ansible/playbooks/prod_web.yml

#!/usr/bin/env ansible-playbook
# vim:ft=ansible:
# Webservers
---
- hosts: web1.dallas.mgmt.desu.ltd
  gather_facts: no
  module_defaults:
    docker_container:
      state: started
      restart_policy: unless-stopped
      pull: yes
  tasks:
    - name: ensure docker network
      docker_network: name=web
      tags: [ docker ]
    - name: include tasks for applications
      include_tasks: tasks/{{ item }}
      with_items:
        - app/redis.yml
        - web/9iron.yml
        - web/desultd.yml
        - web/gitea.yml
        - web/nextcloud.yml
        - web/srv.yml
      tags: [ always ]
  roles:
    - role: backup
      vars:
        backup_s3backup_list_extra:
          - /app/gitea/gitea
          - /data
          - /srv/desu.ltd
        backup_s3backup_exclude_list_extra:
          - /var/lib/gitea/log
          - /data/gitea/data/gitea/log
      tags: [ backup ]
    - role: git
      vars:
        git_repos:
          - repo: https://git.desu.ltd/salt/gitea-custom
            dest: /data/gitea/data/gitea/custom
      tags: [ web, git ]
    - role: ingress
      vars:
        ingress_servers:
          # desu.ltd
          - name: desu.ltd
            proxy_pass: http://desultd:80
            locations:
              - location: /.well-known/matrix/server
                contents: |
                  default_type application/json;
                  return 200 '{"m.server":"matrix.desu.ltd:443"}';
              - location: /.well-known/matrix/client
                contents: |
                  default_type application/json;
                  return 200 '{"m.homeserver":{"base_url":"https://matrix.desu.ltd"}}';
          - name: git.desu.ltd
            proxy_pass: http://gitea:3000
          - name: nc.desu.ltd
            directives:
              - "add_header Strict-Transport-Security \"max-age=31536000\""
              - "client_max_body_size 0"
            proxy_pass: http://nextcloud:80
            locations:
              - location: "^~ /.well-known"
                contents: |
                  location = /.well-known/carddav { return 301 /remote.php/dav/; }
                  location = /.well-known/caldav  { return 301 /remote.php/dav/; }
                  location ^~ /.well-known        { return 301 /index.php$uri; }
                  try_files $uri $uri/ =404;
          # 9iron
          - name: www.9iron.club
            directives:
              - "return 301 $scheme://9iron.club$request_uri"
          - name: 9iron.club
            proxy_pass: http://9iron:80
          - name: srv.9iron.club
            proxy_pass: http://srv:80
      tags: [ web, docker, ingress ]
- hosts: web2.dallas.mgmt.desu.ltd
  gather_facts: no
  module_defaults:
    docker_container:
      state: started
      restart_policy: unless-stopped
      pull: yes
  pre_tasks:
    - name: ensure docker network
      docker_network: name=web
      tags: [ docker ]
    - name: include tasks for applications
      include_tasks: tasks/{{ item }}
      with_items:
        - app/redis.yml
        - web/jenkins.yml
        - web/pleroma.yml
      tags: [ always ]
  roles:
    - role: backup
      vars:
        backup_s3backup_list_extra:
          - /data
      tags: [ backup ]
    - role: ingress
      vars:
        ingress_servers:
          - name: cowfee.moe
            proxy_pass: http://pleroma:4000
          - name: jenkins.desu.ltd
            locations:
              - location: "/"
                contents: |
                  proxy_set_header Host $host;
                  proxy_set_header X-Real-IP $remote_addr;
                  proxy_set_header X-Forwarded-Proto https;
                  proxy_set_header X-Forwarded-Port 443;
                  proxy_pass http://jenkins:8080;
              - location: |
                  ~ "^/static/[0-9a-fA-F]{8}\/(.*)$"
                contents: |
                  rewrite "^/static/[0-9a-fA-F]{8}\/(.*)" /$1 last;
              - location: "/userContent"
                contents: |
                  root /data/jenkins/home/;
                  if (!-f $request_filename) {
                    rewrite (.*) /$1 last;
                    break;
                  }
      tags: [ web, docker, ingress ]
- hosts: web3.dallas.mgmt.desu.ltd
  module_defaults:
    docker_container:
      state: started
      restart_policy: unless-stopped
      pull: yes
  tasks:
    - name: ensure docker network
      docker_network: name=web
      tags: [ docker ]
    - name: include tasks for applications
      include_tasks: tasks/{{ item }}
      with_items:
        - app/redis.yml
        - web/netbox.yml
        - web/ara.yml
      tags: [ always ]
  roles:
    - role: backup
      vars:
        backup_s3backup_list_extra:
          - /data
      tags: [ backup ]
