Damien/arista-evpn-vxlan-clab
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add Grafana monitoring stack with gNMI telemetry and Network Weathermap #17

Closed
Damien wants to merge 28 commits from feature/grafana-monitoring into main
pull from: feature/grafana-monitoring
merge into: Damien:main
Conversation 0 Commits 28 Files Changed 17 +2918
Damien commented 2025-12-16 13:27:19 +00:00
Owner
Copy Link

Summary

Adds a complete monitoring stack for the EVPN-VXLAN fabric using gNMI streaming telemetry with gnmic, Prometheus, and Grafana Network Weathermap.

Components

  • gnmic: Collects gNMI telemetry from all 10 switches (port 6030)
  • Prometheus: Stores metrics with 15-day retention
  • Grafana: Visualizes with Weathermap plugin pre-installed

Files Added

  • monitoring/docker-compose.yml - Stack orchestration
  • monitoring/gnmic/gnmic.yaml - gNMI collector config for all devices
  • monitoring/prometheus/prometheus.yml - Scrape configuration
  • monitoring/grafana/dashboards/ - Pre-built dashboards including weathermap
  • monitoring/deploy.sh - Quick deployment script

Usage

cd monitoring
docker-compose up -d
# Grafana: http://localhost:3000 (admin/admin)

Collected Metrics

  • Interface counters (in/out octets, packets, errors)
  • BGP neighbor state
  • MLAG status
  • System metrics (CPU, memory)
## Summary Adds a complete monitoring stack for the EVPN-VXLAN fabric using gNMI streaming telemetry with gnmic, Prometheus, and Grafana Network Weathermap. ## Components - **gnmic**: Collects gNMI telemetry from all 10 switches (port 6030) - **Prometheus**: Stores metrics with 15-day retention - **Grafana**: Visualizes with Weathermap plugin pre-installed ## Files Added - `monitoring/docker-compose.yml` - Stack orchestration - `monitoring/gnmic/gnmic.yaml` - gNMI collector config for all devices - `monitoring/prometheus/prometheus.yml` - Scrape configuration - `monitoring/grafana/dashboards/` - Pre-built dashboards including weathermap - `monitoring/deploy.sh` - Quick deployment script ## Usage ```bash cd monitoring docker-compose up -d # Grafana: http://localhost:3000 (admin/admin) ``` ## Collected Metrics - Interface counters (in/out octets, packets, errors) - BGP neighbor state - MLAG status - System metrics (CPU, memory)
Damien added 9 commits 2025-12-16 13:27:19 +00:00
Add monitoring stack README 2762a5040b
Add gnmic configuration for gNMI telemetry collection 442211ed5b
Add Prometheus configuration da5a8997d3
Add Grafana datasource provisioning for Prometheus 35123308c2
Add Grafana dashboard provisioning configuration 6f873c8584
Add Grafana fabric overview dashboard c975945d27
Add Docker Compose for monitoring stack c12bd2a701
Add monitoring stack deployment script 1c08b156d6
Add Network Weathermap dashboard template 92e8556e1f
Damien added 1 commit 2025-12-16 13:44:43 +00:00
Fix gnmic config: remove mlag and vxlan subscriptions (not available via OpenConfig on cEOS) d01598f9ce
Damien added 1 commit 2025-12-16 14:19:41 +00:00
fix(gnmic): rewrite config with correct parameters and simplified paths 5fdf374fa4 
- Remove invalid 'add-target: target' (must be overwrite|if-not-present|empty)
- Enable debug mode for troubleshooting
- Simplify interface paths to /interfaces/interface/state (Arista compatible)
- Simplify system paths to /system/state
- Remove complex BGP path that may not work on cEOS
- Add retry and timeout parameters for reliability
- Add expiration to prevent stale metrics
- Add skip-verify for insecure connections
- Increase sample intervals for stability
Damien added 1 commit 2025-12-16 14:21:46 +00:00
fix(gnmic): remove skip-verify (mutually exclusive with insecure) 6c08b9ecf7 
The flags --insecure and --skip-verify are mutually exclusive in gNMIc.
Since we're using insecure connections (no TLS), skip-verify is not needed.
Damien added 1 commit 2025-12-16 14:26:03 +00:00
fix(grafana): correct metric names in weathermap queries ca55e2ff59 
Changed from:
- gnmic_interfaces_interface_state_counters_out_octets
- gnmic_interfaces_interface_state_counters_in_octets
- target label

