Redesign mermaid test with consistent style guidelines

- Define 5-color semantic palette (Okabe-Ito derived, color-blind safe)
- Use classDef for reusable styles instead of per-node statements
- Limit node text to 2 lines max, move specs to tables
- Consistent node shapes by role (rounded for network, stadium for external)
- Dashed arrows for control plane, solid for data plane
- Tinted subgraph fills matching dominant role color

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Maarten 2026-02-24 17:17:39 +01:00
parent 07c289d75c
commit 6293f7ab79
1 changed files with 166 additions and 119 deletions

View File

@ -5,136 +5,160 @@ Delete this file after confirming.
--- ---
## Test 1: HAProxy Architecture (flowchart, top-down) ## Style Guidelines
Original: `notes/haproxy-guide.md` lines 13-44 All diagrams in this project follow a consistent style based on semantic
color-coding by element role, color-blind-safe colors (Okabe-Ito derived),
and minimal text per node.
**Color palette** (5 semantic colors + 1 neutral):
| Role | Color | Hex | Node shape |
|------|-------|-----|------------|
| Cluster node / VM | Teal | `#009E73` | Rectangle `["..."]` |
| Instance / container | Sky blue | `#56B4E9` | Rectangle `["..."]` |
| Network infrastructure | Blue | `#0072B2` | Rounded `("...")` |
| Load balancer / proxy | Amber | `#E69F00` | Rectangle `["..."]` |
| Management / control | Mauve | `#CC79A7` | Rectangle `["..."]` |
| External / entry point | Light gray | `#f5f5f5` | Stadium `(["..."])` |
**Subgraph fills**: light tint of the dominant role color, darker stroke.
**Rules**:
- Max 2 lines of text per node (name + one detail). Move specs to tables.
- Top-down (`TD`) for hierarchies; left-right (`LR`) for lateral/peer relationships.
- Solid arrows (`-->`) for data/traffic; dashed (`-.->`) for control/management.
- Use `classDef` for styling, never per-node `style` statements.
- No emojis or unicode symbols in labels (renderer compatibility).
- Edge labels: 1-3 words max, only when relationship isn't obvious.
---
## Test 1: HAProxy Architecture
Original: `notes/haproxy-guide.md`
```mermaid ```mermaid
flowchart TD flowchart TD
client["Client / LAN Traffic"] client(["Client / LAN"])
vip["UPLINK VIP Address\n192.168.103.200"] vip["VIP 192.168.103.200"]
ovnlb["OVN Load Balancer\n(distributes to HA pair\nfor failover)"] ovnlb("OVN Load Balancer")
ha1["HAProxy 01\n10.10.10.50\n(container)"] ha1["HAProxy 01\n10.10.10.50"]
ha2["HAProxy 02\n10.10.10.51\n(container)"] ha2["HAProxy 02\n10.10.10.51"]
ng1["nginx-01\n.60 :80"] ng1["nginx-01 · .60"]
ng2["nginx-02\n.61 :80"] ng2["nginx-02 · .61"]
ng3["nginx-03\n.62 :80"] ng3["nginx-03 · .62"]
client --> vip --> ovnlb client --> vip
ovnlb --> ha1 vip --> ovnlb
