sascha.dodenhoeft/expertfab-infra
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Initial commit: Infrastructure documentation

Browse Source 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Sascha Dodenhöft
2026-05-03 13:46:47 +02:00
commit bbe86c55d9
9 changed files with 568 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

43
README.md Normal file
View File

@@ -0,0 +1,43 @@
# ExpertFab – Infrastruktur Dokumentation
Zentrale Dokumentation der ExpertFab IT-Infrastruktur.
## Inhalt
| Dokument | Beschreibung |
|---|---|
| [docs/infrastructure.md](docs/infrastructure.md) | VM-Inventar, Proxmox-Hosts, Netzwerk |
| [docs/access.md](docs/access.md) | SSH-Zugänge, API-Token, Credentials |
| [docs/k3s.md](docs/k3s.md) | K3s-Cluster, Namespaces, Ingresses, ERPNext |
| [docs/storage.md](docs/storage.md) | Longhorn Storage, PVCs, StorageClasses |
## Schaubilder
| Datei | Beschreibung |
|---|---|
| [diagrams/k8s_cluster.png](diagrams/k8s_cluster.png) | Gesamte Infrastruktur (Proxmox + K3s) |
| [diagrams/storage_architecture.png](diagrams/storage_architecture.png) | Longhorn Storage Architektur |
Schaubilder neu generieren:
```bash
cd diagrams
python3 k8s_cluster_diagram.py
python3 storage_diagram.py
```
## Schnellzugriff
```bash
# Proxmox
ssh -i ~/.ssh/hetzner_key root@10.42.70.1
# K3s Control Plane
ssh -i ~/.ssh/hetzner_key sd@10.42.71.50
# kubectl (am Control Plane)
sudo KUBECONFIG=/etc/rancher/k3s/k3s.yaml kubectl get pods -A
# kubectl via Proxmox (ohne sudo-Passwort)
ssh -i ~/.ssh/hetzner_key root@10.42.70.1 \
"qm guest exec 119 -- bash -c 'KUBECONFIG=/etc/rancher/k3s/k3s.yaml kubectl get pods -A'"
```

BIN
diagrams/k8s_cluster.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 410 KiB

97
diagrams/k8s_cluster_diagram.py Normal file
View File

@@ -0,0 +1,97 @@
from diagrams import Cluster, Diagram, Edge
from diagrams.k8s.network import Ingress
from diagrams.k8s.compute import Deployment, StatefulSet
from diagrams.k8s.storage import PVC, StorageClass
from diagrams.onprem.network import Opnsense
from diagrams.onprem.certificates import CertManager
from diagrams.onprem.proxmox import ProxmoxVE
from diagrams.onprem.vcs import Gitea
from diagrams.generic.os import Windows
graph_attr = {
"fontsize": "13",
"pad": "0.6",
"nodesep": "0.5",
"ranksep": "1.0",
"splines": "ortho",
}
with Diagram(
"ExpertFab Infrastruktur",
filename="k8s_cluster",
outformat="png",
show=False,
direction="TB",
graph_attr=graph_attr,
):
# ── Physische Ebene ───────────────────────────────────────────────────────
with Cluster("Proxmox Cluster efproxcl02 – 95.156.232.42"):
with Cluster("efproxcl02n01 – 64 vCPU / 128 GB"):
fw = Opnsense("efscfw01\nOPNsense / FW\n10.42.70.1")
n01_kctl = ProxmoxVE("efsckubctl\nK3s control plane")
n01_kn02 = ProxmoxVE("efsckubnode02\nK3s worker")
n01_smtp = ProxmoxVE("efsmtprelay")
n01_trade = ProxmoxVE("eftrade01")
n01_git = Gitea("efgit01 / Gitea\ngit.expertfab.de")
with Cluster("efproxcl02n02 – 64 vCPU / 128 GB"):
n02_kn1 = ProxmoxVE("efsckubnode1\nK3s worker")
n02_af = ProxmoxVE("efscairflow01\nAirflow")
n02_veeam = ProxmoxVE("efscveeam01\nVeeam Backup")
n02_dc = Windows("efscdc01\nDomain Controller")
n02_print = ProxmoxVE("efscprint01\nPrintserver")
n02_moni = ProxmoxVE("efscmoni01\nMonitoring")
# ── Logische K3s-Ebene ────────────────────────────────────────────────────
# K3s läuft auf: efsckubctl (control) + efsckubnode1 + efsckubnode02
with Cluster("K3s Cluster (efsckubctl · efsckubnode1 · efsckubnode02)"):
cert = CertManager("cert-manager\nLet's Encrypt")
traefik = Ingress("Traefik Ingress\n10.42.71.60")
cert >> traefik
with Cluster("erpnext – expertfab.de / www.expertfab.de"):
nginx = Deployment("Nginx")
gunicorn = Deployment("Gunicorn")
workers = Deployment("Workers\ndefault / short / long")
mariadb = StatefulSet("MariaDB 10.6")
df_cache = Deployment("DragonflyDB\ncache")
df_queue = Deployment("DragonflyDB\nqueue")
nginx >> gunicorn >> workers
[gunicorn, workers] >> mariadb
[gunicorn, workers] >> df_cache
workers >> df_queue
with Cluster("paperless – docs.expertfab.de"):
paperless = Deployment("Paperless-NGX")
with Cluster("zitadel – auth.expertfab.de"):
zitadel = Deployment("Zitadel")
with Cluster("rabbitmq – api.expertfab.de"):
fastapi = Deployment("FastAPI")
with Cluster("coworkbase – coworkbase.de"):
cowork = Deployment("Coworkbase")
with Cluster("qubicticker – qubicticker.qchief.io"):
ticker = Deployment("Qubicticker")
traefik >> [nginx, paperless, zitadel, fastapi, cowork, ticker]
with Cluster("Longhorn Storage"):
sc = StorageClass("longhorn")
pvc_mariadb = PVC("MariaDB 10 Gi RWO")
pvc_sites = PVC("Sites 10 Gi RWX")
pvc_logs = PVC("Logs 5 Gi RWX")
pvc_queue = PVC("Queue 2 Gi RWO")
sc >> [pvc_mariadb, pvc_sites, pvc_logs, pvc_queue]
mariadb >> Edge(style="dashed") >> pvc_mariadb
gunicorn >> Edge(style="dashed") >> pvc_sites
nginx >> Edge(style="dashed") >> pvc_logs
df_queue >> Edge(style="dashed") >> pvc_queue
# ── Verbindungen ──────────────────────────────────────────────────────────
fw >> traefik
n01_git >> Edge(label="image pull", style="dashed") >> nginx

