Skip to content
Pre-release
Pre-release

@kmova kmova released this Sep 9, 2018 · 156 commits to master since this release

Getting Started

Prerequisite to install

  • Kubernetes 1.9.7+ is installed
  • Make sure that you have completed the below installation steps with cluster admin context. The installation will involve creating a new Service Account and assigning to OpenEBS components.
  • Make sure iSCSI Initiator is installed on the Kubernetes nodes.
  • NDM helps in discovering the devices attached to Kubernetes nodes, which can be used to create storage pools. If you like to exclude some of the disks from getting discovered, update the filters on NDM to exclude paths before installing OpenEBS.

Using kubectl

kubectl apply -f https://openebs.github.io/charts/openebs-operator-0.7.0.yaml

Using OpenEBS Helm Charts (will be deprecated in the coming releases)

helm repo add openebs-charts https://openebs.github.io/charts/
helm repo update
helm install openebs-charts/openebs

For more details refer to the documentation at: https://docs.openebs.io/

Note: Kubernetes stable/openebs helm chart and other charts still point to 0.6 and efforts are underway to update them to 0.7.

Quick Summary on changes

  • Node Disk Manager that helps with discovering block devices attached to nodes
  • Alpha support for cStor Storage Engines
  • Updated CRDs for supporting cStor as well as pluggable storage control plane
  • Jiva Storage Pool called default and StorageClass called openebs-jiva-default
  • cStor Storage Pool Claim called cstor-sparse-pool and StorageClass called openebs-cstor-sparse
  • There has been a change in the way volume storage policies can be specified with the addition of new policies like:
    • Number of Data copies to be made
    • Specify the nodes on which the Data copies should be persisted
    • Specify the CPU or Memory Limits per PV
    • Choice of Storage Engine : cStor or Jiva

Sample Storage Pool Claims, Storage Class and PVC configurations to make use of new features can be found here: Sample YAMLs

Detailed release notes are maintained in Project Tracker Wiki.

Limitations

  • cStor Target or Pool pods can at times be stuck in a Terminating state. They will need to be manually cleaned up using kubectl delete with 0 sec grace period.
  • Jiva target to Replica message protocol has been enhanced to handle the write errors. This change in the data exchanges causes the older replicas to be incompatible with the newer target and vice versa. The upgrade involves shutting down all the replicas before launching them with the new version. Since the volume requires the target and at least 2 replicas to be online, chances of volumes getting into the read-only state during upgrade are higher. A manual intervention will be required to recover the volume.
  • For OpenEBS volumes configured with more than 1 replica, at least more than half of the replicas should be online for the Volume to allow Read and Write. In the upcoming releases, with cStor data engine, Volumes can be allowed to Read/Write when there is at least one replica in the ready state.
  • This release contains a preview support for cloning an OpenEBS Volume from a snapshot. This feature only supports single replica for a cloned volume, which is intended to be used for temporarily spinning up a new application pod for recovering lost data from the previous snapshot.
  • While testing for different platforms, with a three-node/replica OpenEBS volume and shutting down one of the three nodes, there was an intermittent case where one of the 2 remaining replicas also had to be restarted.
  • The OpenEBS target (controller) pod depends on the Kubernetes node tolerations to reschedule the pod in the event of node failure. For this feature to work, TaintNodesByCondition alpha feature must be enabled in Kubernetes. In a scenario where OpenEBS target (controller) is not rescheduled or is back to running within 120 seconds, the volume gets into a read-only state and a manual intervention is required to make the volume as read-write.
  • The current version of OpenEBS volumes are not optimized for performance sensitive applications.

For a more comprehensive list of open issues uncovered through e2e, please refer to open issues.

Assets 2
You can’t perform that action at this time.