High Availability
High Availability#
Replicated PV Mayastor 2.0 enhances High Availability (HA) of the volume target with the nexus switch-over feature. In the event of the target failure, the switch-over feature quickly detects the failure and spawns a new nexus to ensure I/O continuity. The HA feature consists of two components:
- HA node agent (which runs in each csi- node) and
- Cluster agent (which runs alongside the agent-core).
The HA node agent looks for I/O path failures from applications to their corresponding targets. If any such broken path is encountered, the HA node agent informs the cluster agent. The cluster-agent then creates a new target on a different (live) node. Once the target is created, the node-agent establishes a new path between the application and its corresponding target. The HA feature restores the broken path within seconds, ensuring negligible downtime.
warning
The volume's replica count must be higher than one for a new target to be established as part of switch-over.
To ensure optimal HA in Replicated PV Mayastor clusters:
Applications constrained to nodes with the
io-engine label(openebs.io/engine=mayastor) will have the Nexus preferably placed on the same node where the application is scheduled. If theio-enginepod on a node is in a bad state, the Nexus may be placed on a different node.The kernel parameter
nvme_core.multipath=Yis mandatory to enable HA functionality. Without this configuration, volume target failover is not possible.
How do I disable this feature?#
info
We strongly recommend keeping this feature enabled.
The HA feature is enabled by default; to disable it, pass the parameter --set=agents.ha.enabled=false with the helm install command.