安装oVirt 4.5.4超融合集群时遇'str object'无'vgname'属性错误

环境信息 #

• 部署目标：3节点oVirt 4.5.4超融合集群，每节点配备1块SSD（用于安装oVirt节点）+1块HDD
• 本地环境：Python 3.11.2、Ansible 2.14.2
• 使用角色：gluster.infra/roles/backend_setup（与官方仓库内容一致）

问题描述 #

执行自定义Playbook时，在gluster.infra/roles/backend_setup/tasks/get_vg_groupings.yml的分组任务阶段触发致命错误，所有节点均提示'str object' has no attribute 'vgname'，多次尝试无法解决。

Playbook配置 #

hc_nodes:
  hosts:
    node210.com:
      gluster_infra_volume_groups:
        - vgname: gluster_vg_sda
          pvname: /dev/sda
      gluster_infra_mount_devices:
        - path: /gluster_bricks/engine
          lvname: gluster_lv_engine
          vgname: gluster_vg_sda
        - path: /gluster_bricks/data
          lvname: gluster_lv_data
          vgname: gluster_vg_sda
        - path: /gluster_bricks/vmstore
          lvname: gluster_lv_vmstore
          vgname: gluster_vg_sda
      blacklist_mpath_devices:
        - sda
      gluster_infra_thick_lvs:
        - vgname: gluster_vg_sda
          lvname: gluster_lv_engine
          size: 100G
      gluster_infra_thinpools:
        - vgname: gluster_vg_sda
          thinpoolname: gluster_thinpool_gluster_vg_sda
          poolmetadatasize: 1G
      gluster_infra_lv_logicalvols:
        - vgname: gluster_vg_sda
          thinpool: gluster_thinpool_gluster_vg_sda
          lvname: gluster_lv_data
          lvsize: 150G
        - vgname: gluster_vg_sda
          thinpool: gluster_thinpool_gluster_vg_sda
          lvname: gluster_lv_vmstore
          lvsize: 150G
    node211.com:
      gluster_infra_volume_groups:
        - vgname: gluster_vg_sda
          pvname: /dev/sda
      gluster_infra_mount_devices:
        - path: /gluster_bricks/engine
          lvname: gluster_lv_engine
          vgname: gluster_vg_sda
        - path: /gluster_bricks/data
          lvname: gluster_lv_data
          vgname: gluster_vg_sda
        - path: /gluster_bricks/vmstore
          lvname: gluster_lv_vmstore
          vgname: gluster_vg_sda
      blacklist_mpath_devices:
        - sda
      gluster_infra_thick_lvs:
        - vgname: gluster_vg_sda
          lvname: gluster_lv_engine
          size: 100G
      gluster_infra_thinpools:
        - vgname: gluster_vg_sda
          thinpoolname: gluster_thinpool_gluster_vg_sda
          poolmetadatasize: 1G
      gluster_infra_lv_logicalvols:
        - vgname: gluster_vg_sda
          thinpool: gluster_thinpool_gluster_vg_sda
          lvname: gluster_lv_data
          lvsize: 150G
        - vgname: gluster_vg_sda
          thinpool: gluster_thinpool_gluster_vg_sda
          lvname: gluster_lv_vmstore
          lvsize: 150G
    node212.com:
      gluster_infra_volume_groups:
        - vgname: gluster_vg_sda
          pvname: /dev/sda
      gluster_infra_mount_devices:
        - path: /gluster_bricks/engine
          lvname: gluster_lv_engine
          vgname: gluster_vg_sda
        - path: /gluster_bricks/data
          lvname: gluster_lv_data
          vgname: gluster_vg_sda
        - path: /gluster_bricks/vmstore
          lvname: gluster_lv_vmstore
          vgname: gluster_vg_sda
      blacklist_mpath_devices:
        - sda
      gluster_infra_thick_lvs:
        - vgname: gluster_vg_sda
          lvname: gluster_lv_engine
          size: 100G
      gluster_infra_thinpools:
        - vgname: gluster_vg_sda
          thinpoolname: gluster_thinpool_gluster_vg_sda
          poolmetadatasize: 1G
      gluster_infra_lv_logicalvols:
        - vgname: gluster_vg_sda
          thinpool: gluster_thinpool_gluster_vg_sda
          lvname: gluster_lv_data
          lvsize: 150G
        - vgname: gluster_vg_sda
          thinpool: gluster_thinpool_gluster_vg_sda
          lvname: gluster_lv_vmstore
          lvsize: 150G
  vars:
    gluster_infra_disktype: JBOD
    gluster_set_selinux_labels: true
    gluster_infra_fw_ports:
      - 2049/tcp
      - 54321/tcp
      - 5900/tcp
      - 5900-6923/tcp
      - 5666/tcp
      - 16514/tcp
    gluster_infra_fw_permanent: true
    gluster_infra_fw_state: enabled
    gluster_infra_fw_zone: public
    gluster_infra_fw_services:
      - glusterfs
    gluster_features_force_varlogsizecheck: false
    cluster_nodes:
      - node210.com
      - node211.com
      - node212.com
    gluster_features_hci_cluster: '{{ cluster_nodes }}'
    gluster_features_hci_volumes:
      - volname: engine
        brick: /gluster_bricks/engine/engine
        arbiter: 0
      - volname: data
        brick: /gluster_bricks/data/data
        arbiter: 0
      - volname: vmstore
        brick: /gluster_bricks/vmstore/vmstore
        arbiter: 0

