Hello
This is my first setup with Apache Beam and trying to use the flink runner
for it. Following the example of
https://github.com/apache/flink-kubernetes-operator/tree/main/examples/flink-beam-example
I got my app running inside my local minikube. The app is pretty stateless
- minimal transformation between two IO.
Now, while I was testing things, my Mac decided to upgrade and thus
minikube was also restarted. After the restart, it seems like the
Flinkdeployment was still installed but no jobmanager or task managers pods
were running and doing get flinkdeployment shows it to be:
Job Status: Reconciling and Reconciliation: Deployed.
Description of the job is as follows:
Name: myapp-beam-example
Namespace: default
Labels: <none>
Annotations: <none>
API Version: flink.apache.org/v1beta1
Kind: FlinkDeployment
Metadata:
Creation Timestamp: 2023-02-17T05:38:33Z
Finalizers:
flinkdeployments.flink.apache.org/finalizer
Generation: 3
Managed Fields:
API Version: flink.apache.org/v1beta1
Fields Type: FieldsV1
fieldsV1:
f:metadata:
f:finalizers:
.:
v:"flinkdeployments.flink.apache.org/finalizer":
f:spec:
f:job:
f:state:
f:jobManager:
f:replicas:
Manager: fabric8-kubernetes-client
Operation: Update
Time: 2023-02-17T05:38:33Z
API Version: flink.apache.org/v1beta1
Fields Type: FieldsV1
fieldsV1:
f:spec:
.:
f:flinkConfiguration:
.:
f:taskmanager.numberOfTaskSlots:
f:flinkVersion:
f:image:
f:job:
.:
f:args:
f:entryClass:
f:jarURI:
f:upgradeMode:
f:jobManager:
.:
f:resource:
.:
f:cpu:
f:memory:
f:serviceAccount:
f:taskManager:
.:
f:resource:
.:
f:cpu:
f:memory:
Manager: kubectl-create
Operation: Update
Time: 2023-02-17T05:38:33Z
API Version: flink.apache.org/v1beta1
Fields Type: FieldsV1
fieldsV1:
f:metadata:
f:annotations:
.:
f:kubectl.kubernetes.io/last-applied-configuration:
f:spec:
f:job:
f:parallelism:
Manager: kubectl-client-side-apply
Operation: Update
Time: 2023-02-17T05:45:11Z
API Version: flink.apache.org/v1beta1
Fields Type: FieldsV1
fieldsV1:
f:status:
.:
f:clusterInfo:
.:
f:flink-revision:
f:flink-version:
f:jobManagerDeploymentStatus:
f:jobStatus:
.:
f:jobId:
f:jobName:
f:savepointInfo:
.:
f:lastPeriodicSavepointTimestamp:
f:savepointHistory:
f:startTime:
f:state:
f:updateTime:
f:reconciliationStatus:
.:
f:lastReconciledSpec:
f:lastStableSpec:
f:reconciliationTimestamp:
f:state:
f:taskManager:
.:
f:labelSelector:
f:replicas:
Manager: fabric8-kubernetes-client
Operation: Update
Subresource: status
Time: 2023-02-18T10:15:42Z
Resource Version: 192678
UID: 71b851e5-4b99-434a-9027-d528eb8b1180
Spec:
Flink Configuration:
taskmanager.numberOfTaskSlots: 1
Flink Version: v1_16
Image: myapp-beam:latest
Job:
Args:
--runner=FlinkRunner
--bootstrapServers=kafka-service:9092
Entry Class: com.netskope.myapp.myapp
Jar URI: local:///opt/flink/usrlib/beam-runner.jar
Parallelism: 2
State: running
Upgrade Mode: stateless
Job Manager:
Replicas: 1
Resource:
Cpu: 1
Memory: 2048m
Service Account: flink
Task Manager:
Resource:
Cpu: 1
Memory: 2048m
Status:
Cluster Info:
Flink - Revision: DeadD0d0 @ 1970-01-01T01:00:00+01:00
Flink - Version: 1.16.1
Job Manager Deployment Status: MISSING
Job Status:
Job Id: adb5186e96f6d36637f919fb5d9aad70
Job Name: myapp-flink-0217055545-790af232
Savepoint Info:
Last Periodic Savepoint Timestamp: 0
Savepoint History:
Start Time: 1676613348905
State: RECONCILING
Update Time: 1676628937180
Reconciliation Status:
Last Reconciled Spec:
{"spec":{"job":{"jarURI":"local:///opt/flink/usrlib/beam-runner.jar","parallelism":2,"entryClass":"com.netskope.myapp.myapp","args":["--runner=FlinkRunner","--bootstrapServers=kafka-service:9092"],"state":"running","savepointTriggerNonce":null,"initialSavepointPath":null,"upgradeMode":"stateless","allowNonRestoredState":null},"restartNonce":null,"flinkConfiguration":{"taskmanager.numberOfTaskSlots":"1"},"image":"myapp-beam:latest","imagePullPolicy":null,"serviceAccount":"flink","flinkVersion":"v1_16","ingress":null,"podTemplate":null,"jobManager":{"resource":{"cpu":1.0,"memory":"2048m"},"replicas":1,"podTemplate":null},"taskManager":{"resource":{"cpu":1.0,"memory":"2048m"},"replicas":null,"podTemplate":null},"logConfiguration":null,"mode":null},"resource_metadata":{"apiVersion":"
flink.apache.org/v1beta1
","metadata":{"generation":3},"firstDeployment":false}}
Last Stable Spec:
{"spec":{"job":{"jarURI":"local:///opt/flink/usrlib/beam-runner.jar","parallelism":2,"entryClass":"com.netskope.myapp.myapp","args":["--runner=FlinkRunner","--bootstrapServers=kafka-service:9092"],"state":"running","savepointTriggerNonce":null,"initialSavepointPath":null,"upgradeMode":"stateless","allowNonRestoredState":null},"restartNonce":null,"flinkConfiguration":{"taskmanager.numberOfTaskSlots":"1"},"image":"myapp-beam:latest","imagePullPolicy":null,"serviceAccount":"flink","flinkVersion":"v1_16","ingress":null,"podTemplate":null,"jobManager":{"resource":{"cpu":1.0,"memory":"2048m"},"replicas":1,"podTemplate":null},"taskManager":{"resource":{"cpu":1.0,"memory":"2048m"},"replicas":null,"podTemplate":null},"logConfiguration":null,"mode":null},"resource_metadata":{"apiVersion":"
flink.apache.org/v1beta1
","metadata":{"generation":3},"firstDeployment":false}}
Reconciliation Timestamp: 1676612715537
State: DEPLOYED
Task Manager:
Label Selector: component=taskmanager,app=myapp-beam-example
Replicas: 2
Events: <none>
tried to search for this on Github and in the mailing list archives,
without luck.
Just trying to understand what this means and how will it affect things if
some failures like this happen in production.. If somebody can point me to
the relevant document about this, that would be helpful too.
--
Ritesh
