Hi, I have a supermicro X10DRD-iNT which has two U.2 connectors on it.
I got two Intel 750 SSD's, 2.5" 400Gb. I can't get these to work.
If I plug both in, computer freezes at the last bit of POST, saying ready to boot...
If I plug one in (doesn't matter which one, i've ruled out faulty cables/connectors), it boots up fine.
then I'll have to hot-plug the other one once windows loaded, which half the time freezes windows.
I have managed to boot it all the way to windows with both connected twice over three days and probably hundreds of reboots.
nVME drivers installed, RST installed, SSD toolkit says they're both fine the few times i've been able to see them. updated firmware on both.
flashed BIOS to latest update, cleared CMOS. fiddled around with the UEFI settings.
supermicro support said something about the SSD's not having any option ROM. setting this to EFI or Disabled makes no difference.
i'm on windows 8.1 at the moment but also purchased a windows 10 pro license. both behave the same.
I don't have the OS on these disks, it's on a samsung evo on sata.
We understand you have two Intel® SSD 750 2.5" drives and when both are connected, the system freezes while posting.
Based on the information you provided and the troubleshooting performed on your side, we can see that the problem is most likely related to either a power issue or BIOS configuration, this because when you test the drives as single, they both work without issues.
We checked the https://www.supermicro.com/manuals/motherboard/C600/MNL-1712.pdf motherboard's manual and we understand that the two P1-NVMe ports are supported by CPU1; However, we did not find any BIOS setting to maybe discard a situation with power or bandwidth feature.
Are you only using these two NVMe* drives with the Samsung* drive or do you have more drives plugged in? If you have more connected, please let us know.
Have you contacted Supermicro* to verify this situation (A power or BIOS issue), it is clear for us that the SSDs are fine since both work separately.
We'll be waiting for your response.
Hello Nestor C
thanks for your reply.
I have only the samsung and the two Intel drives connected, correct.
I also thought it might be power related so earlier I put in a new feed from the (1200W)PSU so I had one Intel drive on one cable and the other plus the samsung on another cable. didn't do anything, so I then put each drive on a separate feed. no result.
I get the feeling this is more of a windows-issue than bios post. it's where that turqoise windows logo shows up and the spinning dots appear. either it stops there or goes on. but i don't know.
I have been emailing with supermicro support and I'm waiting for them to reply to my last update but that probably won't be til tomorrow.
i've been messing around with this all day and it's just refusing to work stable. everything is healthy but something is making it crash and act very strange.
All of a sudden a chassi fan stopped spinning and it came back on when I pulled some stuff out. So i'm thinking insufficient power supply. and especially with the whole 1 Intel connected boots fine, 2 completely freezes.
but I have a 1200W corsair, can that be possible? there's only the 1500W left in their range, don't know if anyone else makes more powerful psu's but it seems a bit crazy? maybe I should get a second one and just put all peripherals on that..
I have 2 xeon cpu's
2 noctua cpu fans,
3 chassi fans all 140mm i think.
1 samsung ssd
2 intel 750 ssd
and that's about it at the moment.
I haven't even connected the gpu yet which is a quadro p5000.
the PSU feels a bit on the warm side considering i've not really stressed it, as I can barely get in to windows.
this sucks so much, I had it running for maybe 20mins earlier today and managed to benchmark with cinebench, came out over 4000.
and I tested the seq. speed of the intel drives which looked fantastic! just really want to get this up and running.
any input, anyone?
This is a really strange situation since the PSU you have should be able to handle the server fine with both SSDs connected.
It would be nice to try a different PSU just to discard any issue with it. We would like also to ask you if you have the latest NVMe* drivers installed already? If not, https://downloadcenter.intel.com/download/26833/NVMe-Drivers-for-Intel-SSDs here is the link.
One thing you can check with Supermicro is to verify if there is a tool to diagnose the hardware, we cannot discard as well that the U.2 ports may be the culprit but only when both drives are connected.
We'll be waiting for your response.
ehrm, you discontinued the 750's now!? ..brilliant, thanks.
anyway, i've messed around with this a heck of a lot more and it's nothing to do with power.
either it's a faulty cable, i've ordered a new one but i don't have much hope, or, judging by the number of complaints, this is all a bit hit and miss and i've ended up with crappy disks or something is conflicting in the build, in which case i'll have to replace them.
i'll keep you posted.
