Handling Date-Formatted Files in Ansible: Dynamic Path Manipulation Techniques

Working with date-formatted files is a common requirement in automation, especially for log rotation, backups, and time-based file operations. While shell scripts easily handle date +%format commands, Ansible requires different approaches due to its idempotent nature.

Direct shell-style command substitution like date +%y_%m_%d won't work in Ansible tasks because:

Ansible evaluates the playbook before execution
The backtick syntax isn't processed by Jinja2 templating engine
File module parameters need pre-evaluated strings

Using Ansible Facts

- name: Create dated file using ansible_date_time
  file:
    path: "/path/somefile.{{ ansible_date_time.date }}"
    state: touch

Custom Date Formatting

- name: Create file with custom date format
  file:
    path: "/path/somefile.{{ '%Y-%m-%d' | strftime }}"
    state: touch

For Archive Operations

- name: Unzip dated archive
  unarchive:
    src: "/path/zomezip.{{ '%y_%m_%d' | strftime }}.tar.gz"
    dest: /target/path
    remote_src: yes

Variable Pre-calculation

- name: Set date variable
  set_fact:
    today_date: "{{ '%Y_%m_%d' | strftime }}"

- name: Use pre-calculated date
  file:
    path: "/path/backup_{{ today_date }}.log"
    state: touch

Timezone-aware Dates

- name: Get timezone-specific date
  set_fact:
    tz_date: "{{ lookup('pipe', 'TZ=Asia/Tokyo date +%Y-%m-%d') }}"

- name: Create Tokyo-time file
  file:
    path: "/path/tokyo_{{ tz_date }}.log"

Remember that strftime uses Python's formatting, not shell's date
For complex date math, consider using the datetime Python module
Test date formats thoroughly as YAML can interpret certain patterns as numbers

When working with log rotations, backups, or time-stamped archives in Ansible, you'll frequently encounter files named with date patterns like backup_23_11_15.tar.gz. The immediate temptation is to use shell-style command substitution (date +%y_%m_%d) directly in playbooks - but this approach fails because Ansible evaluates Jinja2 templates before shell interpretation.

Instead of shell commands, leverage Ansible's built-in strftime filter combined with the ansible_date_time fact:

- name: Create dated directory
  file:
    path: "/backups/archive_{{ ansible_date_time.date }}"
    state: directory

- name: Extract timestamped archive
  unarchive:
    src: "/path/to/backup_{{ '%y_%m_%d' | strftime }}.tar.gz"
    dest: "/restore_location/"
    remote_src: yes

For more complex date patterns, combine multiple filters:

- name: Process weekly backups
  file:
    path: "/storage/week_{{ ansible_date_time.isoweekday }}_of_{{ '%B' | strftime }}.log"
    state: touch

When working across timezones, explicitly set the reference time:

- name: UTC timestamped operation
  vars:
    utc_time: "{{ '%Y-%m-%dT%H:%M' | strftime(now(utc=true)) }}"
  copy:
    src: "/tmp/datafile"
    dest: "/archive/{{ utc_time }}_dataset.json"

1. Missing Facts: Ensure gather_facts: true (default) or manually set ansible_date_time
2. Windows Compatibility: Use win_file module with %Y%m%d format
3. Idempotency: When creating files, combine with stat module to check existence first

- name: Check if timestamped file exists
  stat:
    path: "/logs/app_{{ '%Y%j' | strftime }}.log"
  register: log_check

- name: Create log if missing
  file:
    path: "/logs/app_{{ '%Y%j' | strftime }}.log"
    state: touch
  when: not log_check.stat.exists