ovnlb --> ha2 ovnlb --> ha1 & ha2
ha1 --> ng1 ha1 & ha2 --> ng1 & ng2 & ng3
ha1 --> ng2
ha1 --> ng3
ha2 --> ng1
ha2 --> ng2
ha2 --> ng3
style vip fill:#4a90d9,color:#fff classDef external fill:#f5f5f5,color:#333,stroke:#999
style ovnlb fill:#7b68ee,color:#fff classDef network fill:#0072B2,color:#fff,stroke:#005a8e
style ha1 fill:#2ecc71,color:#fff classDef lb fill:#E69F00,color:#fff,stroke:#b87d00
style ha2 fill:#2ecc71,color:#fff classDef instance fill:#56B4E9,color:#fff,stroke:#3a8fbf
style ng1 fill:#e67e22,color:#fff
style ng2 fill:#e67e22,color:#fff class client external
style ng3 fill:#e67e22,color:#fff class vip,ovnlb network
class ha1,ha2 lb
class ng1,ng2,ng3 instance
``` ```
--- ---
## Test 2: Bridge Topology (subgraphs for nodes) ## Test 2: Bridge Topology
Original: `notes/networking-guide.md` lines 81-100 Original: `notes/networking-guide.md`
```mermaid ```mermaid
flowchart TD flowchart TD
subgraph node1["Node 1 (net-node-01)"] subgraph node1["Node 1 · net-node-01"]
br1["incusbr0\n10.0.0.1/24"] c1["c1 · .202"] & c2["c2 · .40"] --- br1("incusbr0\n10.0.0.1/24")
c1["c1\n.202"] br1 --> nat1["NAT · 192.168.1.209"]
c2["c2\n.40"]
nat1["NAT → ens18\n192.168.1.209"]
c1 --- br1
c2 --- br1
br1 --> nat1
end end
subgraph node2["Node 2 (net-node-02)"] subgraph node2["Node 2 · net-node-02"]
br2["incusbr0\n10.0.0.1/24"] c3["c3 · .52"] --- br2("incusbr0\n10.0.0.1/24")
c3["c3\n.52"] br2 --> nat2["NAT · 192.168.1.150"]
nat2["NAT → ens18\n192.168.1.150"]
c3 --- br2
br2 --> nat2
end end
nat1 --- lan nat1 & nat2 --- lan(("LAN\n192.168.1.0/24"))
nat2 --- lan
lan(("LAN\n192.168.1.0/24"))
node1 ~~~ node2 classDef instance fill:#56B4E9,color:#fff,stroke:#3a8fbf
classDef network fill:#0072B2,color:#fff,stroke:#005a8e
classDef node fill:#009E73,color:#fff,stroke:#007a5e
linkStyle 5 stroke:red,stroke-dasharray:5 class c1,c2,c3 instance
linkStyle 6 stroke:red,stroke-dasharray:5 class br1,br2,lan network
class nat1,nat2 node
style node1 fill:#e6f5f0,stroke:#009E73
style node2 fill:#e6f5f0,stroke:#009E73
``` ```
> **Key point:** Each node has its own bridge with the same subnet. The Each node has its own bridge with the same subnet (10.0.0.1/24).
> bridges are NOT connected to each other. The bridges are **not** connected to each other — cross-node traffic fails.
--- ---
## Test 3: OVN Topology (control plane + nodes + tunnels) ## Test 3: OVN Topology
Original: `notes/networking-guide.md` lines 196-227 Original: `notes/networking-guide.md`
```mermaid ```mermaid
flowchart TD flowchart TD
subgraph cp["OVN Control Plane"] subgraph cp["OVN Control Plane"]
ovnc["ovn-central\novn-northd + ovsdb-server\nNB: tcp:192.168.1.209:6641\nSB: tcp:192.168.1.209:6642"] ovnc["ovn-central\nNB :6641 · SB :6642"]
end end
