home/username/scripts/myscript.lock /home/username/scripts/myscript.sh >> /home/username/scripts/logs/myscript.log 2>&1

The -n option makes flock exit immediately when another copy of the job already holds the lock.
To discard both standard output and errors:

0 2 * * * /home/username/scripts/myscript.sh > /dev/null 2>&1

Keep logs while testing: Discard output only after confirming that the job works correctly. Retaining command output makes failures easier to troubleshoot.

Troubleshoot Cron Jobs

If a job does not run as expected:

Execute the complete command manually over SSH.
Execute date '+%F %T %Z %z' and confirm that the cron expression uses the service's system time.
Execute crontab -l and confirm that the entry was saved.
Confirm that all referenced paths are correct and absolute.
Confirm that the script is executable.
Confirm that the log directory exists and is writable.
Check the job's log for errors.
Add environment variables such as PATH or SHELL at the top of the crontab only when the command requires them.
Confirm that an existing flock lock represents a currently running process before attempting to bypass it.
For a practical scheduled-backup example, see External Backup Script using Rclone Mount.

If you require further assistance, you can open a support ticket here.

Crontab Explainer
Understand the Crontab Format
Use Special Characters
Prepare Commands and Scripts
Configure Cron Jobs
Add Cron Jobs
Troubleshoot Cron Jobs