k8s-rollouts

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Progressive Delivery with Argo Rollouts & Flagger

借助Argo Rollouts与Flagger实现渐进式交付

Manage progressive deployments using kubectl-mcp-server's rollout tools (11 tools).

使用kubectl-mcp-server的发布工具（共11个）管理渐进式部署。

When to Apply

适用场景

Use this skill when:

User mentions: "canary", "blue-green", "progressive delivery", "Argo Rollouts", "Flagger"
Operations: rolling out new versions, traffic splitting, automated rollbacks
Keywords: "gradual rollout", "traffic shift", "analysis run", "promote", "abort"

在以下场景中使用本技能：

用户提及："canary"（金丝雀）、"blue-green"（蓝绿）、"progressive delivery"（渐进式交付）、"Argo Rollouts"、"Flagger"
操作场景：新版本发布、流量拆分、自动回滚
关键词："gradual rollout"（逐步发布）、"traffic shift"（流量切换）、"analysis run"（分析运行）、"promote"（升级）、"abort"（终止）

Priority Rules

优先级规则

Priority	Rule	Impact	Tools
1	Detect Argo Rollouts installation first	CRITICAL	`rollouts_detect_tool`
2	Check rollout status before promoting	HIGH	`rollout_status_tool`
3	Monitor analysis runs for failures	HIGH	`analysis_runs_list_tool`
4	Abort immediately on critical failures	CRITICAL	`rollout_abort_tool`

优先级	规则	影响级别	工具
1	先检测Argo Rollouts的安装情况	关键	`rollouts_detect_tool`
2	升级前检查发布状态	高	`rollout_status_tool`
3	监控分析运行是否出现故障	高	`analysis_runs_list_tool`
4	出现严重故障时立即终止	关键	`rollout_abort_tool`

Quick Reference

快速参考

Task	Tool	Example
Detect Argo Rollouts	`rollouts_detect_tool`	`rollouts_detect_tool()`
List rollouts	`rollouts_list_tool`	`rollouts_list_tool(namespace)`
Get rollout status	`rollout_status_tool`	`rollout_status_tool(name, namespace)`
Promote rollout	`rollout_promote_tool`	`rollout_promote_tool(name, namespace)`

任务	工具	示例
检测Argo Rollouts	`rollouts_detect_tool`	`rollouts_detect_tool()`
列出所有发布	`rollouts_list_tool`	`rollouts_list_tool(namespace)`
获取发布状态	`rollout_status_tool`	`rollout_status_tool(name, namespace)`
升级发布	`rollout_promote_tool`	`rollout_promote_tool(name, namespace)`

Check Installation

检查安装

python

rollouts_detect_tool()

python

rollouts_detect_tool()

Argo Rollouts

Argo Rollouts相关操作

List Rollouts

列出所有发布

python

rollouts_list_tool(namespace="default")

python

rollouts_list_tool(namespace="default")

Shows:

显示内容：

- Rollout name

- 发布名称

- Strategy (canary/blueGreen)

- 策略（金丝雀/蓝绿）

- Status

- 状态

- Desired/Ready replicas

- 期望/就绪副本数

undefined

undefined

Get Rollout Details

获取发布详情

python

rollout_get_tool(name="my-rollout", namespace="default")

python

rollout_get_tool(name="my-rollout", namespace="default")

Shows:

显示内容：

- Spec (strategy, steps)

- 规格（策略、步骤）

- Status (phase, conditions)

- 状态（阶段、条件）

- Current step

- 当前步骤

undefined

undefined

Check Rollout Status

检查发布状态

python

rollout_status_tool(name="my-rollout", namespace="default")

python

rollout_status_tool(name="my-rollout", namespace="default")

Returns detailed status with:

返回详细状态，包括：

- Current step index

- 当前步骤索引

- Canary weight

- 金丝雀流量权重

- Stable/canary replicasets

- 稳定版/金丝雀版ReplicaSet

undefined

undefined

Promote Rollout

升级发布

python

undefined

python

undefined

Promote to next step

升级到下一步

rollout_promote_tool(name="my-rollout", namespace="default")

Full promote (skip remaining steps)

完全升级（跳过剩余步骤）

rollout_promote_tool(name="my-rollout", namespace="default", full=True)

undefined

rollout_promote_tool(name="my-rollout", namespace="default", full=True)

undefined

Abort Rollout

终止发布

python

rollout_abort_tool(name="my-rollout", namespace="default")

python

rollout_abort_tool(name="my-rollout", namespace="default")

Reverts to stable version

回滚到稳定版本

undefined

undefined

Retry Rollout

重试发布

python

rollout_retry_tool(name="my-rollout", namespace="default")

python

rollout_retry_tool(name="my-rollout", namespace="default")

Retry failed rollout

重试失败的发布

undefined

undefined

Restart Rollout

重启发布

python

rollout_restart_tool(name="my-rollout", namespace="default")

python

rollout_restart_tool(name="my-rollout", namespace="default")

Triggers new rollout with same spec

使用相同规格触发新的发布

undefined

undefined

Analysis Runs

分析运行

python

undefined

python

undefined

List analysis runs

列出分析运行记录

analysis_runs_list_tool(namespace="default")

Analysis runs verify rollout health:

