Content in Intel Communities

S3420GPRX RMM3 Keyboard problem

webadmin@intel.com — Sat, 27 Apr 2013 19:43:13 GMT

The keyboard doesn't seem to work properly *some of the time* prior to the OS booting. i.e. I can't get into the BIOS setup or use our PXE boot menu. Sometimes however, it'll work fine and I can do both. It seems to work most of the time immediately after the AC cord was pulled. Power cycling or resetting from the web interface doesn't make a difference.

Once the OS has booted, then things work fine all the time.

We just got a few of these servers a few weeks ago and all of them have this problem, so I don't think it's a hardware failure... They're all running the latest firmware, the firmware update was performed after the RMM3 lite modules were installed. There are no IP address conflicts (nic5 is dedicated to the BMC) or other network issues, we're connecting to the web interface over LAN.

Any suggestions or ideas?

The 5000X Chipset and DIMM Ranks (RFI)

webadmin@intel.com — Fri, 26 Apr 2013 00:15:56 GMT

Several workstations including those from Apple (2006~2008) use the 5000X MCH.

This is a request for information (RFI) regarding the Apple implementation of the MCH in the MacPro1,1 and MacPro2,1 machines, how DIMM ranks work in general, and how DIMM ranks work in these particular systems.

The perplexities mount when attempting to select the proper DIMMs for these systems. My specific question is if quad rank DIMMs would work with the 5000X (Chipset containing the Memory Controller Hub "MCH") or not. By reading both user comments and Intel data sheets it remains a large and looming question. Intel's documentation specifically references only single and dual rank DIMMs. But AFAIK the way ranking works it's mostly about the total number of ranks and not so much the ranks per DIMM.

Intel® 5000X Chipset Memory Controller Hub (MCH) Datasheet

The MCH provides four channels and each channel can support up to 4 Dual Ranked DIMMs. That's 4x4x2 or 32 total ranks in 16 total DIMMs. Apple is only using two of the four channels (two sets of four - two 4-DIMM risers) for a total of 8 DIMMs as we all know. My uncertainty stems from the question of whether or not the channels have dual boundaries or not and the documentation doesn't specify.

In other words, if I'm only using 8 total DIMMs can I still use all 32 ranks, or would I still be limited to 8 ranks per channel? In more modern chipsets I believe it is the former and I could indeed use up all the ranks even on a single chip if I wanted - and if such memory existed. But on the 5000X I'm not sure.

Thanks!

Tesselator

S3200sh ipmi configuration

webadmin@intel.com — Thu, 18 Apr 2013 20:28:08 GMT

Has anyone configured ipmi lan interface on s3200sh. I was using http://osdir.com/ml/hardware.ipmitool.devel/2006-07/msg00019.html to configure s5000pal but this procedure doesn't apply to S3200sh.

Regards.

PS: I have posted this message a few minutes ago but i am not sure if have posted it or not so apalogize for double post if it happens

Intel Modular Server - MFSYS25

webadmin@intel.com — Tue, 26 Mar 2013 10:24:33 GMT

We have an Intel Modular Server and are looking to expand the storage by bolting on an additional unit. Has anyone got any recommendations as to what units work well.

Can any units be managed through the Intel Server management system or do they all have to be managed seperately?

Thanks

Lee

S5000PSL dual Xeon server board BIOS update hangs at 0% done

webadmin@intel.com — Wed, 27 Mar 2013 13:02:20 GMT

I need to update the BIOS on an S5000PSL server board.

I'm booting form FreeDOS via USB and running: BIOS101.BAT, Which warns me about a jumper, which I verified is correct, then extracts some stuff, then begins the BIOS update and hangs at:

Updating the system BIOS...0% Done

It says 3 min max. I left it for 1 hour. I tried 4 times, same thing.

Attached pic shows screen at hang.

Can't use the windows version as this is a Linux box. Can't use the "One-Boot Flash Update Utility" as that only supports red had and Suse and this is a Debian box.

Come on Intel!!! This shouldn't be so friggin difficult. All I want is for the $#!@#%# board to sleep, which is specifically fixed in the release notes for this firmware rev that I can't install.

I'm trying to install from the contents of S5000PSL_DOS_update_BIOS101_BMC68_FRUSDR48.zip

dmidecode reports the current version as: S5000.86B.07.00.0079.060520071732

The full dmidecode output is attached.

