R6515 and samsung PM9a3

Hi, thanks for choosing Dell. This is unheard of, let's wait around to see if any other users could provide some insight. Wish you a good one.

Could you please handle this as DELL too? 

I am unsure there are many relevant users that will be able to provide information about this.

I don't think this is completely unheard of. In the very recent months, DELL release an update for PM9A1 (https://www.dell.com/support/home/en-sg/drivers/driversdetails?driverid=v22yd) so I wouldn't be surprised that PM9A3 needs a similar update.

We have 8 new R6515 servers and 16 PM9A3 disks.

Only the disks that have the GDC5502Q (Feb 2022) firmware fail when inserted in DELL R6515 servers.

Hi,

we did not had this failure till you reported this.

Regards Martin

I see. Is there any chance this could be investigated?

Samsung PM9A3 are very standard datacenter disks

Hello jleni,

I sorry to see that issue on the drives. I'm not seeing that we validated the PM9A3 for the R6515  but it could help if you could provide the DP/N number from the label (the Dell part number)? 

Can you post a picture of the label?

I would also recommend make sure the system firmware is up to date: iDRAC, BIOS, PERC controller and backplane.

Amazing answer Andrey!!!

We actually just tried using ASUS servers and they work great!! 

btw, how did you manage to get GDC5602Q or any other firmware? I have been trying to find some other version and it has been impossible!! Did you get it from Samsung? Do you have some link?

We had similar issue with both R640 and R740XD with nvme backplanes. After long research it became obvious that there is a problem with PM9A3 model and some Dell backplane. They are randomly got "errormod"ed during using. We tried to flash them to GDC5602Q, no difference.

It is not a problem to get them back to normal state (linux utility and nvme format is your friend). However it is totally not production ready.

Other samsung models like PM983 working fine. From another side, PM9A3 is also working just great on other server vendors (for instance, asus).

We end up with migrating it to asus storages and using Micron nvme with Dell.

From what we know from investigation, PM9A3 is very sensitive to power and it might be the reason.

we got it from samsung, but its under nda I believe. anyway, its not helped with the issue and we have no release notes as well.. no reasons to worry

While I know this topic is old, I can confirm across a large fleet of systems that GDC5502Q was definitely the culprit.  We've upgraded all disks to GDC5602Q and have yet to see an ERRORMOD failure across ~1k PM9A3 disks.