分析运行用于验证发布健康状态：

- Prometheus metrics

- Prometheus指标

- Web hooks

- Web钩子

- Custom jobs

- 自定义任务

undefined

undefined

Create Canary Rollout

创建金丝雀发布

python

kubectl_apply(manifest="""
apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: my-rollout
  namespace: default
spec:
  replicas: 5
  strategy:
    canary:
      steps:
      - setWeight: 20
      - pause: {duration: 1m}
      - setWeight: 40
      - pause: {duration: 1m}
      - setWeight: 60
      - pause: {duration: 1m}
      - setWeight: 80
      - pause: {duration: 1m}
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
      - name: app
        image: my-app:v2
        ports:
        - containerPort: 8080
""")

python

kubectl_apply(manifest="""
apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: my-rollout
  namespace: default
spec:
  replicas: 5
  strategy:
    canary:
      steps:
      - setWeight: 20
      - pause: {duration: 1m}
      - setWeight: 40
      - pause: {duration: 1m}
      - setWeight: 60
      - pause: {duration: 1m}
      - setWeight: 80
      - pause: {duration: 1m}
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
      - name: app
        image: my-app:v2
        ports:
        - containerPort: 8080
""")

Create Blue-Green Rollout

创建蓝绿发布

python

kubectl_apply(manifest="""
apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: my-rollout
  namespace: default
spec:
  replicas: 3
  strategy:
    blueGreen:
      activeService: my-app-active
      previewService: my-app-preview
      autoPromotionEnabled: false
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
      - name: app
        image: my-app:v2
""")

python

kubectl_apply(manifest="""
apiVersion: argoproj.io/v1alpha1
kind: Rollout
metadata:
  name: my-rollout
  namespace: default
spec:
  replicas: 3
  strategy:
    blueGreen:
      activeService: my-app-active
      previewService: my-app-preview
      autoPromotionEnabled: false
  selector:
    matchLabels:
      app: my-app
  template:
    metadata:
      labels:
        app: my-app
    spec:
      containers:
      - name: app
        image: my-app:v2
""")

Flagger

Flagger相关操作

List Canaries

列出所有金丝雀实例

python

flagger_canaries_list_tool(namespace="default")

python

flagger_canaries_list_tool(namespace="default")

Shows:

显示内容：

- Canary name

- 金丝雀实例名称

- Status (Initialized, Progressing, Succeeded, Failed)

- 状态（已初始化、进行中、成功、失败）

- Weight

- 流量权重

undefined

undefined

Get Canary Details

获取金丝雀实例详情

python

flagger_canary_get_tool(name="my-canary", namespace="default")

python

flagger_canary_get_tool(name="my-canary", namespace="default")

Create Flagger Canary

创建Flagger金丝雀实例

python

kubectl_apply(manifest="""
apiVersion: flagger.app/v1beta1
kind: Canary
metadata:
  name: my-canary
  namespace: default
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: my-app
  service:
    port: 80
  analysis:
    interval: 30s
    threshold: 5
    maxWeight: 50
    stepWeight: 10
    metrics:
    - name: request-success-rate
      threshold: 99
      interval: 1m
    - name: request-duration
      threshold: 500
      interval: 1m
""")

python

kubectl_apply(manifest="""
apiVersion: flagger.app/v1beta1
kind: Canary
metadata:
  name: my-canary
  namespace: default
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: my-app
  service:
    port: 80
  analysis:
    interval: 30s
    threshold: 5
    maxWeight: 50
    stepWeight: 10
    metrics:
    - name: request-success-rate
      threshold: 99
      interval: 1m
    - name: request-duration
      threshold: 500
      interval: 1m
""")

Progressive Delivery Workflows

渐进式交付工作流

Canary Deployment

金丝雀部署

python

1. rollouts_list_tool(namespace)
2. # Update image in rollout
3. rollout_status_tool(name, namespace)  # Monitor progress
4. rollout_promote_tool(name, namespace)  # Promote when ready
5. # Or: rollout_abort_tool(name, namespace) if issues

python

1. rollouts_list_tool(namespace)
2. # 在发布配置中更新镜像
3. rollout_status_tool(name, namespace)  # 监控进度
4. rollout_promote_tool(name, namespace)  # 就绪后进行升级
5. # 或者：若出现问题，执行rollout_abort_tool(name, namespace)

Blue-Green Deployment

蓝绿部署

python

1. rollout_get_tool(name, namespace)  # Check current state
2. # Update image
3. rollout_status_tool(name, namespace)  # Wait for preview ready
4. # Test preview service
5. rollout_promote_tool(name, namespace)  # Switch traffic

python

1. rollout_get_tool(name, namespace)  # 检查当前状态
2. # 更新镜像
3. rollout_status_tool(name, namespace)  # 等待预览环境就绪
4. # 测试预览服务
5. rollout_promote_tool(name, namespace)  # 切换流量

Troubleshooting

故障排查

Rollout Stuck

发布停滞

python

1. rollout_status_tool(name, namespace)  # Check current step
2. analysis_runs_list_tool(namespace)  # Check analysis
3. get_events(namespace)  # Check events
4. # If analysis failing:
   rollout_abort_tool(name, namespace)

python

1. rollout_status_tool(name, namespace)  # 检查当前步骤
2. analysis_runs_list_tool(namespace)  # 检查分析运行记录
3. get_events(namespace)  # 检查事件
4. # 若分析运行失败：
   rollout_abort_tool(name, namespace)

Canary Failing Analysis

金丝雀分析运行失败

python

1. analysis_runs_list_tool(namespace)
2. # Check metrics source (Prometheus, etc.)
3. # Verify threshold configuration
4. rollout_retry_tool(name, namespace)  # Retry if transient

python

1. analysis_runs_list_tool(namespace)
2. # 检查指标来源（如Prometheus等）
3. # 验证阈值配置
4. rollout_retry_tool(name, namespace)  # 若为临时故障则重试