vrr/Pulse
2
0
Fork 0
mirror of https://github.com/rcourtman/Pulse.git synced 2026-05-01 21:10:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6
Pulse/docs/UNIFIED_RESOURCES.md
rcourtman 778a2577b6 feat: Pulse v6 release
2026-03-18 16:06:30 +00:00

4.8 KiB
Raw Blame History

Unified Resource Model

Pulse v6 introduces a unified resource model that normalizes all monitored infrastructure — Proxmox VE, Proxmox Backup Server, Proxmox Mail Gateway, Docker, host agents, Kubernetes, and TrueNAS — into a single, consistent data structure.

Why Unified Resources?

In earlier versions, each platform had its own data model, API endpoints, and frontend pages. This created:

  • Duplicate UI code for each platform
  • Inconsistent filtering and search
  • No cross-platform comparison
  • Separate alert logic per platform

The unified model eliminates this by representing every resource as a single Resource struct with a common set of fields plus optional platform-specific extensions.

Core Concepts

Resource

Every monitored entity is a Resource with:

Field Description
id Globally unique identifier
name Display name
type host, vm, container, storage, pool, dataset, disk, service, cluster, pod, deployment
status online, warning, critical, offline, unknown
sources Array of contributing data sources (e.g., ["pve", "agent"])
sourceStatus Per-source health status
metrics CPU, memory, disk, network (when available)
Platform extensions .kubernetes, .truenas, .docker, etc.

Sources

A single resource can be reported by multiple sources. For example, a Proxmox node might have data from both the PVE API and a host agent:

sources: ["pve", "agent"]
sourceStatus:
  pve: { status: "online", lastSeen: "..." }
  agent: { status: "online", lastSeen: "..." }

The aggregate status is computed from all contributing sources.

Data Sources

Source What it feeds
pve Proxmox VE API — nodes, VMs, containers, storage
pbs Proxmox Backup Server — datastores, backups, sync jobs
pmg Proxmox Mail Gateway — mail stats, cluster health
agent Unified agent — host metrics, temperatures, S.M.A.R.T.
docker Docker/Podman — containers, images, networks
kubernetes Kubernetes — clusters, nodes, pods, deployments
truenas TrueNAS — system info, ZFS pools, datasets, snapshots, replication

Unified Navigation

The v6 UI organises pages by task instead of platform:

Page What it shows
Dashboard Overview panels aggregating all sources
Infrastructure All hosts: Proxmox nodes, Docker hosts, K8s nodes, TrueNAS systems, agent-only hosts
Workloads All workloads: VMs, LXC containers, Docker containers, Kubernetes pods
Storage All storage: Proxmox storage, PBS datastores, ZFS pools/datasets, Ceph
Recovery All backup/snapshot artifacts: PBS backups, PVE local dumps, ZFS snapshots, replication
Alerts Unified alert view across all platforms

Every page supports source filtering — click a source badge to see only resources from that platform.

Legacy Route Compatibility

Legacy URLs redirect automatically with toast notifications:

Legacy Route Redirects To
/proxmox/overview /infrastructure
/hosts /infrastructure?source=agent
/docker /workloads?source=docker
/kubernetes /infrastructure?source=kubernetes
/services, /mail /infrastructure?source=pmg

See Migration Guide for the full mapping.

API

Primary Endpoint

GET /api/resources

Returns all unified resources. Supports query parameters:

Parameter Description
type Filter by resource type (host, vm, container, etc.)
source Filter by data source (pve, docker, kubernetes, etc.)
status Filter by status (online, warning, critical, offline)
search Full-text search across name, ID, tags

Resource Details

Individual resource details are available via the unified state WebSocket connection, which pushes real-time updates to the frontend.

Frontend Architecture

The frontend uses SolidJS reactive selectors to derive views from the unified store:

  • useResources() — access the full unified resource list
  • useInfrastructureResources() — hosts filtered for the Infrastructure page
  • useWorkloadResources() — VMs/containers/pods for the Workloads page
  • useStorageResources() — storage pools/datasets for the Storage page
  • useRecoveryResources() — backup/snapshot data for the Recovery page

These selectors read from a single SolidJS store that is updated in real-time via WebSocket.

See Also