BIN
diagrams/storage_architecture.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 196 KiB

82
diagrams/storage_diagram.py Normal file
View File

@@ -0,0 +1,82 @@
from diagrams import Cluster, Diagram, Edge
from diagrams.k8s.compute import Deployment, StatefulSet, Pod, DaemonSet, Job
from diagrams.k8s.storage import PVC, StorageClass
from diagrams.generic.storage import Storage
graph_attr = {
"fontsize": "13",
"pad": "0.8",
"nodesep": "0.6",
"ranksep": "1.2",
}
with Diagram(
"ExpertFab – Longhorn Storage Architektur",
filename="storage_architecture",
outformat="png",
show=False,
direction="TB",
graph_attr=graph_attr,
):
# ── Longhorn System Pods pro Node ─────────────────────────────────────────
with Cluster("K3s Worker Nodes (Longhorn System Pods)"):
with Cluster("efsckubnode1 (4 vCPU / 8 GB)"):
n1_mgr = Pod("longhorn-manager")
n1_csi = Pod("longhorn-csi-plugin")
n1_eng = Pod("engine-image")
n1_inst = Pod("instance-manager")
n1_driver = Deployment("driver-deployer")
n1_ui = Deployment("longhorn-ui (2×)")
n1_attacher = Deployment("csi-attacher (3×)")
n1_prov = Deployment("csi-provisioner (3×)")
n1_resizer = Deployment("csi-resizer (3×)")
n1_snap = Deployment("csi-snapshotter (3×)")
with Cluster("efsckubnode2 (4 vCPU / 8 GB)"):
n2_mgr = Pod("longhorn-manager")
n2_csi = Pod("longhorn-csi-plugin")
n2_eng = Pod("engine-image")
n2_inst = Pod("instance-manager")
n2_smgr1 = Pod("share-manager\n(erpnext RWX)")
n2_smgr2 = Pod("share-manager\n(erpnext-logs RWX)")
n2_backup = Job("daily-backup\n(CronJob)")
# ── StorageClasses ────────────────────────────────────────────────────────
with Cluster("StorageClasses (driver.longhorn.io)"):
sc_erpnext = StorageClass("longhorn-erpnext\nRetain / Immediate")
sc_paperless = StorageClass("longhorn-paperless\nRetain / Immediate")
sc_default = StorageClass("longhorn\nDelete / Immediate")
# ── PVCs pro Namespace ────────────────────────────────────────────────────
with Cluster("PVCs"):
with Cluster("namespace: erpnext"):
pvc_mariadb = PVC("data-erpnext-mariadb-sts-0\n3 Gi RWO")
pvc_sites = PVC("erpnext\n3 Gi RWX")
pvc_logs = PVC("erpnext-logs\n1 Gi RWX")
with Cluster("namespace: paperless"):
pvc_pl_media = PVC("paperless-media\n10 Gi RWO")
pvc_pl_consume = PVC("paperless-consume\n5 Gi RWO")
pvc_pl_data = PVC("paperless-data\n5 Gi RWO")
pvc_pl_pg = PVC("postgres-data\n5 Gi RWO")
with Cluster("namespace: rabbitmq"):
pvc_rmq = PVC("rabbitmq-data-rabbitmq-0\n5 Gi RWO")
with Cluster("namespace: zitadel"):
pvc_zit_pg = PVC("postgres-data-postgres-0\n10 Gi RWO")
# ── StorageClass → PVC ────────────────────────────────────────────────────
sc_erpnext >> [pvc_mariadb, pvc_sites, pvc_logs]
sc_paperless >> [pvc_pl_media, pvc_pl_consume, pvc_pl_data, pvc_pl_pg]
sc_default >> [pvc_rmq, pvc_zit_pg]
# ── Share-Manager bedient die RWX-Volumes ─────────────────────────────────
n2_smgr1 >> Edge(label="serves", style="dashed") >> pvc_sites
n2_smgr2 >> Edge(label="serves", style="dashed") >> pvc_logs
# ── Longhorn Manager koordiniert über beide Nodes ─────────────────────────
n1_mgr >> Edge(style="dotted", color="gray") >> n2_mgr