      # TODO: Replace this with Naemon(?)
    - role: nagios
      vars:
        nagios_matrix_server: "https://matrix.desu.ltd"
        nagios_matrix_room: "!NWNCKlNmOTcarMcMIh:desu.ltd"
        nagios_matrix_token: "{{ secret_nagios_matrix_token }}"
        nagios_data_dir: /data/nagios
        nagios_admin_pass: "{{ secret_nagios_admin_pass }}"
        nagios_contacts:
          - name: matrix
            host_notification_commands: notify-host-by-matrix
            service_notification_commands: notify-service-by-matrix
            host_notification_period: ansible-not-late-at-night
            service_notification_period: ansible-not-late-at-night
            extra:
              - key: contactgroups
                value: ansible
          - name: salt
            host_notification_commands: notify-host-by-email
            service_notification_commands: notify-service-by-email
            extra:
              - key: email
                value: alerts@babor.tech
        nagios_commands:
          # This command is included in the container image
          - name: check_nrpe
            command: "$USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$"
          - name: check_by_ssh
            command: "$USER1$/check_by_ssh -H $HOSTADDRESS$ -F /opt/nagios/etc/ssh_config -t 30 -q -i /opt/nagios/etc/id_ed25519 -l nagios-checker -C \"$ARG1$\""
          - name: notify-host-by-matrix
            command: "/usr/bin/printf \"%b\" \"$NOTIFICATIONTYPE$\\n$HOSTNAME$ is $HOSTSTATE$\\nAddress: $HOSTADDRESS$\\nInfo: $HOSTOUTPUT$\\nDate/Time: $LONGDATETIME$\" | /opt/Custom-Nagios-Plugins/notify-by-matrix"
          - name: notify-service-by-matrix
            command: "/usr/bin/printf \"%b\" \"$NOTIFICATIONTYPE$\\nService $HOSTALIAS$ - $SERVICEDESC$ is $SERVICESTATE$\\nInfo: $SERVICEOUTPUT$\\nDate/Time: $LONGDATETIME$\" | /opt/Custom-Nagios-Plugins/notify-by-matrix"
        nagios_services:
          # Agentless checks
          - name: HTTP
            command: check_http
            hostgroup: tag-nagios-checkhttp
          - name: HTTPS
            command: check_http!--ssl
            hostgroup: tag-nagios-checkhttp
          - name: SSH
            command: check_ssh
          # check_by_ssh checks
          - name: CPU Utilization
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_cpu_stats -w 75 -c 90
          - name: DNS Resolution
            command: check_by_ssh!/usr/lib/nagios/plugins/check_etc_resolv
          - name: Executables in tmp
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_executables_in_tmp
          - name: Last Ansible Play
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_file_age /var/lib/ansible-last-run -w 432000 -c 604800
          - name: Memory Usage
            command: check_by_ssh!/usr/lib/nagios/plugins/check_memory -w 10% -c 5%
            hostgroup: "ansible,!tag-prov-zfs"
          - name: Ping Self over DNS
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_ping_by_hostname
          - name: Reboot Required
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_reboot_required
          - name: Unit atd.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit atd.service
          - name: Unit backup.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit backup.service
            hostgroup: "ansible,!role-hypervisor"
          - name: Unit backup.timer
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit backup.timer
            hostgroup: "ansible,!role-hypervisor"
          - name: Unit cron.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit cron.service
          - name: Unit dbus.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit dbus.service
          - name: Unit ssh.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit ssh.service
          - name: Unit systemd-resolved.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit systemd-resolved.service
            hostgroup: "ansible,!role-hypervisor"
          - name: Users
            command: check_by_ssh!/usr/lib/nagios/plugins/check_users -w 3 -c 5
          # Privileged checks
          # Required because check_disk may attempt to get the free space of
          # restricted mountpoints
          - name: Disk Usage
            command: check_by_ssh!/usr/bin/sudo /usr/lib/nagios/plugins/check_disk -M -u GB -X nfs -X tracefs -X cgroup -X tmpfs -X overlay -X shm -w 15% -c 10% -W 15% -K 10% -A -I '^/run/' -I '^udev$' -I '^/var/lib/kubelet/' -I '^/tmp/.mount_' -I '^/dev/loop'
          # Device type checks
          # R720
          - name: CPU0 Temperature
            command: check_by_ssh!/usr/bin/sudo /usr/local/bin/monitoring-scripts/check_temp -n -w 65 -c 75 --sensor coretemp-isa-0000
            hostgroup: device-type-r720
          - name: CPU1 Temperature
