Differences

This shows you the differences between two versions of the page.

--- torresvault:architecture:overview [2026/01/23 14:00] – nathna
+++ torresvault:architecture:overview [2026/01/23 14:03] (current) – nathna
@@ Line 142: / Line 142: @@
 ----
 ==== 3. Proxmox Cluster Architecture ====
@@ Line 149: / Line 150: @@
 === PVE1 ===
-  * CPU: **TBD**  (fill from `lscpu`)
+  * Hostname: **pve1**
-  * RAM: **TBD**  (fill from `free -h`)
+  * CPU: **Intel Core i5-2500 @ 3.30 GHz (4 cores / 4 threads, 1 socket)**
-  * Disks: **~12–14 × 1TB 2.5" drives** (exact list from `lsblk`)
+  * RAM: **32 GB DDR3L 1600 MHz**
-  * HBA: **TBD** (from `lspci`)
+    * 4 × 8 GB Timetec DDR3L (PC3L-12800) UDIMMs
+  * Disks (approximate):
+    * Multiple **1 TB WDC WD1003FBYX** enterprise HDDs
+    * Multiple **1 TB Seagate ST91000640NS** drives
+    * Total of ~12 × 1 TB disks for VM storage
+  * Storage stack:
+    * System disk on onboard Intel SATA controller
+    * Data disks on GLOTRENDS SATA card, grouped into Proxmox storage (LVM/ZFS + zvols)
+  * HBAs / SATA:
+    * Onboard **Intel SATA controller (RAID mode)**
+    * **ASMedia ASM1064 SATA controller**
+    * **GLOTRENDS SA3112-C 12-Port PCIe x1 SATA Expansion Card**
+  * Networking:
+    * Onboard **Intel 82579LM Gigabit NIC**
+    * **Intel I350 quad-port 1 GbE** PCIe NIC
+    * vmbr interfaces used for:
+      * LAN / management
+      * Cluster interconnect (10.10.10.0/30 link to PVE2)
   * Roles:
-    * Core VMs (Nextcloud, NPM, Jellyfin, Prometheus/Grafana, etc.)
+    * Hosts many of the core VMs (Nextcloud, NPM, Jellyfin, Prometheus/Grafana, etc.)
-    * Part of Proxmox cluster storage
+    * Part of 2-node Proxmox cluster
 === PVE2 ===
-  * CPU: **TBD**
+  * Hostname: **pve2**
-  * RAM: **TBD**
+  * CPU: **Intel Core i5-4570 @ 3.20 GHz (4 cores / 4 threads, 1 socket)**
-  * Disks: **~12–14 × 1TB 2.5" drives**
+  * RAM: **32 GB DDR3L 1600 MHz**
-  * HBA: **TBD**
+    * Same Timetec 4 × 8 GB kit as PVE1
+  * Disks (approximate):
+    * Multiple **1 TB Seagate ST91000640NS** drives
+    * Total of ~12 × 1 TB disks for VM storage
+  * Storage stack:
+    * System disk on onboard Intel 9-Series SATA controller (AHCI)
+    * Data disks on GLOTRENDS SATA card
+  * HBAs / SATA:
+    * **Intel 9 Series SATA controller (AHCI mode)**
+    * **ASMedia ASM1064 SATA controller**
+    * **GLOTRENDS SA3112-C 12-Port PCIe x1 SATA Expansion Card**
+  * Networking:
+    * Same **Intel I350 quad-port 1 GbE** NIC family as PVE1 (4 ports)
+    * Bridges mirror PVE1 layout for easy VM migration
   * Roles:
-    * Redundancy for critical services
+    * Redundant node for critical services
-    * Test / lab workloads
+    * General lab workloads and testing
 === QDevice ===
-  * Hardware: Raspberry Pi
+  * Hardware: **Raspberry Pi**
-  * Purpose: `corosync-qdevice` to avoid split-brain in 2-node cluster
+  * Purpose: runs **corosync-qdevice** to provide quorum for the 2-node Proxmox cluster
+  * Goal: avoid split-brain if one Proxmox node goes offline
-=== Future: NAS / Proxmox Hybrid ===
+=== Future: NAS / Proxmox Hybrid (Define 7 XL) ===
-  * Case: Fractal Design Define 7 XL
+  * Case: **Fractal Design Define 7 XL**
-  * Motherboard: ASRock Rack X570D4U
+  * Motherboard: **ASRock Rack X570D4U**
-  * CPU: Ryzen (exact model TBD)
+  * CPU: **Ryzen (exact model TBD)**
-  * RAM: TBD
+  * RAM: **TBD (planned upgrade path from 32 GB → higher)**
-  * Disks: 16 × 6TB SAS enterprise drives via dual HBAs
+  * Disks: **16 × 6 TB SAS enterprise drives** via dual HBAs
   * Network:
     * 1 × dedicated management port
-    * 2 × 1GbE
+    * 2 × 1 GbE
-    * 2 × 10GbE (Intel X550)
+    * 2 × 10 GbE (Intel X550)
   * Role:
-    * High-capacity NAS
+    * High-capacity NAS for the cluster
-    * Proxmox node for storage-heavy workloads
+    * Additional Proxmox node for storage-heavy workloads
-    * “Set it and forget it” anchor of TorresVault 2.0
+    * Long-term “set it and forget it” anchor of **TorresVault 2.0**
+=== Future: Services / Mini-PC Cluster ===
-=== Future: Services Cluster (Mini PCs) ===
+  * **2 × MINISFORUM UM890 Pro** mini PCs
+  * Planned roles:
+    * Lightweight Kubernetes / k3s or Docker swarm node(s)
+    * Local AI / automation services
+    * Offload non-critical or experimental workloads from PVE1/PVE2
-  * 2 × MINISFORUM UM890 Pro mini PCs
-  * Planned use:
-    * k3s or Docker-based service cluster
-    * Local AI / automation workloads
-    * Possibly offload lighter services from main Proxmox nodes
 ----