IBM m1015 Problems...

eli_singer

New Member
Feb 9, 2013
26
0
0
well let's dive right in.
installed the IBM M1015 on the top PCIe in the motherboard (Intel DZ77RE-75K - BIOS version is the latest- 0064).
the end result would be having 4-6 4TB drives (for now) working in RAID 10.

this is what i get when i start the machine:
LSI MegaRAID SAS-MFI BIOS
Version 4.31.00 (June 08 2012)
F/W initializing devices
FW Package 20.10.1-0119

at first i only had 2 drives so i started by configuring it through webBIOS for RAID 1 (starting easy).
I'm using X2 HGST 4TB 7200rpm.

OS is win 7 ultimate X64.
installed win 7 drivers, driver version in device manager says: 6.504.4.0 dated to 05/10/2012 from LSI Corp.

MSM version was 12.10.01.02

it seemed to be working since i saw the drive in Windows and was able to work with it, only i keep getting this message when opening MSM:




if it matters, i initialized the drive in Windows Disk Managmnent and not in MSM.
anyways, i got the 2 other drives and i wanted to try the whole 4 drives array in RAID 10. that's when it really got worse.

when i have the 4 drives connected i can't get pass the F/W initialization screen. when it finishes the percentage count it just freezes after the 'FW Package 20.10.1-0119' line and starts beeping like crazy, only reset or power down can get it out of it.

so i said, maybe i'll try to delete the former RAID 1 array to get everything to square one and start it all over again. so i disconnected the 2 new drives and was able to delete the former RAID 1 array in the Web BIOS- weirdly, i had no mouse when i got to Web BIOS and it is like this ever since...
it didn't help. when i tried connecting the new drives again the same thing happened, freeze and beeps.

so i figured, maybe there's something wrong with one of the new drives? so i pulled out no.4.
this time i got through the whole POST process (drive number 3 was identified as JBOD automatically) but in the end, instead of entering Web BIOS i got a black screen with a blinking cursor on the bottom left...
if i'm not trying to get into Web BIOS, it starts windows normally, as always i get the fatal error message as mentioned above.

i was able to convert this drive to GPT and make a simple volume out of it in Windows Disk Management with no problems, just as if it was connected to a SATA port on my Motherboard.

so, i tried doing the same with drive no.4. i disconnected drive no.3 (the recent JBOD drive) and connected no.4.
this time, instead of a blinking cursor on a black screen, i was able to get into Web BIOS!
but, when i pressed 'start' on the first Web BIOS screen to choose an adapter it got crazy. it freezes or parts of the screen disappears and practically, i had no other choice but to reset the system. one weird thing to mention is that sometimes i get the mouse working here again for a few seconds before it freezes...

again, like with drive no.3, when i didn't try to get into Web BIOS windows started normally and i was able to configure the drive to work as a separate drive.
of course when i tried hooking all 4 drives again i still get the freeze and the beeps...


so to conclude, so far we have 4 working drives, 2 of them can work in RAID, the 2 others can only work as JBOD since i can't even get into Web BIOS to configure them, but they can't all work together anyway. only drives 1,2,3 or 1,2,4.

so i tried disconnecting drives 1 and 2 while having 3 and 4 plugged in, maybe i'll find something...
well, the weirdest thing happened, i got this screen:



when i tried entering the configuration utility it froze on this screen:



and i got the same old beeps. damn...

then, i unplugged drive no.3 so i would only have drive 4 plugged in.
got the same screen saying all the disks from my previous configuration are gone, this time it did got me into Web BIOS though i had nothing to do there since i only had one drive connected as JBOD...

so i could get into windows and have that one drive working.
i tried the same thing with only drive no.3 connected but i couldn't get into Web BIOS this time, instead i got the black screen and the blinking cursor again!
when i didn't tried to get into Web BIOS i was able to get into windows normally (of course with the regular fatal error message...).

so i thought, maybe it's the SFF-8087 cable that is defected?
and to my surprise, when i tried connecting drive no.4 with the end of the SFF-8087 that was connected to drive no.3 (let's call it cable no.3) i got the same black screen and blinking cursor that i got with drive no.3!
to be sure i tried connecting cable no.4 to drive no.4 again to see if i can get into Web BIOS like before. i could, but i got this screen:



i don't know what to make of this...