98
docs/access.md Normal file
View File

@@ -0,0 +1,98 @@
# ExpertFab – Zugangsdaten & Verbindungen
> **SSH-Key:** `~/.ssh/hetzner_key` (für alle Server)
---
## Proxmox
| Zugang | Wert |
|----------------|-------------------------------------------|
| Web-UI | https://95.156.232.42:8006 |
| SSH (Node 1) | `ssh -i ~/.ssh/hetzner_key root@10.42.70.1` |
| SSH (Node 2) | `ssh -i ~/.ssh/hetzner_key root@10.42.70.2` |
| Benutzer | `root` |
---
## K3s Cluster
| Zugang | Wert |
|---------------------|--------------------------------------------------------|
| SSH Control Plane | `ssh -i ~/.ssh/hetzner_key sd@10.42.71.50` |
| kubectl (am Node) | `sudo KUBECONFIG=/etc/rancher/k3s/k3s.yaml kubectl …` |
| kubectl via Proxmox | `ssh root@10.42.70.1 "qm guest exec 119 -- bash -c 'KUBECONFIG=/etc/rancher/k3s/k3s.yaml kubectl …'"` |
**K3s Version:** v1.34.6+k3s1
**Container Runtime:** containerd 2.2.2
**OS:** Ubuntu 24.04.4 LTS
### Nodes
| Node | IP | Rolle |
|---------------|--------------|----------------|
| efsckubadm | 10.42.71.50 | control-plane |
| efsckubnode1 | 10.42.71.51 | worker |
| efsckubnode2 | 10.42.71.52 | worker |
> Proxmox-VM `efsckubctl` = K3s-Node `efsckubadm`
---
## Paperless-NGX
| Zugang | Wert |
|-------------|----------------------------------------------|
| URL | https://docs.expertfab.de |
| API Token | `3960b56c7c56d21af06af7d32e49613d8e7f78c8` |
| API Header | `Authorization: Token <token>` |
---
## ERPNext
| Zugang | Wert |
|--------------|-------------------------------------------------------------------|
| URL | https://expertfab.de / https://www.expertfab.de |
| Helm Chart | frappe/erpnext 8.0.14 (ERPNext v15) |
| Namespace | `erpnext` |
| Auth Token | Airflow Variable: `ErpnextAuthToken` |
| Base URL | Airflow Variable: `ErpnextBaseurl` |
| Docker Image | git.expertfab.de/expertfab/customdocker:1.0.2 |
| Image Pull | Secret `gitea-registry` im Namespace `erpnext` |
---
## Gitea
| Zugang | Wert |
|--------|-----------------------------|
| URL | https://git.expertfab.de |
| VM | efgit01 (Proxmox n01 / 110) |
---
## SMTP (Accounting)
| Parameter | Airflow Variable |
|-----------|---------------------------|
| Server | `smtpAccountingServer` |
| Port | `smtpAccountingPort` |
| User | `smtpAccountingUser` |
| Password | `smtpAccountingPassword` |
| Sender | `smtpAccountingSenderName`|
---
## Öffentlich erreichbare Dienste
| Dienst | URL | Namespace |
|--------------|------------------------------|--------------|
| ERPNext | https://expertfab.de | erpnext |
| ERPNext | https://www.expertfab.de | erpnext |
| Paperless | https://docs.expertfab.de | paperless |
| Zitadel SSO | https://auth.expertfab.de | zitadel |
| FastAPI | https://api.expertfab.de | rabbitmq |
| Coworkbase | https://coworkbase.de | coworkbase |
| Qubicticker | https://qubicticker.qchief.io| qubicticker |
| Gitea | https://git.expertfab.de | – |

