本文介绍 Zadig 服务相关的操作。Zadig 的服务可以是一组 Kubernetes 资源集合（下面简称 K8s Yaml）、一个完整的 Helm Chart 或者是传统的云主机服务。下面针对分别介绍这三种类型的服务操作。

K8s Yaml 服务

新增服务

项目中选择点击服务部分，进入服务管理页面。

创建服务

K8s Yaml 服务的创建支持平台管理和代码仓库托管两种方式：

  • 平台管理：服务的 Kubenertes Yaml 配置存储到 Zadig 系统中
  • 仓库托管：将服务用到的 Yaml（包括服务配置使用的 Configmap）托管到代码仓库中，通过 Zadig 平台配置好文件的路径目录后，Zadig 从代码仓库同步，支持两种同步方式：
    • 手动同步：点击加载按钮，获取仓库中最新的配置
    • 自动同步：通过配置 Webhook（参阅 配置 GitLab Webhook），Zadig 监听分支上有代码改动，对应的服务都会自动同步。

平台管理

  • 点击新建按钮新建服务。

创建服务

  • 输入新的服务名称。

创建服务

  • 将服务 YAML 填入编辑器并保存。

创建服务

  • 更新环境，该服务会自动加入到选择的环境中。

更新环境

仓库托管

  • 选择仓库托管，选择代码仓库
  • 选择服务配置所在文件目录，加载服务

更新服务

平台管理

  • 修改服务 Yaml并保存。

修改服务

  • 选择相应环境进行更新。

更新环境中的服务

仓库托管

  • 提交服务配置变更到代码仓库

配置变更

  • 变更合并到主干分支后，通过 Webhook 的能力自动同步最新配置到 Zadig 系统。也可以在界面上手动同步服务配置，如下图所示。

服务手动配置同步

  • 在 Zadig 集成环境中，查看服务配置的变更，点击服务更新按钮执行更新操作

服务版本diff 服务更新

删除服务

  • 从服务 模块中将服务配置删除。
  • 更新环境，将删除的服务从相应环境中移除。

删除环境

服务 Yaml 样例

无状态服务

概念：服务运行的实例不会在本地存储需要持久化的数据，并且多个实例对于同一个请求响应的结果是完全一致的。可以参考这篇文章服务 - 图13 (opens new window)了解无状态服务的更多细节。

点击查看

  1. apiVersion: apps/v1
  2. kind: Deployment
  3. metadata:
  4.   name: nginx-deployment
  5. spec:
  6.   selector:
  7.     matchLabels:
  8.       app: nginx
  9.   replicas: 2 # 2 个 Pod 实例
  10.   template:
  11.     metadata:
  12.       labels:
  13.         app: nginx
  14.     spec:
  15.       containers:
  16.       - name: nginx
  17.         image: nginx:1.14.2
  18.         ports:
  19.         - containerPort: 80

有状态服务

概念：服务的实例可以将一部分数据随时进行备份，并且在创建一个新的有状态服务时，可以通过备份恢复这些数据，以达到数据持久化的目的。可以参考这篇文章服务 - 图14 (opens new window)了解有状态服务的更多细节。