Please let me know if there's anything else I can provide.

-Bruce

Mobo S3210SH, which Raid controller to select, LSI or Intel Matrix

webadmin@intel.com — Wed, 20 Mar 2013 07:40:05 GMT

Intel doesn't make life easier. The mobo comes with 2 RAID controllers - which one to select, please help!

There will be RAID 1 (mirror, now 2x160GB SATA drives) in the beginning. SBS 2008 running Sharepoint Services, SQL Express, Exchange 2007, shared file folder for office documents.

All this for 15 people max. CPU now E8400, when needed we can upgrade to Xeon. Of LSI I know that it has a cute email alert option.

Previous Windows 2003 server died and so we are upgrading mobo and cpu and need a new clean new server 2008 installation. HDD's remain now from the previous server.

Which RAID is right for me?

Thanks!

S5500BC - Random shutdown & restart issue

webadmin@intel.com — Wed, 20 Mar 2013 01:01:34 GMT

S5500BC with 1 Xeon 5520, 16 GB RAM, ran fine for first 6+ months then started intermittent restarts. Win2008 Standard reliability log reports several 0x124 errors. RAM swapped & replaced 2 sticks at at time. Intel consistency check run. Microsoft mem checker run. MB replaced. All drivers, firmware & BIOS updated. SEL indicates power loss for several shutdowns but no indication of PS issues - 750 W PS. Microsoft tech support is unable to pin point either. Time of restarts cannot be correlated with any utilities or operations running on the server. Most incidences result in simple restart although most recent was an actual shutdown to power off. SEL file attached. Hope I'm not missing something obvious here! Any ideas out there??

Controller ID: 0 PD Reset: PD = :8, Error = 3, Path = 50:06:00:b0:01:39:06:c0

webadmin@intel.com — Wed, 20 Mar 2013 01:01:32 GMT

I have an Intel Server with a 5500 series motherboard and a SRCSATAWB raid card. I have the raid card setup to send me emails. Today i received this error and wanted to know how to decrypt this message: Controller ID: 0 PD Reset: PD = :8, Error = 3, Path = 50:06:00:b0:01:39:06:c0

S5520SC - No video, POST Code Diagnostic LEDs keep changing fast

webadmin@intel.com — Wed, 20 Mar 2013 00:59:56 GMT

Hello everyone,

I recently bought the intel S5520SC serverboard and have just installed everything.

As you can see in the title the problem is that I don't get any output, also the POST code

diagnostic LEDS keep changing in configuration very fast (like 5 different combinations/second).

I have two Q3QR ES CPU's that I'm using with the motherboard. These should supposedly have the

same specifications as the xeon x5675. I think the cause of the problem is the fact that my BIOS is

version 38 and that these CPU's need at least version 45. At the moment I have only one CPU connected,

as well as it's RAM. When I remove the RAM I get the three beeps that are often mentioned on these forums.

The reason that I created this post is because no one seems to have the fast changing diagnostic leds and

before I buy a xeon 55** to perform the BIOS update I would like to be as sure as possible that this is the problem.

I have already disconnected everything besides the RAM and CPU and I still get the above. I have already tried a different

RAM bar (with the same specs) and another RAM bar (with different specs) but this changes nothing.

I suppose the motherboard does a fast RAM check and afterwards tests the CPU on which it hangs in my situation?

Any insight in this situation will be greatly appreciated!

Thank you.

Kind regards,

Usui

SS4200-e failed raid

webadmin@intel.com — Tue, 19 Mar 2013 19:35:48 GMT

I have an ss4200-e which has marked all 4 drives as non-raid members with data requesting to be over written. Tech support was no help. wipe out the drives and restore from backup.

I have seen posts with this same issue, and that access is possibel and the data is OK.

So what is involved. I do not have an operational linux system but I can build one. Ubuntu seems to be the most discussed.

I have seen this type of issue with RAID sets on other controllers in the past and the trick was to define the array but not initialize. This is what I am hoping to do.

I like this post which implies that I should be able to possibly access the existing linux on the flash card.

http://communities.intel.com/message/68852

If so and I can find the correct commands I should beablt to telnet into the system define the array and without initializing the array see if the data comes up accessable. I know th board is using an ICH7 controller.

Any help on the linux specifics would be appreciated. My last major unix work was in the late 80's and early 90's. the last was an ocassional use of a SCO box a few years ago.

Thanks