Use Logtail to collect application logs from Windows nodes - Container Service for Kubernetes: Central de Documentação do Alibaba Cloud

O Logtail coleta logs de aplicações de containers em execução em nós Windows e os envia ao Simple Log Service (SLS). Este tópico explica como implantar o Logtail como DaemonSet em nós Windows e verifique se a coleta de logs funciona.

Pré-requisitos

Antes de começar, verifique se você tem:

Um cluster ACK criado com a opção Enable Log Service selecionada. Consulte Criar um cluster gerenciado ACK.
Um pool de nós Windows. Consulte Criar um pool de nós Windows.
Um cliente kubectl conectado ao cluster ACK. Consulte Etapa 2: Selecione um tipo de credencial de cluster.

Limitações

O Logtail coleta apenas logs de aplicações impressos no stdout. Se sua aplicação gravar logs em arquivos ou usar serviços de log do Windows, utilize o LogMonitor (guia de uso) para encaminhar esses logs ao stdout antes da coleta pelo Logtail. Para obter uma visão geral conceitual, consulte Introdução ao LogMonitor.

Escolha uma versão do Logtail

Selecione a versão correspondente ao seu cluster:

Versão do Logtail	Cluster aplicável	Seção
V1.1.x (recomendado)	Nós Windows com o rótulo `windows.alibabacloud.com/deployment-topology=2.0`	Instale o Logtail V1.1.x
V1.0.x	Clusters ACK executando Kubernetes 1.18.x e versões anteriores	Instale o Logtail V1.0.x

Instale o Logtail V1.1.x em nós Windows

Importante

O Logtail V1.1.x depende de processos proxy privilegiados e só pode ser executado em nós Windows rotulados com windows.alibabacloud.com/deployment-topology=2.0. Para verifique se seus nós possuem esse rótulo, execute kubectl get nodes --show-labels e procure essa chave.

Aplique o manifesto DaemonSet a seguir para implantar o Logtail em nós Windows:

apiVersion: apps/v1
kind: DaemonSet
metadata:
  labels:
    k8s-app: logtail-ds-windows
  name: logtail-ds-windows
  namespace: kube-system
spec:
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: logtail-ds-windows
  template:
    metadata:
      annotations:
        scheduler.alpha.kubernetes.io/critical-pod: ""
      labels:
        app: logtail-ds-windows
    spec:
      hostNetwork: true
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
            - matchExpressions:
              - key: type
                operator: NotIn
                values:
                - virtual-kubelet
              - key: beta.kubernetes.io/os
                operator: In
                values:
                - windows
              - key: windows.alibabacloud.com/deployment-topology
                operator: In
                values:
                - "2.0"
            - matchExpressions:
              - key: type
                operator: NotIn
                values:
                - virtual-kubelet
              - key: kubernetes.io/os
                operator: In
                values:
                - windows
              - key: windows.alibabacloud.com/deployment-topology
                operator: In
                values:
                - "2.0"
      containers:
      - name: logtail
        # use the below volume configuration since v1.1+.
        command:
        - pwsh.exe
        - -NoLogo
        - -NonInteractive
        - -File
        - entrypoint.ps1
        env:
        - name: POD_NAMESPACE
          valueFrom:
            fieldRef:
              fieldPath: metadata.namespace
        - name: POD_NAME
          valueFrom:
            fieldRef:
              fieldPath: metadata.name
        - name: CONTAINER_NAME
          value: "logtail"
        # --- configure the logtail file configuration generation --- #
        ## "ALICLOUD_LOGTAIL_CONFIG_PATH" specifies the configuration path of logtail,
        ## if the path is blank, the configuration will generate
        ## via "ALICLOUD_LOGTAIL_CONFIG_ITEM__" prefix environment variables.
        ## "ALICLOUD_LOGTAIL_CONFIG_ITEM__" needs to be combined with a type indicator to
        ## display the type of the configuration item, the following indicators are optional.
        ## -       INT__ : 64-bit signed integer.
        ## -      UINT__ : 64-bit unsigned integer.
        ## -      DOUB__ : 64-bit floating-point number.
        ## -      BOOL__ : boolean.
        ## -       STR__ : string, default type.
        ## - {<TYPE>}S__ : array in <TYPE> with vertical bar separated format.
        ## P.S: don't treate "ALICLOUD_LOGTAIL_CONFIG_ITEM__" as a silver bullet,
        ## when the configuration is too complicated,
        ## please mount a detailed configuration file on "ALICLOUD_LOGTAIL_CONFIG_PATH".
        - name: ALICLOUD_LOGTAIL_CONFIG_PATH
          valueFrom:
            configMapKeyRef:
              key: log-config-path
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__DOUB__CPU_USAGE_LIMIT
          valueFrom:
            configMapKeyRef:
              key: cpu-core-limit
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__UINT__MEM_USAGE_LIMIT
          valueFrom:
            configMapKeyRef:
              key: mem-limit
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__UINT__MAX_BYTES_PER_SEC
          valueFrom:
            configMapKeyRef:
              key: max-bytes-per-sec
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__UINT__SEND_REQUESTS_CONCURRENCY
          valueFrom:
            configMapKeyRef:
              key: send-requests-concurrency
              name: alibaba-log-configuration
        # --- configure the logtail configuration --- #
        ## "ALICLOUD_LOG_REGION" specifies the region of the Alibaba Cloud,
        ## it will discover the region automatically if blank.
        ## "ALICLOUD_LOG_USER_ID" is the same as "ALIYUN_LOGTAIL_USER_ID",
        ## which is to specify the uid of Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_PROJECT" is the same as "ALICLOUD_LOG_DEFAULT_PROJECT",
        ## which is to specify the project of Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_MACHINE_GROUP" is the same as "ALIYUN_LOGTAIL_USER_DEFINED_ID"
        ## and "ALICLOUD_LOG_DEFAULT_MACHINE_GROUP",
        ## which is to specify the machine group of Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_ENDPOINT" specifies the endpoint of the Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_ECS_FLAG" specifies whether to log the ECS flags.
        ## "ALICLOUD_LOG_DOCKER_ENV_CONFIG"
        ## "ALICLOUD_LOG_ENV_TAGS" is the same as "ALIYUN_LOG_ENV_TAGS",
        ## which is to specify the environment variables to be recorded,
        ## it is in form of vertical bar separated list.
        - name: ALICLOUD_LOG_REGION
          value: ""
        - name: ALICLOUD_LOG_USER_ID
          valueFrom:
            configMapKeyRef:
              key: log-ali-uid
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_PROJECT
          valueFrom:
            configMapKeyRef:
              key: log-project
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_MACHINE_GROUP
          valueFrom:
            configMapKeyRef:
              key: log-machine-group
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_ENDPOINT
          valueFrom:
            configMapKeyRef:
              key: log-endpoint
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_ECS_FLAG
          value: "true"
        - name: ALICLOUD_LOG_DOCKER_ENV_CONFIG
          value: "true"
        - name: ALICLOUD_LOG_ENV_TAGS
          value: _node_name_|_node_ip_|working_ip
        - name: _node_name_
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: spec.nodeName
        - name: _node_ip_
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: status.hostIP
        - name: working_ip
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: status.hostIP
        # Replace cn-hangzhou in the following image address with the region where your cluster is deployed.
        image: registry-cn-hangzhou-vpc.ack.aliyuncs.com/acs/logtail-windows:v1.1.30-aliyun.1
        imagePullPolicy: IfNotPresent
        resources:
          limits:
            cpu: 1
            memory: 1Gi
          requests:
            cpu: 100m
            memory: 256Mi
        securityContext:
          privileged: false
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
        volumeMounts:
        - name: addon-token
          mountPath: 'c:/var/addon'
        - name: host-binary
          mountPath: 'c:/host/opt/ilogtail/bin'
        - name: project-wins-pipe
          mountPath: \\.\pipe\rancher_wins
      terminationGracePeriodSeconds: 30
      priorityClassName: system-node-critical
      restartPolicy: Always
      tolerations:
      - operator: Exists
      volumes:
      - name: addon-token
        secret:
          secretName: addon.log.token
          items:
            - key: addon.token.config
              path: token-config
      - name: host-binary
        hostPath:
          path: 'c:/opt/ilogtail/bin'
          type: DirectoryOrCreate
      - name: project-wins-pipe
        hostPath:
          path: \\.\pipe\rancher_wins
  updateStrategy:
    type: RollingUpdate

Instale o Logtail V1.0.x em nós Windows

Importante

Esta versão aplica-se somente a clusters ACK que executam Kubernetes 1.18.x e versões anteriores.