点击查看

  1. apiVersion: v1
  2. kind: ConfigMap
  3. metadata:
  4.   name: mysql
  5.   labels:
  6.     app: mysql
  7. data:
  8.   master.cnf: |
  9.     # Apply this config only on the master.
  10.     [mysqld]
  11.     log-bin
  12.   slave.cnf: |
  13.     # Apply this config only on slaves.
  14.     [mysqld]
  15.     super-read-only
  16. ---
  17. # Headless service for stable DNS entries of StatefulSet members.
  18. apiVersion: v1
  19. kind: Service
  20. metadata:
  21.   name: mysql
  22.   labels:
  23.     app: mysql
  24. spec:
  25.   ports:
  26.   - name: mysql
  27.     port: 3306
  28.   clusterIP: None
  29.   selector:
  30.     app: mysql
  31. ---
  32. # Client service for connecting to any MySQL instance for reads.
  33. # For writes, you must instead connect to the master: mysql-0.mysql.
  34. apiVersion: v1
  35. kind: Service
  36. metadata:
  37.   name: mysql-read
  38.   labels:
  39.     app: mysql
  40. spec:
  41.   ports:
  42.   - name: mysql
  43.     port: 3306
  44.   selector:
  45.     app: mysql
  46. ---
  47. apiVersion: apps/v1beta1
  48. kind: StatefulSet
  49. metadata:
  50.   name: mysql
  51. spec:
  52.   selector:
  53.     matchLabels:
  54.       app: mysql
  55.   serviceName: mysql
  56.   # 1 master and 2 slave
  57.   replicas: 3
  58.   template:
  59.     metadata:
  60.       labels:
  61.         app: mysql
  62.     spec:
  63.       initContainers:
  64.       - name: init-mysql
  65.         image: mysql:5.7
  66.         command:
  67.         - bash
  68.         - "-c"
  69.         - |
  70.           set -ex
  71.           # Generate mysql server-id from pod ordinal index.
  72.           [[ `hostname` =~ -([0-9]+)$ ]] || exit 1
  73.           ordinal=${BASH_REMATCH[1]}
  74.           echo [mysqld] > /mnt/conf.d/server-id.cnf
  75.           # Add an offset to avoid reserved server-id=0 value.
  76.           echo server-id=$((100 + $ordinal)) >> /mnt/conf.d/server-id.cnf
  77.           # Copy appropriate conf.d files from config-map to emptyDir.
  78.           if [[ $ordinal -eq 0 ]]; then
  79.             cp /mnt/config-map/master.cnf /mnt/conf.d/
  80.           else
  81.             cp /mnt/config-map/slave.cnf /mnt/conf.d/
  82.           fi
  83.         volumeMounts:
  84.         - name: conf
  85.           mountPath: /mnt/conf.d
  86.         - name: config-map
  87.           mountPath: /mnt/config-map
  88.       - name: clone-mysql
  89.         image: gcr.azk8s.cn/google-samples/xtrabackup:1.0
  90.         command:
  91.         - bash
  92.         - "-c"
  93.         - |
  94.           set -ex
  95.           # Skip the clone if data already exists.
  96.           [[ -d /var/lib/mysql/mysql ]] && exit 0
  97.           # Skip the clone on master (ordinal index 0).
  98.           [[ `hostname` =~ -([0-9]+)$ ]] || exit 1
  99.           ordinal=${BASH_REMATCH[1]}
  100.           [[ $ordinal -eq 0 ]] && exit 0
  101.           # Clone data from previous peer.
  102.           ncat --recv-only mysql-$(($ordinal-1)).mysql 3307 | xbstream -x -C /var/lib/mysql
  103.           # Prepare the backup.
  104.           xtrabackup --prepare --target-dir=/var/lib/mysql
  105.         volumeMounts:
  106.         - name: data
  107.           mountPath: /var/lib/mysql
  108.           subPath: mysql
  109.         - name: conf
  110.           mountPath: /etc/mysql/conf.d
  111.       containers:
  112.       - name: mysql
  113.         image: mysql:5.7
  114.         env:
  115.         - name: MYSQL_ALLOW_EMPTY_PASSWORD
  116.           value: "1"
  117.         ports:
  118.         - name: mysql
  119.           containerPort: 3306
  120.         volumeMounts:
  121.         - name: data
  122.           mountPath: /var/lib/mysql
  123.           subPath: mysql
  124.         - name: conf
  125.           mountPath: /etc/mysql/conf.d
  126.         resources:
  127.           requests:
  128.             cpu: 500m
  129.             memory: 1Gi
