我正在尝试在minikube中运行服务（MetaMap Tagger服务，medpost-skr）。
我有以下Docker文件，它自己运行良好：
FROM openjdk:8-jre
RUN mkdir /usr/share/public_mm
ADD public_mm/MedPost-SKR /usr/share/public_mm/MedPost-SKR
ADD public_mm/bin /usr/share/public_mm/bin
ADD required/scripts /usr/share/public_mm/scripts
ENV PATH=/usr/share/public_mm/bin:$PATH
WORKDIR /usr/share/public_mm
# --- BUILD ---
RUN /usr/share/public_mm/bin/install.sh
# --- START ---
CMD skrmedpostctl start
我有以下服务/部署文件（在argo工作流模板中）来启动服务：
apiVersion: argoproj.io/v1alpha1
kind: Workflow #new type of k8s spec
metadata:
generateName: nlp-adapt-wf- #name of workflow spec
spec:
entrypoint: nlp-adapt-metamap-services #invoke the build template
templates:
- name: nlp-adapt-metamap-services
steps:
- - name: medpost-deployment
template: medpost-server-d
- - name: medpost-service
template: medpost-server-s
- name: medpost-server-d
resource:
action: create
manifest: |
apiVersion: apps/v1
kind: Deployment
metadata:
name: medpost
spec:
selector:
matchLabels:
app: medpost
track: stable
template:
metadata:
labels:
app: medpost
track: stable
spec:
containers:
- image: ahc-nlpie-docker.artifactory.umn.edu/medpost-skr
imagePullPolicy: Never
name: medpost
ports:
- containerPort: 1795
- name: medpost-server-s
resource:
action: create
manifest: |
apiVersion: v1
kind: Service
metadata:
name: medpost
namespace: default
labels:
app: medpost
spec:
selector:
app: medpost
ports:
- name: test1
protocol: TCP
port: 1795
targetPort: 1795
当我查看服务时，我看到：
（基数）D20181472：medpost-skr gms $ kubectl get svc名称类型集群-IP外部-IP端口（S）AGE kubernetes ClusterIP 10.96.0.1 443 / TCP 60d medpost ClusterIP 10.100.189.245 1795 / TCP 26m
当我查看部署时，看起来服务部署从未启动过：
(base) D20181472:medpost-skr gms$ kubectl get deploy
NAME READY UP-TO-DATE AVAILABLE AGE
medpost 0/1 1 0 27m
但是，当我查看pod的日志时，服务似乎已启动，但pod处于CrashLoopBackOff状态：
(base) D20181472:medpost-skr gms$ kubectl get pods
NAME READY STATUS RESTARTS AGE
medpost-859844ddf-sw888 0/1 CrashLoopBackOff 10 28m
(base) D20181472:medpost-skr gms$ kubectl logs medpost-859844ddf-sw888
$Starting skrmedpostctl:
started.
以下是有关pod的更多信息：
kubectl describe pod medpost-859844ddf-sw888
Name: medpost-859844ddf-sw888
Namespace: default
Priority: 0
PriorityClassName: <none>
Node: minikube/10.0.2.15
Start Time: Tue, 02 Apr 2019 14:14:57 -0500
Labels: app=medpost
pod-template-hash=859844ddf
track=stable
Annotations: <none>
Status: Running
IP: 172.17.0.7
Controlled By: ReplicaSet/medpost-859844ddf
Containers:
medpost:
Container ID: docker://208de2cc7c5007b53b63f2a063db13a8df5011488ab0be26726c12de54735043
Image: ahc-nlpie-docker.artifactory.umn.edu/medpost-skr
Image ID: docker://sha256:89ea83d80cd906319afbdd37805228440048aded93156c0fc17f900d1aa6228b
Port: 1795/TCP
Host Port: 0/TCP
State: Waiting
Reason: CrashLoopBackOff
Last State: Terminated
Reason: Completed
Exit Code: 0
Started: Tue, 02 Apr 2019 14:41:17 -0500
Finished: Tue, 02 Apr 2019 14:41:17 -0500
Ready: False
Restart Count: 10
Environment: <none>
Mounts:
/var/run/secrets/kubernetes.io/serviceaccount from default-token-mkpqk (ro)
Conditions:
Type Status
Initialized True
Ready False
ContainersReady False
PodScheduled True
Volumes:
default-token-mkpqk:
Type: Secret (a volume populated by a Secret)
SecretName: default-token-mkpqk
Optional: false
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute for 300s
node.kubernetes.io/unreachable:NoExecute for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal Scheduled 30m default-scheduler Successfully assigned default/medpost-859844ddf-sw888 to minikube
Normal Pulled 28m (x5 over 30m) kubelet, minikube Container image "ahc-nlpie-docker.artifactory.umn.edu/medpost-skr" already present on machine
Normal Created 28m (x5 over 30m) kubelet, minikube Created container
Normal Started 28m (x5 over 30m) kubelet, minikube Started container
Warning BackOff 5m6s (x117 over 30m) kubelet, minikube Back-off restarting failed container
这里编辑是部署：
(base) D20181472:wsd_server gms$ kubectl describe deploy medpost
Name: medpost
Namespace: default
CreationTimestamp: Tue, 02 Apr 2019 18:03:13 -0500
Labels: <none>
Annotations: deployment.kubernetes.io/revision: 1
Selector: app=medpost,track=stable
Replicas: 1 desired | 1 updated | 1 total | 0 available | 1 unavailable
StrategyType: RollingUpdate
MinReadySeconds: 0
RollingUpdateStrategy: 25% max unavailable, 25% max surge
Pod Template:
Labels: app=medpost
track=stable
Containers:
medpost:
Image: ahc-nlpie-docker.artifactory.umn.edu/medpost-skr
Port: 1795/TCP
Host Port: 0/TCP
Environment: <none>
Mounts: <none>
Volumes: <none>
Conditions:
Type Status Reason
---- ------ ------
Available False MinimumReplicasUnavailable
Progressing True ReplicaSetUpdated
OldReplicaSets: <none>
NewReplicaSet: medpost-859844ddf (1/1 replicas created)
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal ScalingReplicaSet 34s deployment-controller Scaled up replica set medpost-859844ddf to 1
不知道该怎么做？
问题是这些服务是作为后台进程启动的。在前台启动它们就是解决方案。