66
docs/infrastructure.md Normal file
View File

@@ -0,0 +1,66 @@
# ExpertFab – VM-Inventar
## Proxmox Cluster
| Hostname | URL / IP | Rolle |
|--------------------|-----------------------------------|-----------------------------|
| efproxcl02 | https://95.156.232.42:8006 | Proxmox Web-UI (Cluster) |
| efproxcl02n01 | 10.42.70.1 | Proxmox Host Node 1 |
| efproxcl02n02 | 10.42.70.2 | Proxmox Host Node 2 |
**Specs je Node:** 64 vCPU / 128 GB RAM
**DNS:** efproxcl02n01.sc.expertfab.de / efproxcl02n02.sc.expertfab.de
---
## VMs – efproxcl02n01
| VM-ID | Name | Status | CPU | RAM | IP | Rolle |
|-------|-----------------|---------|--------|-------|-----------------|------------------------------|
| 100 | efscfw01 | stopped | – | 16 GB | – | OPNsense Backup (inaktiv) |
| 101 | efsmtprelay | running | 4 vCPU | 16 GB | 109.230.227.36 | SMTP Relay |
| 102 | efscweb01 | stopped | – | 16 GB | – | Webserver (inaktiv) |
| 105 | efcacert | running | – | 4 GB | – | CA / Zertifikatsserver |
| 110 | efgit01 | running | – | 8 GB | – | Gitea (git.expertfab.de) |
| 111 | eftrade01 | running | 16vCPU | 16 GB | 10.42.71.102 | Trading-VM |
| 112 | efbtcpay | stopped | – | 8 GB | – | BTCPay Server (inaktiv) |
| 114 | efubutemp | stopped | – | 4 GB | – | Ubuntu Template |
| 115 | efxmr01 | stopped | – | 16 GB | – | Monero Node (inaktiv) |
| 116 | efscbank | stopped | – | 4 GB | – | Bank-VM (inaktiv) |
| 117 | efmatrix01 | running | – | 4 GB | – | Matrix Chat Server |
| 118 | efubu24lts* | – | – | 4 GB | – | Template (Ubuntu 24.04 LTS) |
| 119 | efsckubctl | running | 4 vCPU | 8 GB | 10.42.71.50 | K3s Control Plane |
| 121 | efsckubnode02 | running | 4 vCPU | 8 GB | 10.42.71.52 | K3s Worker Node 2 |
| 150 | efscfw01 | running | 8 vCPU | 16 GB | 109.230.227.38 | OPNsense Firewall (aktiv) |
*Template
## VMs – efproxcl02n02
| VM-ID | Name | Status | CPU | RAM | IP | Rolle |
|-------|-----------------|---------|---------|-------|-----------------|------------------------------|
| 103 | efscdc01 | running | 8 vCPU | 32 GB | 10.42.71.15 | Domain Controller (Windows) |
| 104 | efscveeam01 | running | 16 vCPU | 16 GB | 10.42.71.16 | Veeam Backup Server |
| 106 | efscprint01 | running | – | 8 GB | – | Printserver |
| 107 | eferp01 | stopped | – | 4 GB | – | Alt-ERP (inaktiv) |
| 108 | efscairflow01 | running | – | 8 GB | – | Apache Airflow |
| 109 | efscmoni01 | running | – | 4 GB | – | Monitoring |
| 113 | efbookstack01 | running | – | 8 GB | – | BookStack Wiki |
| 120 | efsckubnode1 | running | 4 vCPU | 8 GB | 10.42.71.51 | K3s Worker Node 1 |
| 122 | efscNffsBackup | running | – | – | – | NFS Backup |
---
## Netzwerk
| Netz | Bereich | Verwendung |
|--------------|-----------------|-----------------------------------|
| Public | 109.230.227.x | Öffentliche IPs (Hetzner) |
| Intern | 10.42.70.x | Proxmox Hosts / Management |
| Intern | 10.42.71.x | VMs / Server |
| K3s Pod-CIDR | 172.16.0.0/16 | Kubernetes Pod-Netzwerk (Flannel) |
| K3s SVC-CIDR | 10.43.0.0/16 | Kubernetes Services |
**Traefik LoadBalancer IP:** 10.42.71.60
**OPNsense** löst `www.expertfab.de` → `10.42.71.60` (intern), `expertfab.de` → öffentliche IP
→ Wichtig: `host_name` in ERPNext muss `https://www.expertfab.de` sein (Hairpin-NAT-Fix)