  130.           limits:
  131.             cpu: 500m
  132.             memory: 1Gi
  133.         livenessProbe:
  134.           exec:
  135.             command: ["mysqladmin", "ping"]
  136.           initialDelaySeconds: 30
  137.           periodSeconds: 10
  138.           timeoutSeconds: 5
  139.         readinessProbe:
  140.           exec:
  141.             # Check we can execute queries over TCP (skip-networking is off).
  142.             command: ["mysql", "-h", "127.0.0.1", "-e", "SELECT 1"]
  143.           initialDelaySeconds: 5
  144.           periodSeconds: 2
  145.           timeoutSeconds: 1
  146.       - name: xtrabackup
  147.         image: gcr.azk8s.cn/google-samples/xtrabackup:1.0
  148.         ports:
  149.         - name: xtrabackup
  150.           containerPort: 3307
  151.         command:
  152.         - bash
  153.         - "-c"
  154.         - |
  155.           set -ex
  156.           cd /var/lib/mysql
  158.           # Determine binlog position of cloned data, if any.
  159.           if [[ -f xtrabackup_slave_info && "x$(<xtrabackup_slave_info)" != "x" ]]; then
  160.             # XtraBackup already generated a partial "CHANGE MASTER TO" query
  161.             # because we're cloning from an existing slave. (Need to remove the tailing semicolon!)
  162.             cat xtrabackup_slave_info | sed -E 's/;$//g' > change_master_to.sql.in
  163.             # Ignore xtrabackup_binlog_info in this case (it's useless).
  164.             rm -f xtrabackup_slave_info xtrabackup_binlog_info
  165.           elif [[ -f xtrabackup_binlog_info ]]; then
  166.             # We're cloning directly from master. Parse binlog position.
  167.             [[ `cat xtrabackup_binlog_info` =~ ^(.*?)[[:space:]]+(.*?)$ ]] || exit 1
  168.             rm -f xtrabackup_binlog_info xtrabackup_slave_info
  169.             echo "CHANGE MASTER TO MASTER_LOG_FILE='${BASH_REMATCH[1]}',\
  170.                   MASTER_LOG_POS=${BASH_REMATCH[2]}" > change_master_to.sql.in
  171.           fi
  173.           # Check if we need to complete a clone by starting replication.
  174.           if [[ -f change_master_to.sql.in ]]; then
  175.             echo "Waiting for mysqld to be ready (accepting connections)"
  176.             until mysql -h 127.0.0.1 -e "SELECT 1"; do sleep 1; done
  178.             echo "Initializing replication from clone position"
  179.             mysql -h 127.0.0.1 \
  180.                   -e "$(<change_master_to.sql.in), \
  181.                           MASTER_HOST='mysql-0.mysql', \
  182.                           MASTER_USER='root', \
  183.                           MASTER_PASSWORD='', \
  184.                           MASTER_CONNECT_RETRY=10; \
  185.                         START SLAVE;" || exit 1
  186.             # In case of container restart, attempt this at-most-once.
  187.             mv change_master_to.sql.in change_master_to.sql.orig
  188.           fi
  190.           # Start a server to send backups when requested by peers.
  191.           exec ncat --listen --keep-open --send-only --max-conns=1 3307 -c \
  192.             "xtrabackup --backup --slave-info --stream=xbstream --host=127.0.0.1 --user=root"
  193.         volumeMounts:
  194.         - name: data
  195.           mountPath: /var/lib/mysql
  196.           subPath: mysql
  197.         - name: conf
  198.           mountPath: /etc/mysql/conf.d
  199.         resources:
  200.           requests:
  201.             cpu: 100m
  202.             memory: 100Mi
  203.       volumes:
  204.       - name: conf
  205.         emptyDir: {}
  206.       - name: config-map
  207.         configMap:
  208.           name: mysql
  209.   volumeClaimTemplates:
  210.   - metadata:
  211.       name: data
  212.     spec:
  213.       accessModes: ["ReadWriteOnce"]
  214.       resources:
  215.         requests:
  216.           storage: 10Gi

Helm Chart 服务

文档建设中，敬请期待……

传统云主机服务

文档建设中，敬请期待……