Description
Kubernetes provider is dynamically creating new stream for each container for example for that config, but the resulting stream will have the same id
streams:
- id: filestream-kubernetes.container_logs
data_stream:
dataset: kubernetes.container_logs
type: logs
prospector.scanner.symlinks: true
paths:
- '/var/log/containers/*${kubernetes.container.id}.log'It is possible in a non managed policy to bypass that problem by providing an id with a dynamic variable
like this
streams:
- id: >-
filestream-kubernetes.container_logs-${kubernetes.container.id}
data_stream:
dataset: kubernetes.container_logs
type: logs
prospector.scanner.symlinks: true
paths:
- '/var/log/containers/*${kubernetes.container.id}.log'How should we solve that for managed solutions?
Potential solution
A fix should be done on the input side of things, we should first discuss what mechanism will be used for this.
Follow up of: elastic/kibana#129851
Description
Kubernetes provider is dynamically creating new stream for each container for example for that config, but the resulting stream will have the same
idIt is possible in a non managed policy to bypass that problem by providing an id with a dynamic variable
like this
How should we solve that for managed solutions?
Potential solution
A fix should be done on the input side of things, we should first discuss what mechanism will be used for this.
Follow up of: elastic/kibana#129851