96
docs/k3s.md Normal file
View File

@@ -0,0 +1,96 @@
# ExpertFab – K3s Cluster
**Schaubild:** [../diagrams/k8s_cluster.png](../diagrams/k8s_cluster.png)
---
## Cluster-Übersicht
| Eigenschaft | Wert |
|-------------------|----------------------------------|
| Distribution | K3s v1.34.6+k3s1 |
| Container Runtime | containerd 2.2.2 |
| OS | Ubuntu 24.04.4 LTS |
| Kernel | 6.8.0-107-generic |
| Ingress | Traefik (websecure / TLS) |
| TLS | cert-manager + Let's Encrypt |
| Storage | Longhorn (verteiltes Block-Storage)|
| CNI | Flannel |
| LB | MetalLB → 10.42.71.60 |
---
## Nodes
| Name | IP | Proxmox-VM | Proxmox-Host | Rolle |
|---------------|--------------|---------------|---------------|---------------|
| efsckubadm | 10.42.71.50 | efsckubctl | efproxcl02n01 | control-plane |
| efsckubnode1 | 10.42.71.51 | efsckubnode1 | efproxcl02n02 | worker |
| efsckubnode2 | 10.42.71.52 | efsckubnode02 | efproxcl02n01 | worker |
---
## Namespaces
| Namespace | Inhalt |
|------------------|-----------------------------------------|
| `erpnext` | ERPNext + MariaDB + DragonflyDB |
| `paperless` | Paperless-NGX + PostgreSQL |
| `zitadel` | Zitadel SSO + PostgreSQL |
| `rabbitmq` | RabbitMQ + FastAPI |
| `coworkbase` | Coworkbase |
| `qubicticker` | Qubic Ticker |
| `longhorn-system`| Longhorn Storage |
| `cert-manager` | cert-manager (Let's Encrypt) |
| `metallb-system` | MetalLB LoadBalancer |
| `kube-system` | Traefik, CoreDNS, etc. |
---
## Ingresses (Traefik)
| Host | Namespace | TLS |
|-----------------------------|--------------|-----|
| expertfab.de | erpnext | ✓ |
| www.expertfab.de | erpnext | ✓ |
| docs.expertfab.de | paperless | ✓ |
| auth.expertfab.de | zitadel | ✓ |
| api.expertfab.de | rabbitmq | ✓ |
| coworkbase.de | coworkbase | ✓ |
| www.coworkbase.de | coworkbase | ✓ |
| qubicticker.qchief.io | qubicticker | ✓ |
**ClusterIssuer:** `letsencrypt`
**TLS Secret (ERPNext):** `expertfab-tls`
---
## ERPNext Deployment
| Komponente | Typ | Details |
|--------------------|------------|--------------------------------------------|
| Nginx | Deployment | Frontend, `frappeSiteNameHeader=expertfab.de` |
| Gunicorn | Deployment | Web Workers |
| Worker default | Deployment | Background Jobs |
| Worker short | Deployment | Kurze Jobs |
| Worker long | Deployment | Lange Jobs |
| MariaDB | StatefulSet| v10.6, PVC: 3Gi RWO |
| DragonflyDB cache | Deployment | Redis-kompatibel |
| DragonflyDB queue | Deployment | Redis-kompatibel, PVC: 2Gi RWO (Anmerkung¹)|
> ¹ Helm-Values definieren 2Gi für Queue, tatsächliche PVC zeigt 3Gi Sites-Volume als RWX
**Helm Chart:** frappe/erpnext 8.0.14
**Custom Image:** git.expertfab.de/expertfab/customdocker:1.0.2
**Apps:** erpnext, hrms, payments, webshop, ecommerce_integrations, efrevolutgateway
---
## DNS-Besonderheit (Hairpin-NAT)
OPNsense löst nur `www.expertfab.de` → `10.42.71.60` (intern) auf.
`expertfab.de` (ohne www) → öffentliche IP → von innen nicht erreichbar.
**Fix:** Frappe `host_name` = `https://www.expertfab.de`
**Pfad:** `/home/frappe/frappe-bench/sites/expertfab.de/site_config.json`
**Betrifft:** wkhtmltopdf PDF-Generierung (hängt sonst 120s → 504)

