feat: 迁移基础设施到Nomad和Podman并重构配置

refactor: 更新Ansible Playbooks以支持Nomad集群
docs: 更新文档反映从Docker Swarm到Nomad的迁移
ci: 更新Gitea工作流以支持Podman构建
test: 添加Nomad作业测试文件
build: 更新Makefile以支持Podman操作
chore: 清理旧的Docker Swarm相关文件和配置

playbooks/add-beijing-node-prefix.yml

@@ -0,0 +1,69 @@
+---
+- name: Add Beijing prefix to LXC node names in Nomad configuration
+  hosts: beijing
+  become: yes
+
+  vars:
+    node_prefixes:
+      influxdb: "bj-influxdb"
+      warden: "bj-warden"
+      hcp1: "bj-hcp1"
+      hcp2: "bj-hcp2"
+    tailscale_ips:
+      influxdb: "100.100.7.4"
+      warden: "100.122.197.112"
+      hcp1: "100.97.62.111"
+      hcp2: "100.116.112.45"
+
+  tasks:
+    - name: Stop Nomad service
+      systemd:
+        name: nomad
+        state: stopped
+
+    - name: Get current node name from inventory
+      set_fact:
+        current_node_name: "{{ inventory_hostname }}"
+        new_node_name: "{{ node_prefixes[inventory_hostname] }}"
+        tailscale_ip: "{{ tailscale_ips[inventory_hostname] }}"
+
+    - name: Display node name change
+      debug:
+        msg: "Changing node name from {{ current_node_name }} to {{ new_node_name }}, using Tailscale IP {{ tailscale_ip }}"
+
+    - name: Update node name in Nomad configuration
+      lineinfile:
+        path: /etc/nomad.d/nomad.hcl
+        regexp: '^name\s*='
+        line: 'name = "{{ new_node_name }}"'
+        insertafter: 'datacenter = "dc1"'
+        state: present
+
+    - name: Validate Nomad configuration
+      shell: nomad config validate /etc/nomad.d/nomad.hcl
+      register: config_validation
+      failed_when: config_validation.rc != 0
+
+    - name: Start Nomad service
+      systemd:
+        name: nomad
+        state: started
+
+    - name: Wait for Nomad to be ready on Tailscale IP
+      wait_for:
+        port: 4646
+        host: "{{ tailscale_ip }}"
+        delay: 10
+        timeout: 60
+
+    - name: Wait for node registration
+      pause:
+        seconds: 15
+
+    - name: Display new configuration
+      shell: cat /etc/nomad.d/nomad.hcl | grep -E "^(datacenter|name|bind_addr)\s*="
+      register: nomad_config_check
+
+    - name: Show updated configuration
+      debug:
+        var: nomad_config_check.stdout_lines

playbooks/fix-duplicate-plugin-dir.yml

@@ -0,0 +1,56 @@
+---
+- name: Fix duplicate plugin_dir configuration
+  hosts: nomadlxc,hcp
+  become: yes
+
+  tasks:
+    - name: Stop Nomad service
+      systemd:
+        name: nomad
+        state: stopped
+
+    - name: Remove duplicate plugin_dir lines
+      lineinfile:
+        path: /etc/nomad.d/nomad.hcl
+        regexp: '^plugin_dir = "/opt/nomad/plugins"'
+        state: absent
+
+    - name: Ensure only one plugin_dir configuration exists
+      lineinfile:
+        path: /etc/nomad.d/nomad.hcl
+        regexp: '^plugin_dir = "/opt/nomad/data/plugins"'
+        line: 'plugin_dir = "/opt/nomad/data/plugins"'
+        insertafter: 'data_dir = "/opt/nomad/data"'
+        state: present
+
+    - name: Validate Nomad configuration
+      shell: nomad config validate /etc/nomad.d/nomad.hcl
+      register: config_validation
+      failed_when: config_validation.rc != 0
+
+    - name: Start Nomad service
+      systemd:
+        name: nomad
+        state: started
+
+    - name: Wait for Nomad to be ready
+      wait_for:
+        port: 4646
+        host: localhost
+        delay: 10
+        timeout: 60
+
+    - name: Wait for plugins to load
+      pause:
+        seconds: 15
+
+    - name: Check driver status
+      shell: |
+        export NOMAD_ADDR=http://localhost:4646
+        nomad node status -self | grep -A 10 "Driver Status"
+      register: driver_status
+      failed_when: false
+
+    - name: Display driver status
+      debug:
+        var: driver_status.stdout_lines

