R720XD 24 X 1.9TB SSD CONFIGURATION

Papanos,

Would you verify if the server is up to date on BIOS, iDrac, as well as the H710p?

Also, I don't see a disk limitation to the controller, but I do see a limitation of 16 on Virtual Disks to the raid groups, but that is a limitation of the server.

Hello Chris, 

Bios Version is 2.9.0

iDrac FW is 2.63.60 Build 1

Currently installed is the H710mini is 21.3.5-0002

Hi,

What RAID level are you using for your 24 drives in the array? Did you create the RAID in the RAID BIOS or LCC?

Crashes, by meaning the server restarts itself or there is an error message? What SSD are they? Are they Dell certified drives? Do they have Dell PN#? The RAID controller H710 mini, is it an upgrade or it's original with the server?

Hello Joey, 

We are trying to setup the ssd in RAID 5 in bios and the LCC. 

Both attempts after saving resulted in the server restarting and then on boot the health indicator and the electrical indicator starts blinking and the server shuts down. 

The H710 mini was already installed in the server. 

The SSDs are the Kingston DC450R 1.9TB ones. They are not Dell certified but up to 15 they work just fine. 

The problems start over the 15 disks thus meaning the internal storage exceeds 30TB. 

The Power Supplies were changes to 1100W ones as well to support all the drives. 

One error we see on the the logs is the BP15V PG voltage outside of range. 

Hello,

do you have some OCZ SSD drives also? Or all SSD are Kingston?

We had some issue with OCZ drives that cause this BP15B error (issue was on the OCZ).

I would check if there is some new firmware on the Kingston site (I know that there is SSD manager to update it)

Marco

Hello Marco, 

The SSD were already tested and all are the latest firmware on the SSD manager.

The thing is we would face this problem from the start and not after the 15 disk mark . this is the strange thing. 

Can you test those SSD in another Dell system?

Thanks
MArco

Hello Marco, 

8 of the1.9tb were tested on an R420 and all work as should. 

we tried one more thing and updated the firmware on the PSU. The raid 5 was completed but then after a reboot the server shutdown would not boot and the electrical and health indicator started blinking. 

One thing we noticed after many tries is once the server enters the state were the electrical and health indicators starts blinking you have you to remove some ssds

In this case when the server was left with 2x  600GB DELL 2.5 AND 13 X 1.9TB KINGSTON the server started again normaly. 

Also after research regarding the The system board BP1 5V PG voltage is outside of range error i saw that on R740/R740xd it was solved after updating the CPLD version to 1.0.6 

The cpld version on ours is 1.0.3 maybe that is a problem ? i could not locate this anywhere to download and try though for R720xd

Hi,

Thanks for your troubleshooting on this, and I did have a thought that the back plane would not be able to handle the amount of active drives with it's power consumption. That's the reason why that the drives are detected and when they are in the RAID array, the drives are active. 

Regards to the CPLD update, I don't think there is any updates for the server. But if you could, do send a private note to @DELL-Chris H your service tag of the system.

By the way, could you check if you are able to update the SSD firmware that you have there. 

hello Joey,

Regarding power usage i have found the bellow and did a compare with the bellow Dell Drive. 

Dell K5P0T / Samsung SM863a MU 512n MZ7KM1T9HMJP0D3 1.92TB 6Gbps 2.5" MLC SATA SSD --> 

Active Write -->  3.0 W /Avg Write on Kingston is 3.44W

Active Read --> 2.5W / Avg Read on Kingston is 1.42W

Idle --> 1.4W/ Idle on Kingston is 1.22W

The only consumption that is over the Dell branded ssd is the Write one but i still dont understand how it would create a problem during boot. if it would create a problem that would be during creation of the Virtual Disk where most of the writing happens but this never created a problem nor logged an error. 

the exact point that the server hangs is when the server is booting up and does what i believe a quick check on all installed hdd/ssd. 

Papanos,

As they aren't certified drives we don't have the ability, or or information, to be certain of the reason they are having an issue. If you had certified SSDs we can test to see if they allow more than 15 drives without issue, beyond doing that that we can't be certain the issue isn't the drives. 

Also, regarding the drives being seen. There may be some compatibility with the drive in the form of communication, but it may not extend to functionality. So the controller may see them, but not talk to them correctly, or vice versa. 

Hello Chris, 

Please let me know which 1.9TB drives are compatible so all 22 of them work on the R720xd. 

The site shows a couple different 1.8TB drives, seen here. If they specifically need to be 1.9TB let me know and I can give you some internal part numbers.