Network Management

The Network Management module in eEKAS provides a centralized, intuitive interface for configuring
 and monitoring all network interfaces and connections across the cluster. In a Ceph-basedbased, high-availability
 environment, reliable and well-optimizedstructured networking is criticala fundamental requirement for both storage performance and service
 continuity. This module ensuresenables administrators can easilyto define, adjust, and troubleshoot network configurations
 without interruptingdisrupting running services.

To ensure stable cluster operation, eEKAS requires a minimum of two logically separated networks. These networks serve distinct purposes and are critical to maintaining availability, consistency, and predictable performance under both normal operation and failure conditions.

Cluster / Internal Communication Network: This network is used for node-to-node communication, cluster coordination, and heartbeat signaling. It allows the cluster to continuously monitor node health, detect failures, and trigger automated failover actions. A dedicated and low-latency internal network is essential to ensure fast failure detection and stable high-availability behavior.

Ceph Storage Network: The Ceph network is used exclusively for storage-related traffic such as data replication, recovery, and rebalancing between nodes. Isolating this traffic prevents storage operations from interfering with client access and ensures consistent performance, especially during rebuild or recovery scenarios.

In eEKAS,addition to these mandatory networks, one or more client-facing networks are typically configured to provide access to storage services such as SMB, NFS, iSCSI, NVMe-oF, and S3. Separating client traffic from internal and storage replication traffic significantly improves scalability, security, and overall system stability.

For block and file services, eEKAS networking is built around IP Groups—logical collections of one or more IP addresses that
 can be assigned to specific services such as S3, SMB, NFS, iSCSI, or NVMe-oF.services. IP Groups are highly flexible:
they can be moved between cluster nodes during planned maintenance or automatically during failover events, ensuring
 uninterrupted client connectivity.

Key Capabilities

• Centralized Interface Management – Configure physical interfaces, bonded interfaces for
  redundancy/ redundancy and performance, and VLAN-tagged interfaces across all nodes from a single screen.interface.
• VLAN Support – Create and assign VLANs to logicallyLogically separate traffic (e.g.,types such as management, cluster communication, storage
   replication, and client access)access without addingrequiring moreadditional physical NICs. VLANs help segment and prioritize traffic while
  improving security.
• Multi-Network SupportArchitecture – IsolateEnforce storagestrict backendseparation trafficbetween frominternal cluster traffic, Ceph replication traffic, and client-facing trafficservices to enhanceimprove both
  performance and stability.resilience.
• High Availability Integration – IPAutomatic Groupsfailover automaticallymechanisms failensure overservices betweenremain nodesavailable ineven case ofduring node
  failure, keepingor clienthardware connections active.failures.
• Real-Time Monitoring – TrackMonitor link state,status, bandwidth usage,utilization, and traffic statisticspatterns for proactive
  troubleshooting. diagnostics and capacity planning.

By combining powerfula configurationmulti-network toolsarchitecture with centralized management and automated failover logic, eEKAS Network Management ensures that your
cluster maintains stable, high-performance connectivity—supportingconnectivity for both the storage backend and the client-facing
services that relydepend on it.

Example VLAN Scenario

• VLAN 10 – Management traffic:& Cluster Communication: clusterNode administration,management, monitoring, and control.heartbeat traffic.
• VLAN 20 – Ceph Storage replicationNetwork: traffic:Replication, recovery, and internal Ceph internal communication between nodes.communication.
• VLAN 30 – Client access traffic:Access: SMB/NFS/iSCSI/SMB, NFS, iSCSI, NVMe-oFoF, services.and optional S3 access.

This layoutseparation keepsensures eachthat trafficcritical typecluster logicallyand storage operations remain isolated, from client workloads, reducing contention andwhile improving bothperformance, performancepredictability, and security. security.

YouThe model can extendbe the modelextended with additional networks or VLANs for S3, backup traffic, dedicated S3 access, or other specialized workloads as needed.required.