Hi,
The drives left today for RMA at the place I bought them from - so hopefully this is just something to expand my knowledge, rather than needing to encourage them to accept the RMA and send me new drives. . .
I bought 8x WD 6TB Red Pro (7200rpm) drives for my new server, as I’ve had good experiences with WD Red Pro dives in the past… and they had stock of the 6TB…
On arrival, I connected them up to my HBA (LSI 9400-16i) and two were not detected… there was also a lot of noise / vibration and some ‘beeping’ (for want of a better word).
Did some checking - switching connections on the MiniSAS to SATA cables, and the problem followed the two drives and not the cables / ports. So I’m happy enough that it is the HDDs and not the rest of the setup. Also, the two drives are the cause of the ‘beeps’.
Interestingly (and something that the place I bough them from also noted as a potential blip for the RMA process), both drives are detectable from the motherboard in my Windows PC. . . so it seems to be an HBA related issue(?)
So, just expanding my knowledge - is this a common issue or a bit weird???
Double check your power setup. Adding an HBA adds quite some demand for power to your system. Maybe it overwhelms your PSU or the cabling to the HDDs.
It took me quite some time to figure out a similar issue.
It is an overkill 1600w PSU, so I don’t think that’s the issue…
The MiniSAS to SATA cables have 4 different SATA ports, and the problem drives were an issue regardless of which port they were plugged into, and whether there were any other SATA drives plugged into the HBA.
I did also have 4 ‘good’ drives plugged into a single cable and all were detected fine.
I did also update the HBA’s firmware and BIOS once I first detected an issue… so it shouldn’t be that either.
It may have been more “testing” than actual testing, but that are the checks I went through… like I said, the drives are back at the supplier now, so I only have the ‘good’ ones to do anything extra with. (And I think some of the noise / vibration is the old TT Armor showing its age)
‘Drive 8’ was supposed to be a cold spare, so only entered testing when we wanted to check whether it also had a potential issue.
Well, the RMA was accepted and replacement drives arrived today…
I have just moved (well, almost completed the move) to a new case, so the PCIe cards are in different slots now, but the new drives all detect fine on the same hardware / same MiniSAS to SATA cables… so pretty confident that it was the drives
Would still like to know why Windows was able to detect them and the HBA would not.
It could be that the HBA tried to run a test, and the defective drives returned an error? Or, bad voltage on the data line from the HBA, and the new drives are more tolerant?
These seem pretty unlikely, and I am not knowledgeable on the subject, but I would think both of these should be possible.
Starting to think it’s either a bad batch or the 6TB WD Red Pros are not great…
I turned on SMART checks in TrueNAS and it detected a SMART error on one of the replacement drives.
After checking with WD’s tools, I’ve got 5 drives with SMART errors. . . in other words, our of a total of 10 drives, I’ve got 3 that seem to be OK.
The 2nd RMA was based on SMART errors, 4 of the original drives, and 1 of the 1st set of replacements.
This 3rd RMA is based on more SMART errors, 2 of the 2nd set of replacements. . .
At this point, I would highly suggest that you download and run the WD Dashboard and check for SMART errors on any new 6TB WD Red Pros - if not keep away from them all together.