tried connecting drives 3 and 4 with cables 1 and 2- got black screen and blinking cursor if attempted to enter Web BIOS (better than freeze and beeps i got with cables 3 and 4), if i didn't try to get into Web BIOS, then i got to windows normally.
tried connecting drives 1 and 2 with cables 3 and 4- got everything working like it was originally with drives 1 and 2 connected to cables 1 and 2.

so i have no idea if it's the cable. seems unlikely.


last thing i tried was to switch the SAS end of the cable to the second slot on the M1015.
when having all 4 drives connected i got the black screen and blinking cursor if attempted to enter Web BIOS (better than freeze and beeps i got with the other SAS slot), if i didn't try to get into Web BIOS i managed, for the first time, to get into windows and see all 4 drives! although not in the configuration i wanted since i couldn't get into Web BIOS.
and of course, i still get the fatal error message...

but! i was able to configure the whole array as i wanted through MSM!! HURRAY!
one drive group with 2 spans in it, 2 drives per span.





so, now i'm transferring data to the new array. question is, what the hell is wrong here? why doesn't it work as it should and how to fix it?
what's with that fatal error message i keep getting?

any help would be much appreciated since i'm afraid this whole thing could crumble on a moment notice...

thank you for taking the time to get through this ordeal :)

Eli
 

eli_singer

New Member
Feb 9, 2013
26
0
0
well i get this when trying to update fw:



no idea why...


- i closed the other thread, it was more of finding out the voltage problem which is a different topic.

Eli
 
Last edited:

Lost-Benji

Member
Jan 21, 2013
424
23
18
The arse end of the planet
Now I know I have asked this and haven't seen a response, have you taken the car to another system and tested it?

First place to look.
http://download.intel.com/support/motherboards/desktop/sb/biosglossarybymenu_v13.pdf
Look at page with this setting and see what yours is set at:
PEG Allow > x1
Now looksy at page 16 of this:
http://downloadmirror.intel.com/21240/eng/dz77re-75k_TechProdSpec02.pdf
You can see that the PCI-E slots are driven from the CPU (normal for this socket) and there is a "SWITCH" sitting on the last 8 lanes of top slot to divert them to bottom slot.
I would be placing a small video card in the TOP slot and put the M1015 into the BOTTOM slot.

The board is not a board I would be using for anything other than video cards like it was intended. The board is an enthusiasts board, not aimed at the market wanting stable "Server" style systems that just work.

Another option is to source another M1015 and test to see if wrong board or just a bad card. They are on eBay for a reason usually.
 

eli_singer

New Member
Feb 9, 2013
26
0
0
i have no other working machine with the specs to support the card to check it with.

PEG is enabled.

at first i tried putting the card in the second slot but it wasn't recognized so i switched it to the first slot an it was recognized fine since then.

i don't have another M1015 and i'm trying to avoid spending that money without knowing what the problem is...

Eli
 

Lost-Benji

Member
Jan 21, 2013
424
23
18
The arse end of the planet
What I am getting at (sorry, I should have been clearer above) is that there maybe two possible issues. The first is that these boards are like the famous Gigabytes and ASUS boards that ONLY worked with GPU's no other cards. This was the way the these cards BIOS's reported to the board. Hence why I suggest putting any kind of video card in top slot to trick the board into playing ball.
The other reason maybe that the PCI-E switches may not toggle if nothing is in top slot to feed the last 8 lanes of top slot to the bottom slot.
 

eli_singer

New Member
Feb 9, 2013
26
0
0
but if the card is working (not without problems, but still working and recognized) than it's not a PCI-E issue right?
i have the GTX 660 on the top slot and the M1015 on the second. both should be getting 8 lanes.

Eli
 

mobilenvidia

Moderator
Sep 25, 2011
1,804
120
63
New Zealand
My GTX470 happily shares slots/PCIe lanes in my Z77 Mobo.
As works fine in my testing mobo an old P55

The switch to share lanes is built into the CPU, so little can go wrong here, more PCIe bridges may throw a curve ball at it, but also unlikely as my ASUS Z77 WS had no issues with PLX bridge.
 

Lost-Benji

Member
Jan 21, 2013
424
23
18
The arse end of the planet
but if the card is working (not without problems, but still working and recognized) than it's not a PCI-E issue right?
i have the GTX 660 on the top slot and the M1015 on the second. both should be getting 8 lanes.

Eli
Yes, both should be seeing 8 lanes each. If you have video working but the RAID/HBA card is having a spit, it leaves two options; Failed/bad card or touchy board. Hence why I have said, try in another system. If plays up there, replace card.