subgraph n1["Node 1 (net-node-01)"] subgraph n1["Node 1 · net-node-01"]
ctrl1["ovn-controller\novs-vswitchd"] ctrl1["ovn-controller"] ~~~ ls1
oc1["c1 (.2)"] c1["c1 · .2"] & c2["c2 · .3"] --- ls1("logical switch\n10.10.10.0/24")
oc2["c2 (.3)"]
ls1["OVN logical switch\n10.10.10.0/24"]
oc1 --- ls1
oc2 --- ls1
ctrl1 ~~~ ls1
end end
subgraph n2["Node 2 (net-node-02)"] subgraph n2["Node 2 · net-node-02"]
ctrl2["ovn-controller\novs-vswitchd"] ctrl2["ovn-controller"] ~~~ ls2
oc3["c3 (.4)"] c3["c3 · .4"] --- ls2("logical switch\n10.10.10.0/24")
ls2["OVN logical switch\n10.10.10.0/24"]
oc3 --- ls2
ctrl2 ~~~ ls2
end end
subgraph n3["Node 3 (net-node-03)"] subgraph n3["Node 3 · net-node-03"]
ctrl3["ovn-controller\novs-vswitchd"] ctrl3["ovn-controller"] ~~~ ls3
oc4["c4 (.5)"] c4["c4 · .5"] --- ls3("logical switch\n10.10.10.0/24")
ls3["OVN logical switch\n10.10.10.0/24"]
oc4 --- ls3
ctrl3 ~~~ ls3
end end
ovnc -->|proxy devices| ctrl1 ovnc -.-> ctrl1 & ctrl2 & ctrl3
ovnc -->|proxy devices| ctrl2
ovnc -->|proxy devices| ctrl3
ls1 <-->|Geneve tunnel\n192.168.1.209| ls2 ls1 <-->|Geneve| ls2
ls2 <-->|Geneve tunnel\n192.168.1.150| ls3 ls2 <-->|Geneve| ls3
ls1 <-->|Geneve tunnel\n192.168.1.13| ls3 ls1 <-->|Geneve| ls3
lan(("LAN\n192.168.1.0/24")) n1 & n2 & n3 --- lan(("LAN\n192.168.1.0/24"))
n1 --- lan
n2 --- lan classDef mgmt fill:#CC79A7,color:#fff,stroke:#a36088
n3 --- lan classDef instance fill:#56B4E9,color:#fff,stroke:#3a8fbf
classDef network fill:#0072B2,color:#fff,stroke:#005a8e
classDef node fill:#009E73,color:#fff,stroke:#007a5e
class ovnc mgmt
class ctrl1,ctrl2,ctrl3 node
class c1,c2,c3,c4 instance
class ls1,ls2,ls3,lan network
style cp fill:#f5e6f0,stroke:#CC79A7
style n1 fill:#e6f5f0,stroke:#009E73
style n2 fill:#e6f5f0,stroke:#009E73
style n3 fill:#e6f5f0,stroke:#009E73
``` ```
All instances share a single logical switch connected via Geneve tunnels.
The control plane (dashed lines) manages the data plane (solid lines).
--- ---
## Test 4: Geneve Tunnel Mesh ## Test 4: Geneve Tunnel Mesh
Original: `notes/ovn-deep-dive.md` lines 331-349 Original: `notes/ovn-deep-dive.md`
```mermaid ```mermaid
graph LR graph LR
@ -142,69 +166,92 @@ graph LR
n2(("oc-node-02\n.141")) n2(("oc-node-02\n.141"))
n3(("oc-node-03\n.142")) n3(("oc-node-03\n.142"))
n1 <-->|"Geneve\nUDP 6081\nBFD ✓"| n2 n1 <-->|"Geneve 6081"| n2
n2 <-->|"Geneve\nUDP 6081\nBFD ✓"| n3 n2 <-->|"Geneve 6081"| n3
n1 <-->|"Geneve\nUDP 6081\nBFD ✓"| n3 n1 <-->|"Geneve 6081"| n3
style n1 fill:#3498db,color:#fff classDef chassis fill:#009E73,color:#fff,stroke:#007a5e
style n2 fill:#2ecc71,color:#fff
style n3 fill:#e74c3c,color:#fff class n1,n2,n3 chassis
``` ```
Full mesh — every pair has a Geneve tunnel (UDP 6081) with BFD health
monitoring. Tunnel keys are set per-packet from the OVN datapath.