Aplique o manifesto DaemonSet a seguir para implantar o Logtail em nós Windows:

apiVersion: apps/v1
kind: DaemonSet
metadata:
  labels:
    k8s-app: logtail-ds-windows
  name: logtail-ds-windows
  namespace: kube-system
spec:
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: logtail-ds-windows
  template:
    metadata:
      labels:
        app: logtail-ds-windows
    spec:
      affinity:
        nodeAffinity:
          requiredDuringSchedulingIgnoredDuringExecution:
            nodeSelectorTerms:
            - matchExpressions:
              - key: type
                operator: NotIn
                values:
                - virtual-kubelet
              - key: beta.kubernetes.io/os
                operator: In
                values:
                - windows
              - key: windows.alibabacloud.com/deployment-topology
                operator: DoesNotExist
            - matchExpressions:
              - key: type
                operator: NotIn
                values:
                - virtual-kubelet
              - key: kubernetes.io/os
                operator: In
                values:
                - windows
              - key: windows.alibabacloud.com/deployment-topology
                operator: DoesNotExist
      containers:
      - name: logtail
        command:
        - powershell.exe
        - -NoLogo
        - -NonInteractive
        - -File
        - entrypoint.ps1
        env:
        # --- configure the logtail file configuration generation --- #
        ## "ALICLOUD_LOGTAIL_CONFIG_PATH" specifies the configuration path of logtail,
        ## if the path is blank, the configuration will generate
        ## via "ALICLOUD_LOGTAIL_CONFIG_ITEM__" prefix environment variables.
        ## "ALICLOUD_LOGTAIL_CONFIG_ITEM__" needs to be combined with a type indicator to
        ## display the type of the configuration item, the following indicators are optional.
        ## -       INT__ : 64-bit signed integer.
        ## -      UINT__ : 64-bit unsigned integer.
        ## -      DOUB__ : 64-bit floating-point number.
        ## -      BOOL__ : boolean.
        ## -       STR__ : string, default type.
        ## - {<TYPE>}S__ : array in <TYPE> with vertical bar separated format.
        ## P.S: don't treate "ALICLOUD_LOGTAIL_CONFIG_ITEM__" as a silver bullet,
        ## when the configuration is too complicated,
        ## please mount a detailed configuration file on "ALICLOUD_LOGTAIL_CONFIG_PATH".
        - name: ALICLOUD_LOGTAIL_CONFIG_PATH
          valueFrom:
            configMapKeyRef:
              key: log-config-path
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__DOUB__CPU_USAGE_LIMIT
          valueFrom:
            configMapKeyRef:
              key: cpu-core-limit
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__UINT__MEM_USAGE_LIMIT
          valueFrom:
            configMapKeyRef:
              key: mem-limit
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__UINT__MAX_BYTES_PER_SEC
          valueFrom:
            configMapKeyRef:
              key: max-bytes-per-sec
              name: alibaba-log-configuration
        - name: ALICLOUD_LOGTAIL_CONFIG_ITEM__UINT__SEND_REQUESTS_CONCURRENCY
          valueFrom:
            configMapKeyRef:
              key: send-requests-concurrency
              name: alibaba-log-configuration
        # --- configure the logtail configuration --- #
        ## "ALICLOUD_LOG_REGION" specifies the region of the Alibaba Cloud,
        ## it will discover the region automatically if blank.
        ## "ALICLOUD_LOG_USER_ID" is the same as "ALIYUN_LOGTAIL_USER_ID",
        ## which is to specify the uid of Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_PROJECT" is the same as "ALICLOUD_LOG_DEFAULT_PROJECT",
        ## which is to specify the project of Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_MACHINE_GROUP" is the same as "ALIYUN_LOGTAIL_USER_DEFINED_ID"
        ## and "ALICLOUD_LOG_DEFAULT_MACHINE_GROUP",
        ## which is to specify the machine group of Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_ENDPOINT" specifies the endpoint of the Alibaba Cloud SLS service.
        ## "ALICLOUD_LOG_ECS_FLAG" specifies whether to log the ECS flags.
        ## "ALICLOUD_LOG_DOCKER_ENV_CONFIG"
        ## "ALICLOUD_LOG_ENV_TAGS" is the same as "ALIYUN_LOG_ENV_TAGS",
        ## which is to specify the environment variables to be recorded,
        ## it is in form of vertical bar separated list.
        - name: ALICLOUD_LOG_REGION
          value: ""
        - name: ALICLOUD_LOG_USER_ID
          valueFrom:
            configMapKeyRef:
              key: log-ali-uid
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_PROJECT
          valueFrom:
            configMapKeyRef:
              key: log-project
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_MACHINE_GROUP
          valueFrom:
            configMapKeyRef:
              key: log-machine-group
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_ENDPOINT
          valueFrom:
            configMapKeyRef:
              key: log-endpoint
              name: alibaba-log-configuration
        - name: ALICLOUD_LOG_ECS_FLAG
          value: "true"
        - name: ALICLOUD_LOG_DOCKER_ENV_CONFIG
          value: "true"
        - name: ALICLOUD_LOG_ENV_TAGS
          value: _node_name_|_node_ip_|working_ip
        - name: _node_name_
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: spec.nodeName
        - name: working_ip
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: status.hostIP
        - name: _node_ip_
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: status.hostIP
        # Replace cn-hangzhou in the following image address with the region where your cluster is deployed.
        image: registry-cn-hangzhou-vpc.ack.aliyuncs.com/acs/logtail-windows:v1.0.20
        imagePullPolicy: IfNotPresent
        resources:
          limits:
            cpu: 1
            memory: 1Gi
          requests:
            cpu: 100m
            memory: 256Mi
        securityContext:
          privileged: false
        terminationMessagePath: /dev/termination-log
        terminationMessagePolicy: File
        volumeMounts:
        - name: docker-pipe
          mountPath: \\.\pipe\docker_engine
        - name: docker-data
          mountPath: c:/ProgramData/docker
          readOnly: true
        - mountPath: c:/logtail_host
          name: root
          readOnly: true
      terminationGracePeriodSeconds: 30
      priorityClassName: system-node-critical
      restartPolicy: Always
      terminationGracePeriodSeconds: 30
      tolerations:
      - operator: Exists
      volumes:
      - name: docker-pipe
        hostPath:
          path: \\.\pipe\docker_engine
      - name: docker-data
        hostPath:
          path: c:/ProgramData/docker
          type: Directory
      - name: root
        hostPath:
          path: c:/
          type: Directory
  updateStrategy:
    rollingUpdate:
      maxUnavailable: 10%
    type: RollingUpdate

