Unable to mount multiple sub datasets
k8s: 1.25.4 fluid: 0.9.1
When I try to mount two subsets in one pod, it shows timeout
dataset:
subdataset:
pod event:
@lgy1027 Could you please provide the yaml for the dataset and runtime ?
@lgy1027 Could you please provide the yaml for the dataset and runtime ?
dataset.yaml apiVersion: data.fluid.io/v1alpha1 kind: Dataset metadata: name: all namespace: test spec: sharedOptions: alluxio.underfs.s3.disable.dns.buckets: "true" alluxio.underfs.s3.endpoint: xxxxxxxxxx alluxio.underfs.s3.inherit.acl: "false" s3a.accessKeyId: xxxxxxxxx s3a.secretKey: xxxxxxxxxxxxxxx mounts:
- mountPoint: s3://test01 name: test01
- mountPoint: s3://test02 name: test02
apiVersion: data.fluid.io/v1alpha1 kind: AlluxioRuntime metadata: name: all namespace: test spec: replicas: 1 tieredstore: levels: - mediumtype: SSD path: /opt/lgy quota: 10Gi high: "0.95" low: "0.7"
subdateset.yaml apiVersion: data.fluid.io/v1alpha1 kind: Dataset metadata: name: test01 spec: mounts: - mountPoint: dataset://test/all/test01 name: test01
apiVersion: data.fluid.io/v1alpha1 kind: Dataset metadata: name: test02 spec: mounts: - mountPoint: dataset://test/all/test02 name: test02
nginx.yaml apiVersion: v1 kind: Pod metadata: labels: sidecar.istio.io/inject: "false" name: nginx spec: containers: - name: nginx image: nginx volumeMounts: - mountPath: /data/test01 name: test01 - mountPath: /data/test02 name: test02 volumes: - name: test01 persistentVolumeClaim: claimName: test01 - name: test02 persistentVolumeClaim: claimName: test02
@cheyang This bug is caused because sub-dataset' pv use the same volumeHandle name as referenced dataset's pv, therefore k8s think the pod have two same pv resulting in mount timeout. Do we have a plan to fix this or just state in the document that a pod can not use multiple sub datasets pointing to the same dataset?