jmacdowall

I have an unRAID server with about 45T of files on 60T of RAID The server: ASRock Z370 Gaming K6 64GB RAM LSI 9300-16i 12Gbps SAS-3 PCIe x8 HBA P16 IT mode 15 HDDs: 6x 6TB, 9x 4TB (mix of WD, Seagate, Toshiba) Corsair RM Series, RM850, 850 Watt, 80+ Gold Certified P/S ASUS XG-C100C 10G Network Adapter Rosewill 4U Server Chassis 15 Bay Server Case I'm running 6 rsync screen sessions to backup my data to a Debian server hosting a RAID2z ZFS RAID comprised of 7 Seagate 16TB drives. The servers are connected with 10G NICs through a QNAP QSW-M408-4C 10GbE Managed Switch. The multiple rsyncs are copying data from 6 different directories. Running parallel rsyncs gives me the maximum throughput, which tops out around 3Gbps. This runs for about 12 hours and then the server crashes. Cannot be reached by http or ssh. The screen is blank and cannot be resumed by hitting the keyboard. So I hold the power button down. Right now the backup has been running for 6 hours. COREfreq: --- Freq(MHz) Ratio - Turbo -- C0 ---- C1 -- C2:C3 - C4:C6 --- C7 -- Min TMP Max -------------------------------------------------------------------------------------------------------- 000 409.26 ( 4.09) 13.64% 9.01% 5.64% 1.80% 9.52% 74.03% 27 / 33/ 54 001 100.22 ( 1.00) 3.34% 2.21% 5.15% 1.04% 5.43% 86.18% 27 / 31/ 54 002 193.66 ( 1.94) 6.46% 4.26% 5.44% 13.39% 3.62% 73.29% 26 / 30/ 57 003 122.85 ( 1.23) 4.09% 2.71% 5.27% 12.62% 5.27% 74.13% 26 / 30/ 53 004 123.51 ( 1.24) 4.12% 2.73% 8.31% 3.11% 7.56% 78.28% 28 / 34/ 57 005 135.90 ( 1.36) 4.53% 3.00% 8.43% 3.73% 5.99% 78.84% 28 / 31/ 58 006 98.22 ( 0.98) 3.27% 2.16% 6.10% 0.61% 9.88% 81.25% 27 / 31/ 55 007 308.42 ( 3.08) 10.28% 6.81% 14.11% 38.81% 5.21% 35.05% 26 / 30/ 52 ------ % Averages [ 6.22% 4.11% 7.31% 9.39% 6.56% 72.63% ]------------ smartctl disk temps are averaging (Min/Max 27/30) This runs for about 12 hours and then the server crashes. Cannot be reached by HTTP or ssh. The screen is blank. oughput, which tops out at around 3Gbps. When the system crashes I have to power cycle it. It boots and I start the array and then the rsync procs in their screen sessions. Help? tower-diagnostics-20240105-1720.zip

Ah! That's what I was missing! Also, resocketed RAM. Will be in touch!

Recently my unraid server has not stayed connected to the network for more than 24 hours at a time. I can reboot it, restart the array, and it works great for less than 24 hours. I ran memtest for a couple hours with no errors. Attached is the diags. tower-diagnostics-20231213-1123.zip