To:
- gnmic_interfaces_out_octets
- gnmic_interfaces_in_octets  
- source label

These match the actual metrics generated by gNMIc with the simplified
/interfaces/interface/state path and trim-prefixes processor.
Damien added 1 commit 2025-12-16 14:26:36 +00:00
fix(grafana): correct metric names in fabric-overview dashboard b23353bf15 
Changed from:
- gnmic_interfaces_interface_state_counters_* with target label

To:
- gnmic_interfaces_* with source label

Also added:
- Interfaces Monitored stat panel
- MLAG Peer-Link Traffic panel

These match the actual metrics generated by gNMIc.
Damien added 1 commit 2025-12-16 18:51:31 +00:00
Enhance gnmic config for Flow Plugin support with BGP/EVPN telemetry b34b0eed7d
Damien added 1 commit 2025-12-16 18:51:48 +00:00
Enhance Prometheus config with better metric filtering for Flow Plugin b77f461967
Damien added 1 commit 2025-12-16 18:52:08 +00:00
Update docker-compose to use Flow Plugin instead of archived weathermap 011541b7f2
Damien added 1 commit 2025-12-16 18:52:43 +00:00
Create Flow Plugin-based topology dashboard to replace weathermap 903522dd82
Damien added 1 commit 2025-12-16 18:53:42 +00:00
Add comprehensive configuration review documentation 4b657a4e1e
Damien added 1 commit 2025-12-16 18:54:17 +00:00
Add quick start deployment guide for monitoring stack bcb3160c9b
Damien added 1 commit 2025-12-16 19:48:44 +00:00
Fix gnmic config: remove invalid VXLAN/routing paths causing subscription errors f79a3bdd38
Damien added 1 commit 2025-12-16 19:49:25 +00:00
Add documentation on Arista cEOS gNMI path compatibility and troubleshooting 53b585e6b3
Damien added 1 commit 2025-12-16 19:50:04 +00:00
Add comprehensive summary of gnmic configuration fix for Arista compatibility 33407445fb
Damien added 1 commit 2025-12-16 20:06:59 +00:00
Add guide explaining VXLAN monitoring through existing metrics 2ac535cdae
Damien added 1 commit 2025-12-16 20:07:45 +00:00
Add final deployment status and readiness checklist 1b537db918
Damien added 1 commit 2025-12-16 20:14:54 +00:00
Add VXLAN subscription for Arista augmented Vxlan1 interface data 761f10efc4
Damien added 1 commit 2025-12-16 20:15:39 +00:00
Document successful VXLAN telemetry discovery via Vxlan1 interface 8220e55b0e
Damien added 1 commit 2025-12-16 21:16:17 +00:00
fix(gnmic): preserve full OpenConfig path-based metric names 0196b64232 
Changes:
- Remove path-base transform that was stripping metric names to just leaf elements
- Change VXLAN subscription from on_change to sample mode (30s interval)
  to ensure consistent metric collection
- Remove unused event-processors from Prometheus output
- Clean up processor configuration

This fixes metric naming to match Grafana dashboard expectations:
- Before: gnmic_interfaces_out_octets
- After: gnmic_interfaces_interface_state_counters_out_octets

The full path names provide better clarity and match standard OpenConfig
metric naming conventions used in dashboards.
Damien closed this pull request 2025-12-26 14:49:51 +00:00
Damien deleted branch feature/grafana-monitoring 2025-12-26 14:49:51 +00:00

Pull request closed

This pull request cannot be reopened because the branch was deleted.
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
Damien
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Damien/arista-evpn-vxlan-clab#17
Delete Branch "feature/grafana-monitoring"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?