playbooks/fix-podman-driver-config.yml

@@ -0,0 +1,112 @@
+---
+- name: Fix Nomad Podman Driver Configuration
+  hosts: nomadlxc,hcp
+  become: yes
+  vars:
+    nomad_user: nomad
+
+  tasks:
+    - name: Stop Nomad service
+      systemd:
+        name: nomad
+        state: stopped
+
+    - name: Install Podman driver plugin if missing
+      block:
+        - name: Check if plugin exists
+          stat:
+            path: /opt/nomad/data/plugins/nomad-driver-podman
+          register: plugin_exists
+
+        - name: Download and install Podman driver plugin
+          block:
+            - name: Download Nomad Podman driver
+              get_url:
+                url: "https://releases.hashicorp.com/nomad-driver-podman/0.6.1/nomad-driver-podman_0.6.1_linux_amd64.zip"
+                dest: "/tmp/nomad-driver-podman.zip"
+                mode: '0644'
+
+            - name: Extract Podman driver
+              unarchive:
+                src: "/tmp/nomad-driver-podman.zip"
+                dest: "/tmp"
+                remote_src: yes
+
+            - name: Install Podman driver
+              copy:
+                src: "/tmp/nomad-driver-podman"
+                dest: "/opt/nomad/data/plugins/nomad-driver-podman"
+                owner: "{{ nomad_user }}"
+                group: "{{ nomad_user }}"
+                mode: '0755'
+                remote_src: yes
+
+            - name: Clean up temporary files
+              file:
+                path: "{{ item }}"
+                state: absent
+              loop:
+                - "/tmp/nomad-driver-podman.zip"
+                - "/tmp/nomad-driver-podman"
+          when: not plugin_exists.stat.exists
+
+    - name: Update Nomad configuration with correct plugin name and socket path
+      replace:
+        path: /etc/nomad.d/nomad.hcl
+        regexp: 'plugin "podman" \{'
+        replace: 'plugin "nomad-driver-podman" {'
+
+    - name: Update socket path to system socket
+      replace:
+        path: /etc/nomad.d/nomad.hcl
+        regexp: 'socket_path = "unix:///run/user/1001/podman/podman.sock"'
+        replace: 'socket_path = "unix:///run/podman/podman.sock"'
+
+    - name: Add plugin_dir configuration if missing
+      lineinfile:
+        path: /etc/nomad.d/nomad.hcl
+        line: 'plugin_dir = "/opt/nomad/data/plugins"'
+        insertafter: 'data_dir = "/opt/nomad/data"'
+        state: present
+
+    - name: Ensure Podman socket is enabled and running
+      systemd:
+        name: podman.socket
+        enabled: yes
+        state: started
+
+    - name: Start Nomad service
+      systemd:
+        name: nomad
+        state: started
+
+    - name: Wait for Nomad to be ready
+      wait_for:
+        port: 4646
+        host: localhost
+        delay: 10
+        timeout: 60
+
+    - name: Wait for plugins to load
+      pause:
+        seconds: 20
+
+    - name: Check driver status
+      shell: |
+        export NOMAD_ADDR=http://localhost:4646
+        nomad node status -self | grep -A 10 "Driver Status"
+      register: driver_status
+      failed_when: false
+
+    - name: Display driver status
+      debug:
+        var: driver_status.stdout_lines
+
+    - name: Check for Podman driver in logs
+      shell: journalctl -u nomad -n 30 --no-pager | grep -E "(podman|plugin)" | tail -10
+      register: plugin_logs
+      failed_when: false
+
+    - name: Display plugin logs
+      debug:
+        var: plugin_logs.stdout_lines

playbooks/fix-warden-nfs.yml

@@ -0,0 +1,46 @@
+---
+- name: Fix NFS mounting on warden node
+  hosts: warden
+  become: yes
+  tasks:
+    - name: Ensure rpcbind is running
+      systemd:
+        name: rpcbind
+        state: started
+        enabled: yes
+
+    - name: Ensure nfs-client.target is active
+      systemd:
+        name: nfs-client.target
+        state: started
+        enabled: yes
+
+    - name: Create consul-shared directory
+      file:
+        path: /opt/consul-shared
+        state: directory
+        mode: '0755'
+
+    - name: Mount NFS share
+      mount:
+        path: /opt/consul-shared
+        src: snail:/fs/1000/nfs
+        fstype: nfs
+        opts: rw,sync,vers=3
+        state: mounted
+
+    - name: Add to fstab for persistence
+      mount:
+        path: /opt/consul-shared
+        src: snail:/fs/1000/nfs
+        fstype: nfs
+        opts: rw,sync,vers=3
+        state: present
+
+    - name: Verify mount
+      command: df -h /opt/consul-shared
+      register: mount_result
+
+    - name: Display mount result
+      debug:
+        var: mount_result.stdout