--- ---
## Test 5: OVS Bridge Architecture (per-node) ## Test 5: OVS Bridge Architecture
Original: `notes/ovn-deep-dive.md` lines 305-313 Original: `notes/ovn-deep-dive.md`
```mermaid ```mermaid
flowchart LR flowchart LR
subgraph provider["incusovn7 (provider bridge)"] subgraph provider["incusovn7 · provider bridge"]
nic["incusovn7 ← physical NIC"] nic["physical NIC"]
intport["incusovn7b ← internal port"] intport["internal port"]
patch1["patch-...-to-br-int"] patch1["patch to br-int"]
end end
subgraph integration["br-int (integration bridge)"] subgraph integration["br-int · integration bridge"]
veth["veth* ← instance NICs"] veth["instance veth ports"]
ovntun["ovn-* ← Geneve tunnels"] ovntun["Geneve tunnels"]
patch2["patch-br-int-to-..."] patch2["patch to incusovn7"]
brint["br-int ← internal port"]
end end
patch1 <-->|"patch port"| patch2 patch1 <-->|"patch port"| patch2
style provider fill:#e8f4fd,stroke:#3498db classDef prov fill:#0072B2,color:#fff,stroke:#005a8e
style integration fill:#eafaf1,stroke:#2ecc71 classDef integ fill:#009E73,color:#fff,stroke:#007a5e
class nic,intport,patch1 prov
class veth,ovntun,patch2 integ
style provider fill:#e0eef8,stroke:#0072B2
style integration fill:#e6f5f0,stroke:#009E73
``` ```
Each IncusOS node runs two OVS bridges. The provider bridge is the
on-ramp to the physical network; the integration bridge handles all
OVN logical processing (ACLs, NAT, LB, routing).
--- ---
## Test 6: Operations Center Architecture ## Test 6: Operations Center Architecture
Original: `notes/operations-center-guide.md` lines 12-45 Original: `notes/operations-center-guide.md`
```mermaid ```mermaid
flowchart TD flowchart TD
subgraph proxmox["Proxmox VE Host (pve)\nIntel i9-13900HK · 64 GiB · 20 cores"] subgraph proxmox["Proxmox VE Host · i9-13900HK · 64 GiB"]
subgraph cluster["Incus Cluster (OVN Geneve · net-prod 10.10.10.0/24)"] subgraph cluster["Incus Cluster · net-prod 10.10.10.0/24"]
n1["oc-node-01\nVMID 400\n4c/8G/64G\n.140\novn-central"] n1["oc-node-01\nVMID 400 · .140"]
n2["oc-node-02\nVMID 401\n4c/8G/50G\n.141"] n2["oc-node-02\nVMID 401 · .141"]
n3["oc-node-03\nVMID 402\n4c/8G/50G\n.142"] n3["oc-node-03\nVMID 402 · .142"]
end end
oc["oc-server\nVMID 920 · 2c/4G/50G\n.120\nhttps://192.168.102.120\nWeb UI + CLI + REST API"] oc["Operations Center\nVMID 920 · .120"]
end end
vmbr0["vmbr0 (VLAN 69)\n192.168.100.0/22\nGateway: 192.168.100.1"] vmbr0(("VLAN 69\n192.168.100.0/22"))
ext["OVN External IPs\n192.168.103.200-210\nLB: .201 · Fwd: .202"] ext["OVN external IPs\n192.168.103.200-210"]
proxmox --- vmbr0 proxmox --- vmbr0
cluster -.->|"OVN external"| ext cluster -.->|"external gateway"| ext
style proxmox fill:#f5f5f5,stroke:#333 classDef node fill:#009E73,color:#fff,stroke:#007a5e
style cluster fill:#e8f4fd,stroke:#3498db classDef mgmt fill:#CC79A7,color:#fff,stroke:#a36088
style oc fill:#fff3cd,stroke:#e67e22 classDef network fill:#0072B2,color:#fff,stroke:#005a8e
style n1 fill:#d4edda,stroke:#28a745
style n2 fill:#d4edda,stroke:#28a745 class n1,n2,n3 node
style n3 fill:#d4edda,stroke:#28a745 class oc mgmt
class vmbr0,ext network
style proxmox fill:#f5f5f5,stroke:#999
style cluster fill:#e6f5f0,stroke:#009E73
``` ```
| Component | VMID | IP | Specs | Role |
|-----------|------|----|-------|------|
| oc-node-01 | 400 | .140 | 4c/8G/64G | Cluster init + OVN central |
| oc-node-02 | 401 | .141 | 4c/8G/50G | Cluster member |
| oc-node-03 | 402 | .142 | 4c/8G/50G | Cluster member |
| oc-server | 920 | .120 | 2c/4G/50G | Operations Center |