The Intel® SSD 750 series was discontinued on April 30th, 2017. You can check the http://qdms.intel.com/dm/i.aspx/E3A4F190-F8EC-4E32-B731-24EBFD3C62DC/PCN115375-01.pdf product change notification for that.
Now, about the situation, we believe as well this could also be related to a cable; However, it is hard to say since each drive works fine individually.
Please keep us posted, it will be nice if Supermicro* could also check about this situation with other drives, just to confirm the two ports can handle this configuration.
still no luck getting this to work.
here's a rundown.
I have all the latest drivers for RST and nVME.
I can't boot the machine with both drives attached, supermicro confirms both ports are meant to work at the same time.
I can boot with one drive attached.
if I hotplug the drives, one works fine, put in the second, windows gets extremely slow/becomes unresponsive.
the MB has onboard Ethernet, intel X540-AT2. when I hotplug the ssd's,(scan for new hardware in DevMan) network connection shuts down temporarily and comes back on if it's successful and windows doesn't crash. (EventViewer says MEIx64 service shuts down and sometimes RST service terminated unexpectedly)
when windows crashes, it gives a bsod with watchdog violation. hal.dll+3795 and ntoskrnl+1953af
this is from the eventviewer (.NET runtime and Application Error):
Faulting application name: IAStorDataMgrSvc.exe, version: 188.8.131.526, time stamp: 0x51fb0c50
Faulting module name: unknown, version: 0.0.0.0, time stamp: 0x00000000
Exception code: 0xc0000005
Fault offset: 0x04154ca5
Faulting process id: 0x27a0
Faulting application start time: 0x01d30c5b3bc708dc
Faulting application path: C:\Program Files\Intel\Intel(R) Rapid Storage Technology\IAStorDataMgrSvc.exe
Faulting module path: unknown
Report Id: bbf626f4-0836-4c2c-8db2-6f61abd75283
Faulting package full name:
Faulting package-relative application ID:
this is an excerpt from intelSSDTDK log-file, some 1117 error here:
any thoughts on this? I have absolutely no idea what's going on.
drive failure? they're brand new, I've never had anything on them.
15:32:05:063668 [debug] CommandPath_NVME_MSFT.cpp(107) sendCommand: Attempting to send command via IOCTL_STORAGE_QUERY_PROPERTY
15:32:05:063668 [debug] CommandPath_NVME_MSFT.cpp(239) sendDeviceIoControl: About to call DeviceIoControl (IOCTL_STORAGE_QUERY_PROPERTY)
15:32:05:063668 [error] CommandPath_NVME_MSFT.cpp(300) sendDeviceIoControl: IOCTL failed via OS error: 1 = Incorrect function.
15:32:05:063668 [debug] CommandPath_NVME_MSFT.cpp(547) test: Test result: False
15:32:05:063668 [debug] CommandPath_NVME_MSFT.cpp test: Exiting
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(510) getNVMECommandPath: Failed to talk via: nvme::msft
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(516) getNVMECommandPath: Failed to talk NVMe. Trying MMIO.
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(519) getNVMECommandPath: Creating and checking command path: mmio::nsg
15:32:05:063668 [debug] CommandPath_MMIO.cpp(220) test: Entering
15:32:05:063668 [debug] CommandPath_MMIO.cpp(80) sendCommand: Sending command: Identify
Control Code: 2054
General Purpose: 0
Transfer Bytes: 4096
15:32:05:063668 [error] CommandPath_MMIO.cpp(169) sendCommand: IOCTL failed: 1117 = The request could not be performed because of an I/O device error.
15:32:05:063668 [debug] CommandPath_MMIO.cpp(236) test: Test result: False
15:32:05:063668 [debug] CommandPath_MMIO.cpp test: Exiting
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(529) getNVMECommandPath: Failed to talk via: mmio::nsg
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(535) getNVMECommandPath: \\.\SCSI1:. Target: 0. PathId: 0. LUN: 0. has failed to talk NVMe. Moving on.