错误日志 #

TASK [gluster.infra/roles/backend_setup : Group devices by volume group name, including existing devices] ***
task path: /etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/get_vg_groupings.yml:3
fatal: [node210.com]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'str object' has no attribute 'vgname'. 'str object' has no attribute 'vgname'\n\nThe error appears to be in '/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/get_vg_groupings.yml': line 3, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: Group devices by volume group name, including existing devices\n  ^ here\n"}
fatal: [node211.com]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'str object' has no attribute 'vgname'. 'str object' has no attribute 'vgname'\n\nThe error appears to be in '/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/get_vg_groupings.yml': line 3, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: Group devices by volume group name, including existing devices\n  ^ here\n"}
fatal: [node212.com]: FAILED! => {"msg": "The task includes an option with an undefined variable. The error was: 'str object' has no attribute 'vgname'. 'str object' has no attribute 'vgname'\n\nThe error appears to be in '/etc/ansible/roles/gluster.infra/roles/backend_setup/tasks/get_vg_groupings.yml': line 3, column 3, but may\nbe elsewhere in the file depending on the exact syntax problem.\n\nThe offending line appears to be:\n\n\n- name: Group devices by volume group name, including existing devices\n  ^ here\n"}

get_vg_groupings.yml任务文件内容 #

---

- name: Group devices by volume group name, including existing devices
  set_fact:
   gluster_volumes_by_groupname_pre: >-
    {%- set output={} -%}
    {%- for grouper, devicesConf in volume_groups | groupby('vgname') -%}
      {%- set confs=[] -%}    
      {%- for deviceConf in devicesConf -%}
        {%- if deviceConf.pvname is defined -%}
          {%- for device in deviceConf.pvname.split(',') -%}
            {%- set deviceConfCln=dict(devicesConf | first) -%}
            {{- deviceConfCln.__setitem__('pvname',device | trim) -}}
            {{- confs.append(deviceConfCln) -}}
          {%- endfor -%}
        {%- endif -%}
      {%- endfor -%}
      {{- output.__setitem__(grouper, confs) -}}      
    {%- endfor -%}
    {%- if  hostvars[inventory_hostname].ansible_lvm is defined and hostvars[inventory_hostname].ansible_lvm.pvs is defined -%}
      {%- for device, pvs in hostvars[inventory_hostname].ansible_lvm.pvs.items() -%}
        {%- if pvs.vg in output and output[pvs.vg] | selectattr('pvname','equalto',device) | list | count == 0 -%}
          {{- output[pvs.vg].append({'pvname':device, 'vgname':pvs.vg }) -}}
        {%- endif -%}
      {%- endfor -%}
    {%- endif -%}
    {{- output | to_json -}}    

- name: Check if vg block device exists
  shell: >
   {% for pvsname in item.value | ovirt.ovirt.json_query('[].pvname')  %}
   test -b {{ pvsname }} && echo "1" || echo  "0";
   {% endfor %}
  register: vg_device_exists
  loop: "{{gluster_volumes_by_groupname_pre | dict2items}}"

- name: Filter none-existing devices
  set_fact:
   gluster_volumes_by_groupname: >-
    {%- set output={} -%}
    {%- for vgname, devicesConf in gluster_volumes_by_groupname_pre.items() -%}
      {%- for item in vg_device_exists.results | ovirt.ovirt.json_query('[?item.key==`' ~ vgname ~ '`]') -%}
        {%- set confs=[] -%}
        {%- for vgConfig in item.item.value -%}
            {%- if item.stdout_lines is defined and item.stdout_lines[loop.index0] == "1" -%}
              {%- set vgConfigCln=dict(item.item.value | first) -%}
              {{- vgConfigCln.__setitem__('pvname',vgConfig.pvname) -}}
              {{- confs.append(vgConfigCln) -}}
            {%- endif -%}
        {%- endfor -%}
        {{- output.__setitem__(vgname, confs) -}}
      {%- endfor -%}
    {%- endfor -%}
    {{- output | to_json -}}

问题分析与解决建议 #

报错核心原因是任务中使用的volume_groups变量被解析为字符串而非预期的字典列表，导致groupby('vgname')无法访问对象属性。

排查与修复方向： #

1. 变量传递验证：检查角色逻辑是否将gluster_infra_volume_groups正确赋值给volume_groups，若缺失需手动添加：

   # 在Playbook的vars或任务中添加
   volume_groups: "{{ gluster_infra_volume_groups }}"
   

2. 变量冲突排查：确认是否存在同名volume_groups变量被定义为字符串，覆盖了预期的字典列表。可执行以下命令查看变量实际值：

   ansible -m debug -a "var=volume_groups" node210.com
   

3. 版本兼容性调整：当前Ansible 2.14.2与gluster-ansible-infra角色可能存在兼容性问题，尝试降级Ansible到2.12.x或2.13.x版本。
4. 设备路径校验：确认/dev/sda为正确的物理卷设备，避免因设备不存在或路径错误导致变量解析异常。

内容的提问来源于stack exchange，提问作者charnet1019