playbooks/setup-nfs-storage.yml

@@ -0,0 +1,75 @@
+---
+- name: Setup NFS Storage for Consul Cluster
+  hosts: localhost
+  gather_facts: false
+  vars:
+    nfs_server: snail
+    nfs_export_path: /fs/1000/nfs
+    nfs_mount_path: /opt/consul-shared
+  
+  tasks:
+    - name: Install NFS client and mount on master
+      ansible.builtin.shell: |
+        ssh -o StrictHostKeyChecking=no -p 60022 ben@master '
+          echo "3131" | sudo -S apt update && 
+          echo "3131" | sudo -S apt install -y nfs-common &&
+          echo "3131" | sudo -S mkdir -p {{ nfs_mount_path }} &&
+          echo "3131" | sudo -S mount -t nfs {{ nfs_server }}:{{ nfs_export_path }} {{ nfs_mount_path }} &&
+          echo "{{ nfs_server }}:{{ nfs_export_path }} {{ nfs_mount_path }} nfs defaults 0 0" | echo "3131" | sudo -S tee -a /etc/fstab
+        '
+      delegate_to: localhost
+      register: master_result
+
+    - name: Install NFS client and mount on ash3c
+      ansible.builtin.shell: |
+        ssh -o StrictHostKeyChecking=no ben@ash3c '
+          echo "3131" | sudo -S apt update && 
+          echo "3131" | sudo -S apt install -y nfs-common &&
+          echo "3131" | sudo -S mkdir -p {{ nfs_mount_path }} &&
+          echo "3131" | sudo -S mount -t nfs {{ nfs_server }}:{{ nfs_export_path }} {{ nfs_mount_path }} &&
+          echo "{{ nfs_server }}:{{ nfs_export_path }} {{ nfs_mount_path }} nfs defaults 0 0" | echo "3131" | sudo -S tee -a /etc/fstab
+        '
+      delegate_to: localhost
+      register: ash3c_result
+
+    - name: Install NFS client and mount on warden
+      ansible.builtin.shell: |
+        ssh -o StrictHostKeyChecking=no ben@warden '
+          echo "3131" | sudo -S apt update && 
+          echo "3131" | sudo -S apt install -y nfs-common &&
+          echo "3131" | sudo -S mkdir -p {{ nfs_mount_path }} &&
+          echo "3131" | sudo -S mount -t nfs {{ nfs_server }}:{{ nfs_export_path }} {{ nfs_mount_path }} &&
+          echo "{{ nfs_server }}:{{ nfs_export_path }} {{ nfs_mount_path }} nfs defaults 0 0" | echo "3131" | sudo -S tee -a /etc/fstab
+        '
+      delegate_to: localhost
+      register: warden_result
+
+    - name: Test NFS connectivity on all nodes
+      ansible.builtin.shell: |
+        ssh -o StrictHostKeyChecking=no -p 60022 ben@master 'echo "3131" | sudo -S touch {{ nfs_mount_path }}/test-master-$(date +%s) && ls -la {{ nfs_mount_path }}/'
+        ssh -o StrictHostKeyChecking=no ben@ash3c 'echo "3131" | sudo -S touch {{ nfs_mount_path }}/test-ash3c-$(date +%s) && ls -la {{ nfs_mount_path }}/'
+        ssh -o StrictHostKeyChecking=no ben@warden 'echo "3131" | sudo -S touch {{ nfs_mount_path }}/test-warden-$(date +%s) && ls -la {{ nfs_mount_path }}/'
+      delegate_to: localhost
+      register: nfs_test_result
+
+    - name: Display NFS test results
+      ansible.builtin.debug:
+        var: nfs_test_result.stdout_lines
+
+    - name: Create Consul data directories on NFS
+      ansible.builtin.shell: |
+        ssh -o StrictHostKeyChecking=no -p 60022 ben@master 'echo "3131" | sudo -S mkdir -p {{ nfs_mount_path }}/consul-master'
+        ssh -o StrictHostKeyChecking=no ben@ash3c 'echo "3131" | sudo -S mkdir -p {{ nfs_mount_path }}/consul-ash3c'
+        ssh -o StrictHostKeyChecking=no ben@warden 'echo "3131" | sudo -S mkdir -p {{ nfs_mount_path }}/consul-warden'
+      delegate_to: localhost
+      register: consul_dirs_result
+
+    - name: Display setup completion
+      ansible.builtin.debug:
+        msg: 
+          - "NFS setup completed successfully!"
+          - "NFS mount point: {{ nfs_mount_path }}"
+          - "Consul data directories created:"
+          - "  - {{ nfs_mount_path }}/consul-master"
+          - "  - {{ nfs_mount_path }}/consul-ash3c"
+          - "  - {{ nfs_mount_path }}/consul-warden"