January 2nd, 2006, 07:22 PM #1
LSI MegaRAID Elite 1600 (aka Dell Perc 493)
I have the above-named controller card, with both the BBU card and a battery-backed (literally) DIMM (128MB cache). When I try to boot up with it, I get to its 'POST' screen, briefly see "Firmware Initializing" and then it blinks to "Initializing SCSI Controllers" which just repeatedly flashes.
Glancing through the manual, it sounds like I have to charge this thing up for four hours. How do I do that - do I just pop it in and let it feed off standby power, or do I turn the computer on too? I had it do that yesterday for about half an hour but I didn't get anywhere. Any suggestions?
January 2nd, 2006, 08:27 PM #2
Does the card work without the BBU unit?
If I recall correctly, it should continue with the battery backup disabled. Once it's charged, it will activate.
January 2nd, 2006, 08:30 PM #3
I seem to recall that I tried without the BBU card in and got the same thing, but I wasn't at my best yesterday I can try it again.
January 3rd, 2006, 03:06 PM #4
If the card isn't working without the BBU, then, obviously, that has nothing to do with it.
January 3rd, 2006, 03:10 PM #5
I understand that. I'm just saying I can't remember for sure what I tried, as I wasn't at 100% on January 1st. This is a result of the effects of December 31st.
Now that I have my right mind on I'll try tonight.
January 4th, 2006, 12:20 AM #6
Well it's the darndest thing... I moved the card into this system, which has no SCSI drives, and it boots up just fine. Only thing is I can't Ctrl+M into the BIOS.
In the other machine I had it connected to three Hitachi 36GB 15K147s, already in RAID 0 from an LSI 21320-IS. No, the RAID 0 is not a long-term plan and I don't put anything important on it. Anyway, do I need to delete that array first?
P.S. I did not remove the BBU either.
January 4th, 2006, 11:36 PM #7
Even after I updated the card's firmware to the latest posted, the card simply sits on a blinking "Initializing SCSI Controllers". I tried it with all three drives attached, then two, one, and none (both ends of the cable disconnected) and this behavior is unchanged when used in my workstation. In my PC it boots up just fine, no drives there to attach. And I can Ctrl+M/Ctrl+H into it there too. Any known incompatibility with this card and the AMD 8111/8131/8151 chipset?
January 4th, 2006, 11:43 PM #8
I'm not aware of any compatability issues with the AMD chips; I'd suggest contacting LSI (if you have a true blue LSI card).
What's the motherboard here?
January 5th, 2006, 11:24 AM #9
January 6th, 2006, 12:41 PM #10
I typed "megaraid initializing SCSI controllers" into LSI's support search and got this result, I'll try it first before I email them:
Solution Title The MegaRAID controller takes an excessively long time to initialize (hours or more).
Solution ID ict7730
Date Modified 21-May-2005
Goal: The MegaRAID controller takes an excessively long time to initialize (hours or more).
Fix: Verify that Fast Initialization is enabled in the BIOS Configuration Utility (CTRL-M) in the Objects > Adapters menu. Update the firmware to the latest version available at www.lsilogic.com > Support > Download Center > RAID Adapters.
January 6th, 2006, 08:49 PM #11
Have you tried all three PCI slots?
Make sure that the card is in the BIOS boot order. If you have any other option ROMs, try disabling them.
January 6th, 2006, 08:54 PM #12
I will check the option ROMs and boot order tonight. I can only use it in the bottom PCI-X slot - the top is obstructed by my FireGL's heatsink. I did also try the regular PCI slot on the bottom, made no difference.
January 6th, 2006, 09:42 PM #13Originally Posted by sm8000They say technology slows down for no one. I know it outruns my wallet. I figure its because my wallet isn't light enough yet.
TechIMO Folding@home Team #111 - Crunching for the cure!
dulce bellum inexpertis
January 6th, 2006, 09:46 PM #14
No, the card WORKS in the system with no SCSI drives, no error message at all. In the system with SCSI drives, it doesn't work - with or without the drives connected, same error symptoms.
January 6th, 2006, 10:12 PM #15
Fast initialization was indeed enabled. Regardless I cleared config and reset to defaults; booted into Ctrl+M again and made sure Fast init was on. Guess I'll go into the Opteron rig's BIOS now and look at its settings.
January 7th, 2006, 12:21 AM #16
I looked through my BIOS and found this setting: PCI option ROM scan order. It was set to On Board Device First. I changed it to PCI Slot First, then tried the card. No joy. I then changed the boot order to Hard Drive(s)->Bootable Add-in card first, still no joy. I also tried the regular PCI slot at the bottom, wasn't gonna happen. I think all I can do now is email LSI support.
January 7th, 2006, 07:20 PM #17
Did you try disabling all other option ROMs?
If the issue is that there isn't enough address space for all of the option ROMs, disabling the others should alleviate that.
January 7th, 2006, 07:26 PM #18
Yes, I disabled the PXE and set the Sil3114A function to disabled. I even updated the board's BIOS to F5 from F2. Still no luck.
In my copy of this thread over at HardForum, AreEss says the 1600 has an issue with the 8131: http://www.hardforum.com/showthread.php?t=999432
When I Google: megaraid 1600 8131, this first result is someone apparently experiencing the same incompatibility: http://mail-index.netbsd.org/port-am...3/23/0000.html
It's a shame, but it's not so bad in the long run as I was planning to use the MegaRAID elsewhere and the included SATA RAID on this board. But I would like to get the MegaRAID working here.
January 8th, 2006, 12:36 PM #19
- Join Date
- Oct 2001
- Augsburg, Germany
That problem - if it still exists - is hardly related to your problem at all. IOAPIC setup problems (which are either BIOS fsckups or lack of brain in the operating system core).
At the phase you are in - before actual OS boot - the system still operates on legacy PIC interrupt controllers, and anyway, things like SCSI or RAID init do not use interrupts at all yet.
You're not having a UMB space problem either, else you wouldn't even get to see the Ctrl-M prompt.
Still, this kind of problem is a system BIOS bug more often than not. Ask IWill and/or LSI - typically either of the two will have heard of it and provide suggestions.
January 8th, 2006, 01:47 PM #20
I do plan on emailing LSI, however I should clarify that on the problem (Gigabyte) system, I don't even get as far as Ctrl+M. The system simply sits on a blinking "Initializing SCSI Controllers" [Chip 0: Init][Chip 1: Init]. In the socket A system it works in, those Inits turn to Done, and then I can Ctrl+M or Ctrl+H, or just let Windows boot up.
Users Browsing this Thread
There are currently 1 users browsing this thread. (0 members and 1 guests)