ServeRAID 4Mx don't boot... :-(

[wxc]

Hi,I'm french, excuse me for poor english

My ServeRAID 4Mx controller hangs in its post. It is diplaying its bootmessage, but it will not continue. It says: Controller 1 Slot 2, Power-on-Self-Test takes up to 3Min(s) and 0 Sec(s)_ and that's it. It wont boot any futher I can't enter my bios setup either.
My config : asus A7M266-D with 2 PCI 64 bits ports (slot 1 & 2)

BIOS Version 4.84

Any tips are welcome

 

[neilcresswell]

wxc,
Couple of questions...

Does the timer count down...or does the controller actually hang ???
If you unplug the disks does the controller generate a different error ??
Are you able to update the bios/firmware by any means possible ??

N

 

[wxc]

Thinks neilcresswell,

The timer should down.
If I unplug the disks, the controller generate the same error.
On the controller, there is a connector i2c/Jtag.

When I connect the cable on this connector and on the chart where is the rackable disks hard, I do not have any more the message of waiting and the computer posts directly invites it bios.
It is the only means which I found to reach the bios of the chart serveRaid.

 

[neilcresswell]

Has this ever worked in this configuration. It may be that the card does not support the mainboard bios.

Find out the PCI bus speed as the ServeRaid card is limited to 100Mhz and 3.3 Volts.

Try to obtain a BIOS upgrade for the Main board and try to install it with the ServeRaid card removed.

N

 

[npiersma]

Hello again,

I had this problem about 6 months ago. I ended up ditching the card. But I know what the problem is.

I tested this card in a DELL 500SC and DELL 600SC and had the same problem. Then I tested the card in a DELL 2600 and it worked like clockwork. Then I noticed something strange, the DELL SC500/SC600 have GREEN PCI64 buses. The 2600 has a white PCI64 bus. Then I thought let's test. I put the card in a 32Bit PCI slot and guess, same problem, the same message appears, (Controller 1 Slot 2, Power-on-Self-Test takes up to 3Min(s) and 0 Sec(s)_)Some how the when the card is inserted in some PCI buses, it is misinterperted or is just running/connection as PCI32 and not as PCI64.

I have contacted both DELL and IBM and they are not willing to cooperate. DELL says, buy a DELL PERC controller and IBM says buy a IBM server.

So trew away the controller and did't cry a bit.

Niels Piersma
Piersma.COM

 

[hanbo]

I've got the same problem, same motherboard (ASUS A7M266-D).
I put the board in another system (32-bit 33 MHz PCI) and upgraded bios and flash to latest IBM code, 5.11.05. Works OK in other system, but in dual processor Athlon, it still hangs immediately after identifying version.

IBM tech support acknowledged problem, and said to return it place of purchase. I made the mistake of purchasing the board from California Computer Center, who is insisting that they
will not issue an RMA for the board, even though they claim to be "authorized for IBM sales and service", I've got an open case number with IBM, IBM admits "compatibility issue" and isn't planning on fixing it.

 

[carlo2]

Hi,
did you try to enable busmastering on your motherboard.

I've had a similar problem with a Servraid 4H in a HP Visualize x-class Workstation (A1280C)and enabling the busmastering for the particular PCI-Slot helped.

Carlo

 

[teagle]

Sam problem for me on a ASUS A7V8X and a ServeRAID-4Lx...

Always this message:

ServeRAID-4Lx Slot1, Power-on self-test takes up to 3 min(s) 0 sec(s)

If anyone has a great idear to solve this problem, thanks a lot!

 

[lej]

This usually indicates a problem with the signal cable, SCSI bus termination, one of the devices/drives or power to the drive(s). One other important point. This adapter will not support 8 bit SCSI devices by design.

First disconnect the signal cable and press CtL+I during ServeRAID initilization to enter the mini config. Go to Advanced settings and set defaults (initilize config). This will delete any errous config data stored on the adapter.

Second, be sure that the SCSI signal cable is only terminated at each end ONLY. The adapter will terminate the bus at the adapter end. Be sure the drives or devices that will be connected are not terminated or provide Term Power. Term Power is provided by the adapter.

If multiple channels are used do the following with each channel seperately.

Be sure all drives and devices are the same type, either Single Ended (SE) or Low Voltage Differental (LVD). The bus must have only SE or LVD devices, they cannot be mixed.

Next, boot the system with all the drives disconnected but leave the signal cable attached. If the system completes POST reconnect the drives 1 at a time. If the system completes POST with a drive attached, connect another until you find the bad one or all drives are connected.

good luck

 

[Threadsafe]

I also have this problem on my ASUS P4PE (i845PE chipset) motherboard. Nothing I seem to do in the bios remedies the situation, and I can't "CTRL+I" to get into the config of the BIOS because it freezes right away at the post.

The post freezes even if there is no cable connected to the card, and even after I reset the "bad configuration" with my other machine and put it back into this machine (without any devices).

I've spoken to numerous people at IBM, and received no help whatsoever, everybody recants the phrase "we can't guarantee compatibility in non-ibm servers". That's great, you don't hear such rubbish out of nvidia regarding supporting their video cards on only their nforce based systems.

This is infuriating, as it works fine in my old 440BX board with a Celeron 466!

If anybody here has any idea, please let me know! Thanks.