TheBearJew
Cadet
- Joined
- Nov 4, 2022
- Messages
- 5
Hi all, I'll try to keep this short, sweet to the point.
I have an old Lenovo ThinkServer TS140. I bought it used and currently I have it running TrueNas Scale 22.12.3.1.
I'll admit, it's a little jank, but all in all it has been incredibly rock solid. System specs are as follows
CPU Xeon E3-1225 v3 (4core 4thread)
RAM 32GB ECC Memory DDR3
HDD 5x8TB, mixed of NAS Red drives and shucked drives (Pool1) RaidZ2
SSD 3x500GB Samsung Evo 860 (Pool2) RaidZ1
BOOT SSD 1x250GB Samsung Evo 860
Motherboard has 5xSATA
1xPCIE for 4 extra SATA (bought this on Amazon, cost about $20)
550w PSU (fairly new from my old rig. 3 years old 550w EVGA, I replaced the old one with this last month when I changed out the drives).
Recently I upgraded Pool 1 from 2TB to 8TB each. The upgrade went pretty flawless, but during the resilver process, one of the drives (bought brand new) showed 48 write errors. I take the drive out and run a smart test on my main rig, and Crystal Disk Health is showing healthy. (Weird, but okay).
Sucks, but it happens. So, I ordered a replacement drive, this time a WD Red (the original was Seagate Red), swapped it out, and started the process. Then I had the same issue. (I don't remember the error count but it was the same).
What are the odds that I had 2 bad disks, from different brands being faulty? So, I replaced the SATA cable. Still the same error. I unplugged the PCIE card and moved it to the slot underneath, ran a scrub, and low and behold, no issues. Been rock solid for 2 weeks.
Yesterday, I was cloning some data from one Dataset to another Dataset (on the same pool), and during that, I got an email saying there were 2 write errors. Crap, here we go again. I clear the zpool, waited, and about an hour later (it's still cloning), I got another email and it showed 24 errors.
Keep in mind, this has been the same disk.
Back when this first started I bought a replacement PCIE x 6 SATA card, but when I moved the slots, and had no errors, I just put it on the shelf just in case. So this morning, I tried putting the new card in my motherboard, and while my system posts, it does not boot, and does not even allow me to get into the BIOS. (Maybe its a Generation problem, but whatever).
So, as of today, I put the old PCIE x 4 in the graphics card slot (my last slot I have, and haven't yet tried until now) and rebooted the system. Upon boot, TrueNAS is showed the pool as healthy and did an auto resilver. After the resilver TrueNAS is showed 98 errors on the drive and has highlighted it as FAULTED.
My main question, do you think this is a sign of my motherboard failing? PSU power issues? I called my uncle and as a last hail Mary thinks it wouldn't hurt to try a different SATA Power cable. (FYI, the cable that is connected to the drive is a 3 daisy chained single cable plugged directly into the PSU).
Is there anything else you would do that I haven't done?
Thanks
TheBearJew96
I have an old Lenovo ThinkServer TS140. I bought it used and currently I have it running TrueNas Scale 22.12.3.1.
I'll admit, it's a little jank, but all in all it has been incredibly rock solid. System specs are as follows
CPU Xeon E3-1225 v3 (4core 4thread)
RAM 32GB ECC Memory DDR3
HDD 5x8TB, mixed of NAS Red drives and shucked drives (Pool1) RaidZ2
SSD 3x500GB Samsung Evo 860 (Pool2) RaidZ1
BOOT SSD 1x250GB Samsung Evo 860
Motherboard has 5xSATA
1xPCIE for 4 extra SATA (bought this on Amazon, cost about $20)
550w PSU (fairly new from my old rig. 3 years old 550w EVGA, I replaced the old one with this last month when I changed out the drives).
Recently I upgraded Pool 1 from 2TB to 8TB each. The upgrade went pretty flawless, but during the resilver process, one of the drives (bought brand new) showed 48 write errors. I take the drive out and run a smart test on my main rig, and Crystal Disk Health is showing healthy. (Weird, but okay).
Sucks, but it happens. So, I ordered a replacement drive, this time a WD Red (the original was Seagate Red), swapped it out, and started the process. Then I had the same issue. (I don't remember the error count but it was the same).
What are the odds that I had 2 bad disks, from different brands being faulty? So, I replaced the SATA cable. Still the same error. I unplugged the PCIE card and moved it to the slot underneath, ran a scrub, and low and behold, no issues. Been rock solid for 2 weeks.
Yesterday, I was cloning some data from one Dataset to another Dataset (on the same pool), and during that, I got an email saying there were 2 write errors. Crap, here we go again. I clear the zpool, waited, and about an hour later (it's still cloning), I got another email and it showed 24 errors.
Keep in mind, this has been the same disk.
Back when this first started I bought a replacement PCIE x 6 SATA card, but when I moved the slots, and had no errors, I just put it on the shelf just in case. So this morning, I tried putting the new card in my motherboard, and while my system posts, it does not boot, and does not even allow me to get into the BIOS. (Maybe its a Generation problem, but whatever).
So, as of today, I put the old PCIE x 4 in the graphics card slot (my last slot I have, and haven't yet tried until now) and rebooted the system. Upon boot, TrueNAS is showed the pool as healthy and did an auto resilver. After the resilver TrueNAS is showed 98 errors on the drive and has highlighted it as FAULTED.
My main question, do you think this is a sign of my motherboard failing? PSU power issues? I called my uncle and as a last hail Mary thinks it wouldn't hurt to try a different SATA Power cable. (FYI, the cable that is connected to the drive is a 3 daisy chained single cable plugged directly into the PSU).
Is there anything else you would do that I haven't done?
Thanks
TheBearJew96