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp getNVMECommandPath: Exiting
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(442) scanNVME: Checking for remapped RST controllers on \\.\SCSI1:
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(632) getScsiAddressesOfRemappedRstControllers: Calling DeviceIoControl IOCTL_SCSI_MINIPORT (IOCTL_REMAPPORT_GET_PATHIDS_IMPLEMENTED)
15:32:05:063668 [error] DeviceFinderExtension_NVME.cpp(647) getScsiAddressesOfRemappedRstControllers: Failed to call IOCTL_SCSI_MINIPORT (IOCTL_REMAPPORT_GET_PATHIDS_IMPLEMENTED). OS Error: 1117
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp scanNVME: Exiting
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(155) scan: Opening connection to \\.\SCSI2:
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(168) scan: Scanning for NVME support and devices
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(380) scanNVME: Entering
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(246) scanRSTeVMD: Entering
15:32:05:063668 [info] DeviceFinderExtension_NVME.cpp(269) scanRSTeVMD: NVME_GET_NUMBER_OF_PASSTHROUGH_DISKS for RSTe passthrough failed.
15:32:05:063668 [info] DeviceFinderExtension_NVME.cpp(269) scanRSTeVMD: NVME_GET_NUMBER_OF_JOURNALING_DRIVES for RSTe passthrough failed.
15:32:05:063668 [info] DeviceFinderExtension_NVME.cpp(269) scanRSTeVMD: NVME_GET_NUMBER_OF_SPARE_DISKS for RSTe passthrough failed.
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp scanRSTeVMD: Exiting
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(402) scanNVME: \\.\SCSI2: is the controller for \\.\PHYSICALDRIVE3
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(453) getNVMECommandPath: Entering
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(458) getNVMECommandPath: Testing for NVMe support on \\.\SCSI2:. Target: 0. PathId: 0. LUN: 0
15:32:05:063668 [debug] CommandPath_NVME_NSG.cpp(53) CommandPath: Creating async engine for AER
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(462) getNVMECommandPath: Creating and checking command path: nvme::windows_debug
15:32:05:063668 [debug] CommandPath_NVME_Windows_Debug.cpp(51) test: Entering
15:32:05:063668 [debug] CommandPath_NVME_Windows_Debug.cpp(86) test: Calling DeviceIOControl: IOCTL_SCSI_MINIPORT (NVME_DEBUG_DRIVER_IDENTIFICATION)
15:32:05:063668 [debug] CommandPath_NVME_Windows_Debug.cpp(100) test: IOCTL failed via OS Error: 1117 = The request could not be performed because of an I/O device error.
15:32:05:063668 [info] CommandPath_NVME_Windows_Debug.cpp(113) test: Windows Debug NVMe test result: False
15:32:05:063668 [debug] CommandPath_NVME_Windows_Debug.cpp test: Exiting
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(473) getNVMECommandPath: Failed to talk via: nvme::windows_debug
15:32:05:063668 [debug] CommandPath_NVME_NSG.cpp(53) CommandPath: Creating async engine for AER
15:32:05:063668 [debug] DeviceFinderExtension_NVME.cpp(477) getNVMECommandPath: Creating and checking command path: nvme::nsg
15:32:05:063668 [debug] CommandPath_NVME_NSG.cpp(613) test: Entering
15:32:05:063668 [debug] CommandPath_NVME_NSG.cpp(136) sendCommand: Sending command: Identify
Thanks for the logs provided over here. We would like to share this with our engineering department and check if there is something else from our side that can help on the resolution of this situation.
We'll be providing updates as soon as we have some.
Checking the specs of your system, we can see that your motherboard uses the Intel® PCH C612, which is not one of the validated chipsets for the Intel® SSD 750 series, as you can see here:
https://www.intel.com/content/www/us/en/support/memory-and-storage/enthusiast-ssds/000005797.html Before You Buy an Intel® SSD 750 Series
https://www.intel.com/content/www/us/en/support/memory-and-storage/enthusiast-ssds/000006027.html Frequently Asked Questions for Intel® Solid State Drive 750 Series
https://www.intel.com/content/www/us/en/support/memory-and-storage/enthusiast-ssds/000005956.html Physical Installation Instructions for Intel® Solid State Drive 750...
Is there any way that you can maybe use one system board with one of the chipsets mentioned in the links above? (Intel® Z97 Chipset or Intel® X99 Chipset)
Please also check your private messages for further information on this situation.
ah, i see. hm, yeah that might be the case then..
hm.. well i've got two xeon's running on this board.
I don't think there are any x99 or z97 with dual cpu, socket 2011-3 and two u.2 ports.
ok, well my bad. should've checked the specs before buying i guess..