-
Notifications
You must be signed in to change notification settings - Fork 7
PodReaper
-
Remove completed pods (error or successful)
- This is primarily due to the fact that the executor is launched un-managed (the puller does not track the pod id at all)
-
Run on an interval (see CronJob.yaml)
-
Configurable reap age for pods.
N/A - There are no inputs for the pod reaper, only configuration settings.
ResourcePool: Handler common command line arguments
ResourcePool: Handler common environment variables
ResourcePool: Handler common property file values
Property | Value Type | Notes |
---|---|---|
cp.kubernetes.masterUrl | String | Generally not needed if running as a pod. Will default to the cluster the pod is running in. |
cp.kubernetes.namespace | String | |
pod.reap.age.minutes | Number | The age in minutes that a pod must be to be reaped. |
pod.reap.batch.size | Number | The amount of items to delete in a single request to kubernetes. |
- Submission
- Scheduling
-
Execution
the ResourcePool
Agenda
the workflow
Agenda Template
the workflow definition
Customer
Insight
the scheduling queue definition
Operation Progress
the state of the running Agenda operations
Progress
the state of the running Agendas
ResourcePool
the processing resources
Agenda Service
the workflow submission
Progress Service
rolled up agenda progress summary
ResourcePool Service
getting work and updating progress
AgendaReclaimer
restarting stuck Agendas
AgendaRetry
retrying failed Agendas
DataObjectReaper
reaping expired data objects
PodReaper
reaping stuck Kubernetes pods
DevKubernetesSetup
RunLocalExecution
- RunWithMiniKube