The switch to share lanes is built into the CPU, so little can go wrong here, more PCIe bridges may throw a curve ball at it, but also unlikely as my ASUS Z77 WS had no issues with PLX bridge.
No, lanes are switched by lane switches, not the CPU. Look at the link to image below then take note of the four little darlings sitting to the right of the PCI-E x1 slot below to the first PCI-E x16 slot.
http://www.xbitlabs.com/images/mainboards/intel-dz77re-75k-dz77ga-70k/board70k_big.jpg

Already covered in post #5. http://downloadmirror.intel.com/21240/eng/dz77re-75k_TechProdSpec02.pdf Page 16, Block Diagram 1.1.3
 
Last edited:

pettakos

New Member
Oct 9, 2012
15
0
1
Can I populate the first pci-e with a graphics card, if needed, and the other three with the controllers?
 

mobilenvidia

Moderator
Sep 25, 2011
1,804
120
63
New Zealand
Yip
I had my Geforce GTX470 with the 3x LSI controllers, all worked happily together.
GTX470 also worked in other than 1st PCIe slot
With 4x cards the 4x PCIe slots all run at 8x
 

mobilenvidia

Moderator
Sep 25, 2011
1,804
120
63
New Zealand
Try to avoid lightning :mad: , it fried one of the LAN ports and later 3 of the Memory slots refused to work, now the Mobo just gathers dust
I do miss it, and the 4x PCIe 8x slots, since bought an expander so as to run all my drives with a single controller.
 

CFelipe

New Member
Aug 19, 2014
3
0
1
50
Hi guys,

I don't know if Eli solved his problem, but I am with the same effects here, and the only thing we have in common is a RAID card (LSI based) using a regular 16xPCIe slot.
I got the same message from MSM (image above) saying "controller encountered a fatal error and was reset" every boot.
If I use only my first drive, nothing happens and I can work in peace. If I use the disk on PERC H310, a BSOD appear in a few minutes (1A - memory management is most frequent).
To trigger the error, I use robocopy to sync my files, and the BSOD happen.

I tested the memory twice for 8 and 12hours, I stressed the computer with OCCT and prime, with and without OC. No problems.
My PSU is ok (I tested another one too). All computer parts are new, except for the GPU, that was running well in my old computer.
I changed the slot, and the problem persisted. I tryed to every configuration possible in the BIOS, I updated drivers, firmware and BIOS, disabled onboard devices, tryed LSI drivers, Microsoft drivers, removed everything to the point I had only Microsoft drivers installed. Removed even my programs to the point I had only Windows and MSM. Used verifier to check my drivers, reinstalled Windows, and one month of pain, and a lot of nights googling for some answer, I have no answer.

All problems happens after I installed a Dell PERC H310 in my computer, and mounted a RAID 5 with 4x3TB disks. It's was created in my first try, initialized to the end without errors. The RAID is online and has no errors (and keep a lot of data). I tested each of HDDs, one by one, with Seagate Tools. I also changed my cables (I bought 2 sets from Intel).

I did not used the tape trick to put my RAID card working. Every boot is ok, and Windows load to the end without problems.
I tested the Perc in one Dell T320 Server, stressed it for 2 days and had no problems with that.

I can't be defeated without an answer. Why? What?

Any help is welcome. Thanks for read this.

My specs are:
MB: Asus Rampage IV Black Edition.
Mem: 2x8GB Gskill TridentX 2400
CPU: Intel 4930K
First drive: Raid 0 (2xOCZ Vector 150) using Intel RST.
Second drive: RAID 5 (4xST3000DM001) using Perc H310 for Precision Workstation (SubDeviceId 1f78, not the regular 1f4e of Dell Servers)
GPU: Nvidia 560Ti
PSU: Corsair AX1200i
Windows 8.1 Pro 64
 

Patrick

Administrator
Staff member
Dec 21, 2010
12,003
4,987
113
Just wondering, how are the cooling on the LSI cards? My biggest issues with them have been due to a lack of cooling.
 

CFelipe

New Member
Aug 19, 2014
3
0
1
50
I did nothing about it, since there were a 200x200mm fan in front of RAID Card (case side fan at 800RPM), 2x140x140 at front case and 3x140x1400 at top/rear.
After your post, I put a temp sensor in heatsink card and a 40x40mm cooler attached to it.
The heatsink temp was 36ºC/97ºF, during the last BSOD.