86
docs/storage.md Normal file
View File

@@ -0,0 +1,86 @@
# ExpertFab – Longhorn Storage Architektur
**Schaubild:** [../diagrams/storage_architecture.png](../diagrams/storage_architecture.png)
---
## StorageClasses
| Name | Reclaim | Binding | Verwendung |
|---------------------|---------|-----------|--------------------------|
| `longhorn` | Delete | Immediate | RabbitMQ, Zitadel |
| `longhorn-erpnext` | Retain | Immediate | ERPNext (Daten bleiben!) |
| `longhorn-paperless`| Retain | Immediate | Paperless (Daten bleiben!)|
| `longhorn-static` | Delete | Immediate | Manuell provisionierte Volumes |
| `local-path` | Delete | WaitForFirstConsumer | Rancher local-path |
> **Retain** = PV bleibt erhalten wenn PVC gelöscht wird → Schutz vor Datenverlust
---
## PVCs nach Namespace
### namespace: erpnext
| PVC | Größe | Mode | StorageClass |
|----------------------------|-------|------|---------------------|
| data-erpnext-mariadb-sts-0 | 3 Gi | RWO | longhorn-erpnext |
| erpnext | 3 Gi | RWX | longhorn-erpnext |
| erpnext-logs | 1 Gi | RWX | longhorn-erpnext |
### namespace: paperless
| PVC | Größe | Mode | StorageClass |
|--------------------|-------|------|----------------------|
| paperless-media | 10 Gi | RWO | longhorn-paperless |
| paperless-consume | 5 Gi | RWO | longhorn-paperless |
| paperless-data | 5 Gi | RWO | longhorn-paperless |
| postgres-data | 5 Gi | RWO | longhorn-paperless |
### namespace: rabbitmq
| PVC | Größe | Mode | StorageClass |
|--------------------------|-------|------|--------------|
| rabbitmq-data-rabbitmq-0 | 5 Gi | RWO | longhorn |
### namespace: zitadel
| PVC | Größe | Mode | StorageClass |
|-------------------------|-------|------|--------------|
| postgres-data-postgres-0| 10 Gi | RWO | longhorn |
---
## Longhorn Pods nach Node
### efsckubnode1 (10.42.71.51)
| Pod | Typ | Replicas |
|---------------------------|------------|----------|
| longhorn-manager | DaemonSet | 1 |
| longhorn-csi-plugin | DaemonSet | 1 |
| engine-image | DaemonSet | 1 |
| instance-manager | – | 1 |
| longhorn-driver-deployer | Deployment | 1 |
| longhorn-ui | Deployment | 2 |
| csi-attacher | Deployment | 3 |
| csi-provisioner | Deployment | 3 |
| csi-resizer | Deployment | 3 |
| csi-snapshotter | Deployment | 3 |
### efsckubnode2 (10.42.71.52)
| Pod | Typ | Beschreibung |
|-----------------------------|-----------|-------------------------------------|
| longhorn-manager | DaemonSet | – |
| longhorn-csi-plugin | DaemonSet | – |
| engine-image | DaemonSet | – |
| instance-manager | – | – |
| share-manager (erpnext) | – | Bedient RWX-Volume `erpnext` (3Gi) |
| share-manager (erpnext-logs)| – | Bedient RWX-Volume `erpnext-logs` (1Gi) |
| daily-backup | CronJob | Tägliches Backup |
> **share-manager** Pods werden für RWX-Volumes benötigt: Longhorn stellt RWX über NFS-Share-Manager bereit.
> CSI-Controller-Pods (attacher, provisioner, resizer, snapshotter) laufen nur auf `efsckubnode1`.
---
## Hinweise
- Longhorn UI: erreichbar über Port-Forward `kubectl port-forward -n longhorn-system svc/longhorn-frontend 8080:80`
- Replikation: Standard 2 Replicas (beide Worker-Nodes)
- Backups: `daily-backup` CronJob auf efsckubnode2