Arista Cloud Engineer: DANZ Monitoring Fabric (DMF) Specialist (DMF) – Details

Detaillierter Kursinhalt

Arista Monitoring Solutions
  • Network monitoring
    • Needs and Challenges
    • New generation of NPB
  • LANZ - Latency Analyzer
    • Operation, Congestion, Latency, Packet Drops
    • LANZ report from CLI and CloudVision
  • Mirroring and TapAgg
  • Introduction to DMF
    • Single point of management
    • Zero Touch Fabric
    • Dynamic Monitoring of VM Workloads
    • Application identification
    • Integrated packet capture
  • OOB topologies
    • Single-switch, two-level, multi-level fabric
    • Switch Cluster Fabric
    • Multi-site manufacturing
  • Use case
    • Ubiquitous Network Monitoring
    • Multi-site monitoring
DMF deployment
  • Structure components
    • Controllers
    • Switches
    • DMF Nodes - Service, Logger, Analytics
    • Management network
  • Interface connections
  • Structural Design Considerations
    • Platform option - Network Appliance or VM
    • HA Configuration - Standalone or Cluster
  • Structure commissioning sequence
    • Management network, controller, switches, nodes
    • Configuration
Access and configuration of the DMF
  • Access modes - GUI, CLI, REST
  • Configuration Management
    • Setup Run
    • File management
  • System configuration - AAA, SNMP, NTP, others
  • Repair
DMF Policy Configurations
  • Interface - Roles and configuration
  • Policies - Matching Rules, Actions
    • Example - Monitor ICMP and Web Traffic
  • Service Nodes - Managed Services, Build Policy
  • Record Node - Policymaking, Querying, Replaying
  • Policy configuration
Advanced DMF configurations
  • Modes - Packet Matching and VLAN Tagging
  • Interface Types and Options - Rate Limiting
  • Packet Capture - Configuration, PCAP Files
  • Tunnels - Requirements, Setup
  • Monitoring vCenter VMs
  • Deep Packet Matching Capabilities
  • Managed Services - Netflow, UDP Replication, Configuration
  • Monitoring as a Service (MaaS)
  • Provision of VM-based tools
DMF operations
  • Upgrades
    • DMF fabrications, controllers, switches, and nodes
    • Upgrade Procedure and Rollback
  • zero touch
    • Automatic leveling by controller
    • Switches/SN/RN
  • High availability
    • Failure of a service node
  • RMA procedure
    • Change of controller
    • Remove/Add new controller
  • Change management
    • Instant configurations
  • Maintenance
DMF Analysis Nodes
  • DMF Analytics
    • Visibility of the production network
    • Analytics as Flow Collector
  • Deployment
    • Single node cluster
    • Cluster of nodes
  • Internal Stack Overview
    • ELK - ElasticSearch Logstash Kibana
    • ELK stack built into the analytics node
  • Dashboards
    • sFlow, Netflow, TCP, ARP, DHCP, DNS
    • Customizing Dashboards
  • Machine-learning
    • Steps in creating an ML job
    • Types of MLnJob and Data Viewer
    • Single Metric ML Job, Multi-Metric ML Job
    • ML Anomaly Explorer
  • Application Dependency Maps (ADM)
    • Impact analysis
    • Steps in creating an ADM
  • Alerts - Watchmen
    • Threshold Alert, Advanced Alerts, Email Alert
    • Monitoring - Webhook and Slack integration
  • Personalization
    • Infoblox IPAM Integration
  • ELK Basics, Elasticsearch Statistics
  • Analytical clustering
  • Repair
DMF Recorder Node
  • Hardware, Deployment, Capabilities
  • Scaling
    • Added AI
    • Exploitation of external storage by NFS mounting
  • Operations - Configuration, Troubleshooting, Upgrade
  • Performance
  • RBAC, role-based access control
  • Using the Record Node
  • REST-APIs
DMF Troubleshooting
  • Fabric state - controller, switch, policies
  • Transfer - Model, Modes, Pipeline
  • Organizational chart
Contact Support
  • Open a support case
Appendix: Multi Cloud Director (MCD)
  • Dashboards
  • DMF Health Planning and Operations