Borgmatic zabbix template

---

This is an active zabbix template for monitoring borgmatic tasks.

How does it work

• Import the template on the zabbix server
• Apply the template on the client on zabbix server
• Create a file containing the tasks should be written at the client. For example a file created as /etc/ansible/facts.d/borgmatic_tasks.fact:

  {
      "data": [
          {
              "name": "borgmatic_task_name_1",
              "repository": "database",
              "server": "borg_server1.example.com",
              "user": "backup",
              "zabbix_item_key": "db_physical_1:ssh://backup@borg_server1.example.com/./database",
              "zabbix_nodata_warning_threshold": "28h"
          },
          {
              "name": "borgmatic_task_name_2",
              "repository": "database",
              "server": "borg_server2.example.com",
              "user": "backup",
              "zabbix_item_key": "borgmatic_task_name_2:ssh://backup@borg_server2.example.com/./database",
              "zabbix_nodata_warning_threshold": "28h"
          },
          {
              "name": "borgmatic_task_name_3",
              "repository": "binlogs",
              "server": "borg_server1.example.com",
              "user": "backup",
              "zabbix_item_key": "borgmatic_task_name_3:ssh://backup@borg_server1.example.com/./binlogs",
              "zabbix_nodata_warning_threshold": "1d"
          },
          {
              "name": "borgmatic_task_name_3",
              "repository": "binlogs",
              "server": "borg_server2.example.com",
              "user": "backup",
              "zabbix_item_key": "borgmatic_task_name_3:ssh://backup@borg_server2.example.com/./binlogs",
              "zabbix_nodata_warning_threshold": "1d"
          }
      ]
  }

• Run the following on the client

  cd /etc/zabbix && zabbix_sender -vv -c /etc/zabbix/zabbix_agent2.conf -k borgmatic.discovery.task -o "$(cat /etc/ansible/facts.d/borgmatic_tasks.fact)"

• Add the following line under the after_backup, on_error, after_everything and before_everything hook on the borgmatic task file:

    after_backup:
        - echo "`date` - Finished backup."
        # - other custom commands ....
        # Change ${BORGMATIC_TASK_NAME} to the name of the borgmatic configuration file
        - cd /etc/zabbix;/usr/bin/zabbix_sender -c /etc/zabbix/zabbix_agent2.conf -k borgmatic.task.repository.status[${BORGMATIC_TASK_NAME}:{repository}] -o 1 || true
    on_error:
        - echo "`date` - Error while creating a backup."
        - cd /etc/zabbix;/usr/bin/zabbix_sender -c /etc/zabbix/zabbix_agent2.conf -k borgmatic.task.repository.status[${BORGMATIC_TASK_NAME}:{repository}] -o 0 || true
    # Optionally if you use scripts or command in before_everything or after_everything add
    before_everything:
        - myscript || (exit_code="$?";/usr/bin/jq -rc '.data[]|select(.name=="${BORGMATIC_TASK_NAME}").zabbix_item_key' /etc/ansible/facts.d/borgmatic_tasks.fact|while read zabbix_item_key;do cd /etc/zabbix; /usr/bin/zabbix_sender -c /etc/zabbix/zabbix_agent2.conf -k borgmatic.task.repository.status[$zabbix_item_key] -o 0 || true; done; exit "$exit_code")
    after_everything:
        - myscript || (exit_code="$?";/usr/bin/jq -rc '.data[]|select(.name=="${BORGMATIC_TASK_NAME}").zabbix_item_key' /etc/ansible/facts.d/borgmatic_tasks.fact|while read zabbix_item_key;do cd /etc/zabbix; /usr/bin/zabbix_sender -c /etc/zabbix/zabbix_agent2.conf -k borgmatic.task.repository.status[$zabbix_item_key] -o 0 || true; done; exit "$exit_code")

Every time the borgmatic task runs it will send a status flag to the zabbix server. 0 means failure, 1 success. If the server has not received an item for more than zabbix_nodata_warning_threshold an alert will be triggered.