Hey Homelabbers,
I am looking to rebuild my Jonsbo N2 NAS with a couple minor improvements. The main issue I’m facing is a slowly increasing ATA error count for a system that was built less than 2 years ago. I’m suspicious of the SATA port expansion card I’m using as the cause, but a little out of my depths in trying to understand the failures. I also have a decommissioned Hackintosh PC which has a GPU I’m considering integrating into the rebuild.
While rebuilding, I’d like to improve airflow in the entire case to help drive temps. Additionally wouldn’t mind incorporating the graphics card from my decommissioned Hackintosh for improved overhead of the Jellyfin instance to expand the user base to a couple more family members, or buy a graphics card that will improve upon the built in graphics of the CPU without being overly expensive. I also would like to consider virtualizing or migrating the services I run on a Dell Optiplex Micro i5 w/ 16gb ram which runs Home Assistant and a handful of LXCs and VMs into the same server, as the server itself is overbuilt for its current usage and the Optiplex occasionally grounds to a halt when I give it a large influx or data to process -though overall, it might be better off remaining on a separate device for the sake of separation.
Services I run on the NAS: Jellyfin, kasm, pyload-ng, sabnzbd, and an instance of ollama that doesn’t get much love.
Services/VMs I run through Proxmox on the dinky little Optiplex: Home Assistant w/ Frigate and a Coral for image recognition ingesting streams from 4 PoE cameras (ideally part or all of this moves to the NAS’s NVME as the cameras will be constantly writing), *arrs x2 (shared instance, and a non-shared 4k), Immich, Paperless, AdGuard, Vaultwarden, ChangeDetection, Hoarder, Jellyseerr, 2 super small alpine VMs, and a handful of other services not worth mentioning.
Questions for the homelabbers out there:
- My cursory understanding is ATA errors are an early sign of drive failure, but in my case I’m suspicious of the cables and expansion connection 2 of the drives are connected to vs the actual drives. What should I do about the slowly rising error count…? Should I replace drives after fixing the underlying issue…?
- Should I use a different case for better airflow? Ideally it would fit within 17” x 14” x 12” to fit in my rack -I don’t really have anywhere else I can put it where the noise won’t drive someone in the house crazy.
- Are there any better mobos now that support AM4, and my need for 7 drives without the use of a SATA expansion card while being a small-ish board?
The NAS has:
| Part | Model |
|
|
|
| OS | TrueNAS Scale |
| Case | JONSBO N2 |
| CPU | Ryzen Pro 5650GE |
| CPU Cooler | Noctua NH-L9a |
| Mobo | ASUS ROG Strix B550-I Gaming |
| Ram | 2x Kingston Server Premier 32 GB 3200MHz DDR4 ECC |
| RAIDz2 Storage Drives | 5x Western Digital 16TB WD Red Pro NAS Internal Hard Drive HDD - 7200 RPM |
| Apps/Docker Drive | 1x WD_BLACK 1TB SN770 NVMe Internal Gaming SSD Solid State Drive |
| Boot Drive | 1x Kingston 240GB A400 SATA 3 2.5” Internal SSD |
| PSU | BN639 SFX L Power 600W |
| Port Expansion | IO CREST M.2 B+M Key to SATA III 2 Ports Expansion Card Jmicro JMB582 Chipset, Add Two SATA 3.0 Devices to Any M.2 2242 Slot SI-ADA40149IO CREST M.2 B+M Key to SATA III 2 Ports Expansion Card Jmicro JMB582 Chipset, Add Two SATA 3.0 Devices to Any M.2 2242 Slot SI-ADA40149 |
| Sata Cables | ADCAUDX SATA-III Cable:0.5M 6Pcs/Set-SATA Cable 90-Degree Right-Angle SATA-Data Cable SAS/SATA-6Gbps Cable |
The decommissioned PC that’s just laying around has:
| Part | Model |
|
|
|
| Case | NCASE M1 |
| CPU | i7-9700 |
| CPU Cooler | Noctua NH-C14 |
| Mobo | ASRock Z390 Mini ITX |
| Ram | 2x16gb DDR4 3200 |
| Drives | 2x Samsung 970 Evo Plus 1 TB NVME |
| Graphics Card | PowerColor Red Dragon RX 5700 XT 8 GB |
| PSU | Corsair SF600 |
