Problem with new SSD and DL 380 G7

[Explorermf1]

Hi!

I have a DL 380 G7. I bought a couple of new SSD disks, HP 800GB 6G SAS Mainstream Endurance SSD (690823-B21). After installation, system detected disks, I configured RAID 1 via SSA and added to OS. I tried migrate files from other arrays. It was OK I then decided to change the drive bay. I unmounted the drive from OS, removed the array via SSA, shut down the server and replaced the SSDs from 1, 2 bays to 3, 4 bays. And power on... Both SSDs show critical errors. Not recognised by iLO, Smart array p410i and SSA. Integrated management log shows:

Internal Storage Enclosure Device Failure (Bay 3, Box 1, Port 1I, Slot 0)
Internal Storage Enclosure Device Failure (Bay 4, Box 1, Port 1I, Slot 0)

However, the system boots correctly. I reset the iLO. Changed drive bays. Changed server. Situation is not changed. Drive status: fault. It looks like if you add an ssd disk, it will only replace it after crushing. But it can't be like this.

I would appreciate any idea.

[TVVJ]

Hello,

Please swap them back to their original position and see if the operating system is accessible.

Regards,

[Explorermf1]

Hello TVVJ,

It was tried first, but no.... I will say more: the orange LED on these SSDs comes on even before boot, regardless of server and bay position. This makes me wonder if the SSDs have somehow been marked as failures due to the replacement process. If so, is there any way to fix this?

[Tam92]

Hello,

You may check the status of the SSD in iLO. If it is marked as predictive failure, then yes you would need to replace the drive.

As you have swapped the disk bays, so the Logical drive would have moved to failed state, which could you also be a reason for the amber light.

You may launch offline ACU, and check the status of the SSD.

https://support.hpe.com/connect/s/softwaredetails?language=en_US&collectionId=MTX-43871f71327848ab&tab=Installation+Instructions

Thanks,
TAM

[support_s]

Hello,

Let us know if you were able to resolve the issue.

If you have no further query, and you are satisfied with the answer then kindly mark the topic as Solved so that it is helpful for all community members.

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

Thank you for being a HPE valuable community member.

[Explorermf1]

Hello @Tam92 ,

thank you for refer. I'd like to add a bit of information about this problem. After some thought, I decided to install these 2 SSDs in Gen10. SSA on Gen10 has the ability to erase discs. The drive was erased and, miracle of miracles, when installed in G7 it was recognised as new when the server booted. How happy I was. But it was not that simple. I decided to use ORCA this time and only used one SSD. Using ORCA I created a RAID 0 and decided to install the operating system (in this case WS2012R2). Everything went very well. SSA showed that the drive was in perfect condition. I installed the drivers, the drive showed the expected data rate. I installed the necessary software. Started checking everything for fault tolerance. .The system rebooted several times and everything was fine.....

Until I shut down the server. And then I turned it back on. And then, during the first initialisation, the drive went back to critical status. IMI reported:

POST error: 1784-drive array drive failure
POST error: 1785-Drive Array Not Configured

I replaced the Smart Array P410i Embedded with an External one. I erased the drives on the 380 G10 again and ... it happened again. OK, I did the same on the Smart Array Embedded P410i DL380 G7, 385 G7, 587 G7. Everything ended up exactly the same. I swapped bays, used drives individually or together, put both drives in 1+0 RAID, changed cache batteries, changed CMOS battery, reset Bios and p410i. As soon as I shut down or power off the server, the SSDs go into a critical state and lit up with a permanent orange indicator light. All of this happens before the splash scree "HP Proliant" appears. This is what happened when I unpacked the disks, installed them on the first server, created an array, partitioned, checked the functionality, deleted the array, turned off the server and moved the disks to another server to check compatibility. But the first time I didn't understand at what point the problem occurred.

SSD drives can't get past the G7 server's initialisation. As long as they are clean, everything is fine. I can shutdown or power off. Everything is fine. Once you create an array and then shut down the server - critical.

I bought the drives according to QuickSpecs HP ProLiant DL380 and DL385 Generation 7 (G7).
The box labelled HP 800GB 6G SAS Mainstream Endurance SSD (690823-B21) had SSDs p/n 765290-003 in the box.
The servers have the latest firmwares and work quietly with SAS 10K HDDs without causing these problems.

Honestly, I don't know what else would help anymore. Maybe I'm missing something in the case of using SSDs within the G7? I would appreciate any idea.

[Tam92]

Thank you for sharing more information. Thats quite strange.

Would you try installing a supported Windows OS and check. I think Win2012 is not supported on this server. ?

https://www.hpe.com/us/en/collaterals/collateral.a50010742enw.html

https://www.hpe.com/us/en/collaterals/collateral.a50010837enw.html

Check with Windows 2008.

[Explorermf1 · Accepted solution]

Hi Tam92,

Sorry for the long absence from the discussion, I needed some time to finalise this issue.

HPE Win2012 R2 is installed quite easily on the DL 380 G7. But the problem turned out to be something else entirely. I can't say exactly why, but it turns out that the same part number can be incompatible with a particular system. As I described in the question itself, I ordered an HP 800GB 6G SAS Mainstream Endurance SSD (690823-B21) and in the box was model MO0800JDVEV. So for some reason this particular drive was not compatible with DL 380 G7 and DL 385 G7, despite the fact that the servers have all the latest firmwares. The SSD have been replaced by the distributor, but the situation has not changed. I asked the distributor to replace the SSD with another model, MO0800JEFPB, which had the same replacement part number. This solved all my problems. The ssd are recognised, they can be transferred between servers and they can be rebuilt in RAID. It is hard to say exactly where and how the incompatibility arose, but that was the end of the story. Thank you for your active participation in the discussion and for your suggestions.

[Sunitha_Mod]

Hello @Explorermf1,

Great! 

We are glad to hear you found the solution and we appreciate you for keeping us updated.