Drobo

8D Accelerator SSD Failures

Hello-

I have an odd situation. I purchased two Drobo 8Ds in October 2019 and equipped each with five Seagate 10TB HDDs, as well as a Samsung 2TB 860 EVO SSD (Drobo-recommended) for acceleration. Both Drobos are equipped identically. One is used every day as my primary data drive on a 2017 iMac running macOS Mojave. The other is only used every three months or so to serve as a mirrored backup to the primary unit. It is then disconnected and stored offsite. Therefore, while the main 8D is up and running 24x7 since October, the second 8D has only run for a total of ~4 days.

Just before running the Chronosync script to mirror 8D #1 to 8D #2 for the third time since October, the Samsung accelerator SSD was reported as failing by 8D #1. I purchased a brand new exact replacement from Amazon and all seemed to be fine. However, shorty after plugging in 8D #2 (daisy-chained to 8D #1), IT’S Samsung SSD reported a failure. Remember, though purchased at the same time as the initial one that failed, it has probably been used ~1/1,000 of the amount of time.

I put each of the two failed SSDs into an external enclosure and ran Samsung’s SSD diagnostic software. No errors were reported, though I suspect it’s probably because they were connected to the Mac via USB 3.0. I reformatted one of them and it seems to be working fine, though I haven’t attempted to use it long-term.

Before contacting Samsung, I wanted to see if anyone had any ideas. I’m open to the idea that it was a bad lot of SSDs, but it still seems unusual to me that they would both fail at almost the exact same time, even though one was used so much less than the other.

Incidentally, the replacement SSD in 8D #1 has been humming along just fine for the past six weeks.

Thank you!
Dave

Hello-

OP here. I wanted to provide an update on this. I sent back both SSD drives to Samsung who confirmed that nothing is wrong with them. Just last night, the THIRD SSD I purchased failed in 8D#1.

So, in summary: two separate Drobo 8Ds, each with VERY different up times. Three separate Samsung 2TB SSDs used as accelerators (main data drives in both Drobos are spinning HDDs). Three failures – two in 8D#1 and one in 8D#2. Samsung confirmed that the first two SSDs are fine.

Has anyone else experienced failures of the accelerator SSD in a Drobo 8D?

Thank you!
Dave

Same issue. Two different 1TB Samsung SSDs (one 860 Pro and 860 Evo) reported failure within a few days of being inserted but were fine when tried in other devices.

The option did not seem to improve performance so I just gave up on trying it any further. Flawed implementation.

Sorry to not offer any help!

I have had the same problem. 3 Samsung SSD’s failing-- on DROBO 8D. Drobo provided no explanation but replaced the DROBO. Have had ongoing unexplained shutdowns of Drobo with error messages about ejecting drives before removing. Now I have had two Seagate HDs fail within 2 days (both less than 12 months old). I reformatted and check them with Disk Utility and Techtool pro 13. No problems, but replacing them in DROBO still shows they are “Failed.” Seems to be clearly a DROBO problem. I have another support ticket open but am dubious about this product.

First double check the drives aren’t on the list here:
Can I really use any hard drive I want in Drobo?

I’m unfamiliar with the tools mentioned being a Windows/PC user, however, if you’re using an Intel based Mac I’d suggest checking the drives using Seatools for DOS, if they pass (or are repaired by) the Seatools long test, then running the full zero fill should get rid of the failed message.

Instructions for running the tools on an Intel Mac can be found at the bottom of this page:
SeaTools for DOS tutorial | Seagate Support UK
Instructions for running the tests can be found earlier on the same page.

Failed tests using Seatools would be grounds to RMA the drives for replacement & would be accepted as such by Seagate.

Thanks for suggestion. Now the DROBO won’t stay on, can’t be discovered by my Mac, and the lights are flashing red indicating a failure of all the drives. The two drives that had been identified as having “failed” by the DROBO are fine when mounted directly on my Mac. This is a replacement DROBO because I had same disk failure problem on my first 8D after 9 months. The unit seems to be dead. There seem to be a number of other users who have had similar problems. I am hoping I can recover the data on another replacement unit (it is still under warranty) and then I am pretty much done with DROBO.

I hope the OP was able to resolve his issues.

I had come across this thread recently, and could not help but notice his main drives were 5x 10TB Seagates. This means that he had 3 open slots. Perhaps OP was not aware of the Data Aware Tiering feature which is available when 2 SSD’s are inserted into the normal 3.5" drive bays.

I have posted in this chat group, a separate thread addressing this feature.

Hello-

OP here. Thank you to everyone for your responses. As an update, I have given up on using an SSD as an accelerator drive in my setup, with little noticeable impact. The “failures” occurred even on brand new replacement SSDs, and I could determine no discernible pattern to the timing or circumstances around their failures. All drives – both the HDDs used for storage and the SSD used for acceleration – were on Drobo’s recommended list.

@jcfhsu - I’m not sure if I follow your response. In looking at the screen capture of your Drobo Dashboard, one thing I should point out is that I am not using SSDs in any of the data bays. All five of my drives are traditional, spinning HDDs as shown in the screen capture below. As such, I don’t get the warning that you indicate.

Wish I could afford an 8/12 bay unit to investigate these possibilities for potential acceleration, sadly best I have access to is a 5C.