            command: check_by_ssh!/usr/bin/sudo /usr/local/bin/monitoring-scripts/check_temp -n -w 65 -c 75 --sensor coretemp-isa-0001
            hostgroup: device-type-r720
          # Pi 4 4G
          - name: CPU Temperature
            command: check_by_ssh!/usr/bin/sudo /usr/local/bin/monitoring-scripts/check_temp -n -w 65 -c 75 --sensor cpu_thermal-virtual-0
            hostgroup: device-type-pi4b-2g,device-type-pi4b-4g,device-type-pi4b-4g-storage
          # Device role checks
          # hypervisor (which is assumed to be Proxmox)
          - name: PVE Unit pve-firewall.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pve-firewall.service
            hostgroup: role-hypervisor
          - name: PVE Unit spiceproxy.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit spiceproxy.service
            hostgroup: role-hypervisor
          - name: PVE Unit pve-ha-crm.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pve-ha-crm.service
            hostgroup: role-hypervisor
          - name: PVE Unit pvedaemon.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pvedaemon.service
            hostgroup: role-hypervisor
          - name: PVE Unit pvefw-logger.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pvefw-logger.service
            hostgroup: role-hypervisor
          - name: PVE Unit pveproxy.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pveproxy.service
            hostgroup: role-hypervisor
          - name: PVE Unit pve-cluster.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pve-cluster.service
            hostgroup: role-hypervisor
          - name: PVE Unit pvestatd.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit pvestatd.service
            hostgroup: role-hypervisor
          # Tag-specific checks
          # ansible-pull
          - name: Unit ansible-pull.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit ansible-pull.service
            hostgroup: tag-ansible-pull
          - name: Unit ansible-pull.timer
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit ansible-pull.timer
            hostgroup: tag-ansible-pull
          # docker
          - name: Unit docker.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit docker.service
            hostgroup: "ansible,!tag-no-docker"
          - name: Docker Status
            command: check_by_ssh!/usr/bin/sudo /usr/local/bin/monitoring-scripts/check_docker --no-ok --status running
            hostgroup: tag-nagios-checkdocker
          - name: Docker CPU Usage
            command: check_by_ssh!/usr/bin/sudo /usr/local/bin/monitoring-scripts/check_docker --no-ok --cpu 85:90
            hostgroup: tag-nagios-checkdocker
          - name: Docker Memory Usage
            command: check_by_ssh!/usr/bin/sudo /usr/local/bin/monitoring-scripts/check_docker --no-ok --memory 90:95:%
            hostgroup: tag-nagios-checkdocker
          # nagios-checkpgsql
          - name: PSQL
            command: "check_by_ssh!/usr/lib/nagios/plugins/check_pgsql -H localhost -l nagios -p {{ secret_postgresql_monitoring_password }} -w 2 -c 5"
            hostgroup: tag-nagios-checkpgsql
          - name: PSQL Connections
            command: "check_by_ssh!/usr/lib/nagios/plugins/check_pgsql -H localhost -l nagios -p {{ secret_postgresql_monitoring_password }} -w 2 -c 5 -q 'select (select count(*)::float used from pg_stat_activity) / (select setting::int max_conn from pg_settings where name=\\$\\$max_connections\\$\\$)' -W 0.7-0.8 -C 0.8-1.0"
            hostgroup: tag-nagios-checkpgsql
            # https://rhaas.blogspot.com/2020/02/useless-vacuuming.html
          - name: PSQL Old Xacts
            command: "check_by_ssh!/usr/lib/nagios/plugins/check_pgsql -H localhost -l nagios -p {{ secret_postgresql_monitoring_password }} -w 2 -c 5 -q 'select count(*)::float from pg_prepared_xacts where age(transaction) > 5000000' -W 500-1000 -C 1000-1000000"
            hostgroup: tag-nagios-checkpgsql
          - name: Unit postgresql.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit postgresql.service
            hostgroup: tag-nagios-checkpgsql
          # nagios-checkswap
          - name: Swap Usage
            command: check_by_ssh!/usr/lib/nagios/plugins/check_swap -w 20% -c 10%
            hostgroup: tag-nagios-checkswap
          # zerotier
          - name: Unit zerotier-one.service
            command: check_by_ssh!/usr/local/bin/monitoring-scripts/check_systemd_unit zerotier-one.service
            hostgroup: tag-zt-personal
      tags: [ nagios, no-auto ]
    - role: ingress
      vars:
        ingress_servers:
          - name: ara.desu.ltd
            proxy_pass: http://ara:8000
          - name: netbox.desu.ltd
            proxy_pass: http://netbox:8080
          - name: nagios.desu.ltd
            proxy_pass: http://nagios:80
      tags: [ web, docker, ingress ]