Link to home
Start Free TrialLog in
Avatar of tom_szabo
tom_szabo

asked on

H710 Mini not working with WD Black 2.5"?

So I made the mistake of buying an R720xd which I now regret, as good 2.5" drives are so much less common and much more expensive.

We have SAS drives for the OS and important stuff, but we need some cheap storage for other files which dont get accessed much. I wanted to opt for 2x WD Black 2TB in RAID 1, but these do not come in 2.5", so I looked around and found a 2TB Seagate firecuda SSHD which was reasonably priced, bought them and they installed fine, setup the raid, RAID controller seemed happy. Boot into Windows and they instantly drop out of RAID and the bays start flashing orange. Ok fair enough, I tried, time to get the Blacks.

So after some research everyone seems to be of the opinion that the 2.5" WD Blacks are no different from the 3.5" model, even stumbled accross a few threads where people where running the 2.5" models in RAID 10 and other configs. So went ahead and bought 4x 1TB WD Black 2.5" and they are now doing the exact same thing as the firecudas, dropping out when Windows boots.

Is this actually a HDD issue or am I missing something?
ASKER CERTIFIED SOLUTION
Avatar of pgm554
pgm554
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
You need the drove to support TLER deals with responding within a set amount of tine, or the drive is kicked out as failed.
Which raid controller firmware version do you have/raid driver, is it up to date?

Get seagate savio 2.5 drives the 720xd has 24 bays if not mistaken why don't you get 4x1TB in a raid 10  
Make sure you check the tech spec that they are RAID suitable including TLER.
The hybrid SSD+HD type drives are not suitable
FYI

Dell drive firmware is tweaked to work with their controllers ,generic enterprise drives may not perform as well or reliably with non Dell branded drives.
Avatar of Member_2_231077
Member_2_231077

Dell don't do very much tweaking but they do write to a specific codepage with "dell certified" so generic enterprise drives may run OK but show as degraded in OMSA.
They do something as I had a WD desktop drive drive not even get recognized when I was testing drives at one point.
Worked fine on a regular SATA port,but not on a PERC port.
There was a version of firmware Dell deployed that would only recognized drives with "Dell firmware" after many complaints, they released a perc firmware that would interact with any sas/SATA drive but will always be in caution indicating because it is not a "Dell" drive, .....as Andyalder ref.
Unlike desktop systems where application and users interact with the storage and can wait "as long as it takes" raid HW has a specific timing and can not wait for ever for a response from one of the drives making up the raid volume that the command was received and executed and completed.
If a drive does not respond within this time, the raid controller kicks the non-responsive drive out of the volume,
@pgm554, it is only enterprise drives that I was referring to, Dell don't write "dell certified" on desktop drives because as you already said they're both unsupported and unusable.
Avatar of tom_szabo

ASKER

The thing is, for many years we have been running WD Blacks in numerous DELL PERC Raid arrays and never had an issue. Upon further investigation the 2.5" Blacks where not actually falling out of the array, I installed management software and it said the disks where 100% optimal with no errors, even though the front lights where flashing organge and the server screen is saying "drive fault" on those bays. Guess these newer PERC controllers really hate anything that isnt 'Dell Certified'.
Check the raid firmware version and see whether that is the source of your issue.
The non-certified disks should not be flickering Amber on the status light, it has two an activity and power/status light.
The price difference between a black and enterprise should be negligible.

Software RAID is a tossup between the blacks and enterprise,but the tler of hardware raid is where you can get burned,

If you have a drive that is falsely deemed as bad,then you have to do a rebuild,and if during a rebuild you get another drive falsely  flagged ,there goes your data if you're running raid 5.

Enterprise drives a have a better ecc bit rate too ,so having a double parity error is much less likely.
We only ever use RAID 1 so if the array does get dropped all the data should still be there.

I update the controller to the latest firmware and I'm still getting the organge blinking. Weird thing is when I log into iDrac it shows all the drives  as "good" with a green marker next to them, despite the whole server throwing an orange LED because of those HDDs.
Amber blinking, activity light is the effect of the led on the backplane
Commonly the power light flickers Amber when the power light/status flickers it suggest the drive has a predictable failure, check the hardware log using openmanage.
If all drives have the same color flickering light with the  other solid, it might be a non issue.
If you have a spare tray, illuminate the side at the end while looking at the indicators to see whether the two are supposed to be green, Amber or are transparent clear propagating the led color from the backplane?