Implante uma aplicação de teste

Com o Logtail em execução, implante a aplicação de teste abaixo para validar a coleta de logs. Essa aplicação gera saída contínua no stdout, capturada pelo Logtail.

apiVersion: apps/v1
kind: Deployment
metadata:
  labels:
    app: logtail-test
  name: logtail-test
spec:
  replicas: 1
  selector:
    matchLabels:
      app: logtail-test
  template:
    metadata:
      labels:
        app: logtail-test
    spec:
      containers:
      - name: nanoserver
        image: mcr.microsoft.com/windows/servercore:1809
        command: ["powershell.exe"]
        args: ["ping -t 127.0.0.1 -w 10000"]
        env:
      ######### Specify environment variables ###########
        - name: aliyun_logs_logtail-stdout
          value: stdout
        - name: aliyun_logs_logtail-tags
          value: tag1=v1
      #################################
      nodeSelector:
        beta.kubernetes.io/os: windows
      tolerations:
      - effect: NoSchedule
        key: os
        operator: Equal
        value: windows

Verifique a coleta de logs

Após iniciar a aplicação de teste, confirme se o Logtail está coletando sua saída:

Verifique se o Pod de teste está em execução:
```
kubectl get pods -l app=logtail-test
```
Confirme se o Pod está imprimindo logs no stdout:
```
kubectl logs <pod-name>
```
Linhas de saída do ping devem aparecer no log.
Abra o console do SLS e acesse o Logstore do projeto para confirmar a chegada das entradas de log. Para mais detalhes sobre consulta de logs, consulte Guia de consulta e análise de logs.

Próximos passos

Guia de consulta e análise de logs