Rework portfolio around Linux operations, Zabbix monitoring, migration validation, and ELK/Grafana log observability. Add AAP-style LVM resize workflow, Zabbix server/proxy/agent automation assets, Linux/AIX monitoring templates, and updated validation CI.
This commit is contained in:
@@ -0,0 +1,177 @@
|
||||
---
|
||||
- name: Validate decommissioning requirements
|
||||
ansible.builtin.assert:
|
||||
that:
|
||||
- backup_data or not backup_data
|
||||
fail_msg: "Invalid decommissioning configuration"
|
||||
|
||||
- name: Pre-decommissioning checks
|
||||
block:
|
||||
- name: Check node health
|
||||
ansible.builtin.uri:
|
||||
url: http://localhost/health
|
||||
method: GET
|
||||
status_code: 200
|
||||
register: health_check
|
||||
failed_when: false
|
||||
ignore_errors: true
|
||||
when: "'webservers' in group_names"
|
||||
|
||||
- name: Create decommissioning backup directory
|
||||
ansible.builtin.file:
|
||||
path: "/var/backups/decommission-{{ ansible_date_time.iso8601 }}"
|
||||
state: directory
|
||||
mode: '0755'
|
||||
|
||||
- name: Initialize decommissioning log
|
||||
ansible.builtin.file:
|
||||
path: "/var/log/decommission.log"
|
||||
state: touch
|
||||
mode: '0644'
|
||||
modification_time: now
|
||||
access_time: now
|
||||
|
||||
- name: Log decommissioning start
|
||||
ansible.builtin.lineinfile:
|
||||
path: "/var/log/decommission.log"
|
||||
line: "{{ ansible_date_time.iso8601 }} - Starting decommissioning of {{ inventory_hostname }}"
|
||||
state: present
|
||||
|
||||
- name: Graceful application shutdown
|
||||
block:
|
||||
- name: Stop application services
|
||||
ansible.builtin.service:
|
||||
name: "{{ item }}"
|
||||
state: stopped
|
||||
loop: "{{ application_services }}"
|
||||
failed_when: false
|
||||
when: graceful_shutdown
|
||||
|
||||
- name: Wait for connections to drain
|
||||
ansible.builtin.pause:
|
||||
seconds: 30
|
||||
when: graceful_shutdown and ("webservers" in group_names or "loadbalancers" in group_names)
|
||||
|
||||
- name: Export and backup data
|
||||
block:
|
||||
- name: Create config export directory
|
||||
ansible.builtin.file:
|
||||
path: "/var/backups/decommission-{{ ansible_date_time.iso8601 }}/config"
|
||||
state: directory
|
||||
mode: '0755'
|
||||
|
||||
- name: Archive system configuration
|
||||
community.general.archive:
|
||||
path: "{{ config_paths }}"
|
||||
dest: "/var/backups/decommission-{{ ansible_date_time.iso8601 }}/config/system_config.tar.gz"
|
||||
format: gz
|
||||
when: export_config
|
||||
failed_when: false # noqa risky-file-permissions
|
||||
|
||||
- name: Create data backup directory
|
||||
ansible.builtin.file:
|
||||
path: "/var/backups/decommission-{{ ansible_date_time.iso8601 }}/data"
|
||||
state: directory
|
||||
mode: '0755'
|
||||
when: backup_data
|
||||
|
||||
- name: Backup individual data paths
|
||||
community.general.archive:
|
||||
path: "{{ item }}"
|
||||
dest: "/var/backups/decommission-{{ ansible_date_time.iso8601 }}/data/{{ item | regex_replace('/', '_') }}.tar.gz"
|
||||
format: gz
|
||||
loop: "{{ data_paths }}"
|
||||
when: backup_data
|
||||
failed_when: false # noqa risky-file-permissions
|
||||
|
||||
- name: Update monitoring and load balancing
|
||||
block:
|
||||
- name: Remove from load balancer
|
||||
ansible.builtin.debug:
|
||||
msg: "Would remove {{ inventory_hostname }} from load balancer"
|
||||
when: "'webservers' in group_names or 'databases' in group_names"
|
||||
|
||||
- name: Update monitoring alerts
|
||||
ansible.builtin.debug:
|
||||
msg: "Would update monitoring alerts for {{ inventory_hostname }}"
|
||||
when: "'monitoring' not in group_names"
|
||||
|
||||
- name: Clean up application
|
||||
block:
|
||||
- name: Remove application directories
|
||||
ansible.builtin.file:
|
||||
path: "{{ item }}"
|
||||
state: absent
|
||||
loop:
|
||||
- /opt/application
|
||||
- /var/www/html
|
||||
- /var/lib/postgresql
|
||||
- /var/lib/prometheus
|
||||
failed_when: false
|
||||
|
||||
- name: Remove application packages
|
||||
ansible.builtin.apt:
|
||||
name: "{{ item }}"
|
||||
state: absent
|
||||
purge: true
|
||||
loop: "{{ application_packages }}"
|
||||
failed_when: false
|
||||
|
||||
- name: Clean system logs
|
||||
ansible.builtin.shell: |
|
||||
set -o pipefail
|
||||
find /var/log -name "*.log" -type f -size +0 -exec truncate -s 0 {} \;
|
||||
changed_when: false
|
||||
failed_when: false
|
||||
|
||||
- name: Remove SSH credentials
|
||||
ansible.builtin.file:
|
||||
path: "{{ item }}"
|
||||
state: absent
|
||||
loop:
|
||||
- /root/.ssh/authorized_keys
|
||||
- /root/.ssh/known_hosts
|
||||
- /home/infra-admin/.ssh/authorized_keys
|
||||
failed_when: false
|
||||
|
||||
- name: Generate decommissioning report
|
||||
ansible.builtin.template:
|
||||
src: decommission_report.j2
|
||||
dest: "/var/log/decommission_report_{{ ansible_date_time.iso8601 }}.log"
|
||||
mode: '0644'
|
||||
vars:
|
||||
backup_location: "/var/backups/decommission-{{ ansible_date_time.iso8601 }}"
|
||||
|
||||
- name: Send decommissioning notification
|
||||
community.general.mail:
|
||||
host: localhost
|
||||
port: 25
|
||||
to: "{{ decommission_notification_email }}"
|
||||
subject: "Node Decommissioned - {{ inventory_hostname }}"
|
||||
body: |
|
||||
Node {{ inventory_hostname }} has been successfully decommissioned.
|
||||
|
||||
Backup location: /var/backups/decommission-{{ ansible_date_time.iso8601 }}/
|
||||
Services stopped: {{ application_services | join(', ') }}
|
||||
Configuration exported: {{ export_config }}
|
||||
Data backed up: {{ backup_data }}
|
||||
|
||||
See /var/log/decommission_report_{{ ansible_date_time.iso8601 }}.log for details
|
||||
when: decommission_notification_email is defined
|
||||
failed_when: false
|
||||
|
||||
- name: Finalize decommissioning
|
||||
block:
|
||||
- name: Log decommissioning completion
|
||||
ansible.builtin.lineinfile:
|
||||
path: "/var/log/decommission.log"
|
||||
line: "{{ ansible_date_time.iso8601 }} - Decommissioning completed for {{ inventory_hostname }}"
|
||||
state: present
|
||||
|
||||
- name: Perform system shutdown
|
||||
ansible.builtin.reboot:
|
||||
msg: "System scheduled for shutdown after decommissioning"
|
||||
delay: "{{ shutdown_delay }}"
|
||||
when: auto_shutdown | bool
|
||||
async: 1
|
||||
poll: 0
|
||||
Reference in New Issue
Block a user