He IDE to satas adaptors I have seen for cheap (I'm not gonna sped like £10 in one) only support up to 160gb. I think that I am gonna try and get the card sent back as not even a Linux live cd can Format it and in the meantime I can put the drive in my main pc, share it and get the sever to use it as a share since they are bOth on at the same time usually.
Yes, It's still running There has been no memory errors logged too. Apart from the one few days ago. Tqh - Is it possible/how could Josh get the HardwareID and VendorID of the sata card in Live Linux CD (the string that looks like this PCI\VEN_8086&DEV_3A44&CC_0604)? I could have a look if these IDs are actually written inside the *.inf file of the driver. I think I had to manually append them to the inf file and then it would detect my drives.
That is good news. That memory error you got a few days ago doesn't really count since you were running your new DDR333 at DDR400. Please report back after you've had 1-2 weeks of up time. As for Josh's issue, it seems either the card is bad or the SI 3112 firmware has a bug. I've used a bunch of different SI controllers on various on-board and PCI[-X] configurations. I've seen SI 3124 PCI-X card's firmware fail to recognize some basic WD 200GB SATA drives. Recently I also had my trusty SI 3512 PCI card have problems with a WD VelociRaptor 300GB. I could install CentOS Linux on there, but after it rebooted after install, the drive started giving the kernel errors with timeouts or something. The same card works fine with WD Black 320GB and 500GB drives. I guess at some point I should try to contact Rosewill and have them work it out with Silicon Image (since they don't provide end-user support directly)...but I'll try going back one firmware revision since their latest was targeting Windows 7 only. Or it could just be a DMA problem between the controller and the specific hard drive.
TQH - I haven't swapped my RAM sticks yet. (Slowest one to bank 0). I decided to wait and see if it freezes and if it does I'll do it. Instead I rebooted the shut down the system and started it again. In fact I may shut down it today and plug power off and then start it again... If cpuz says RAM freq is 167MHz, does it mean it's running at DDR333? (Why is 333 double 166?, is it because it's a dual channel memory?) EDIT: Rebooted the server twice in a row and I had 1 memory error logged... Should I worry? It's just one... Out of 5 reboots I had 3 memory errors, they all occured somewhen before I was able to log in to the server... I guess RAM is heavily used at startup, that why
TQ, MD and all just a quick post to confirm my server hasn't failed since my last post a couple of days from now will be 2 months running without issues. Used daily as a data store, serves 4 exchange email accounts and webserver hosting my remote desktop support gateway/server. I have not changed anything since last posts. Still no raid, but it works well.. All the best guys LJ
@tqhoang was it possible to make a powernow mod for my processors? Cause it sure is noisy! But I am having loads of problems with it and it is getting REALLY frustrating and I am so close to throwing it out of the window. I have nearly completely gave up on getting a data hard drive to work so I have put it in my pc so I now have a raid 10 yay... But I digress, I tried Ubuntu server and installing that was a bitch but figured out both DVD drives were messed up so I pulled one out of an old pc that was in my shed, Tehe.
Josh, I don't know if that what was meant to happen but in my case PowerNow mod did not reduce any of the fans. I had to install SpeedFan as a service using the SrvAny.exe app from M$ to make it start when the server boots. The fans slow down according to my configuration to 15% so they are really quiet. They only speed up when CPUs are working more than usual and generate more heat. My cpus temps are usually at 30-35 C when idle and running at 1GHz. When the server is doing continouous work then CPUs' temps rise to about 40-45% and that's when SpeedFan increases fans to 70% I think. THQ: The server's still running LJ: Enjoy the server. Hope mine stops freezing too XD
Josh - There is already a BIOS on the first page for the HDAMA-G v2.18 with the PowerNow for Opteron 280 OSP280FAA6CB. Do you not have a way to flash the BIOS from DOS? Try what mateuszd said about controlling the fans...I should probably do the same myself.
LJ - mateuszd has been using the Opteron 250 E4 PowerNow BIOS for well over a week now. I think we figured out that his lockup problem was due to a memory mis-match (had DDR266 instead of the minimum DDR333 or DDR400).
Hello, I have recently acquired a nice storage server with a HDAMA-I motherboard. To solve installation issues I had with Server 2008 Re2, I installed the 1.23 bios (which was already on the board) with the 5.5.0.0 ROM (I had 5.4.something). Server installed without any problems after this. The system was stable for a day or two, however, the past few days I've been having random freezes. When I check the Event log I see a lot of "event 47"'s around the time the system probably stopped responding: "A correct hardware error has occurred. Component: Memory" I have since installed BIOS "HDAMA-I BIOS v1.22 + Silicon Image 3114 ROM v5.5.0.0 + ACPI FACP Fix", and I'm still seeing the occasional event 47, but haven't had a hang up yet in the past hour. I also ran Memtest for a full day when the machine arrived, without any errors. Is there anything else I can check or do? My full specs are: Arima HDAMA-I (with 1.22) 2x Optereon 270 (revision JH9-E6) 4x 1GB 400MHz (clocked at 333) 300GB Maxtor connected to SI3114 (for OS) 3Ware 9550SX-12 with 5x 2TB WD Green I saw that a few pages back somebody had the same issues with a rev-G board, but he fixed it by removing some memory that was overclocked? Not really sure about that.
FWIW, I have the v1.23 + SI ROM v5.5.0.0 (SATA-RAID or SATA-Link) on the first page. You should probably stick with those since that BIOS is reported to have the ACPI PowerNow P-States support built-into the BIOS. Check the SPD of your memory sticks with CPU-Z. Make sure they all are the same or at least DDR333 or faster. Also if they have different timings, put the slower ones in bank0.
That was the BIOS I was using when I succesfully installed Server'08. I'm still getting 47's every minute or so. So I now removed all memory modules except for the one in slot 1 of cpu 1. So far no 47 errors (up time 8 hours). All 4 of my modules are PC3200 CL3 ECC. The all have the same SPD content: 2.5,3,3,7 for 333Mhz and 3,3,3,8 for 400MHz Update: Found a bend pin on processor 2, bend it back (more or less) and reinserted the processor. So far no 47 errors.
HI Tqh. Just a quick report (Up for 8 days and no errors yet) . @knirfie - I had the overclocked memory problem. I fixed it by removing slow memory blocks as mentioned earlier. The system has been running ok since that time with no memory errors (apart from one occuring during bootup, not always though).
Greeting tqhoang & all the other experts, Thanks for sharing your Know how with others. It was an interesting day, to follow all the information. Unfortunately I still have not found what I am looking for... I own two Rackable Systems Chassis (Model C2004), build in is an Arima RioWorks Motherboard Rev G, two Opteron 275, 16GB RAM and a Phantom IV management card with Lantronix xport. No FDD, no optical drive, nothing on the PCI bus. Installed is HDAMA-G BIOS v2.18B (original from Rackable Systems). I boot VMWare ESXi 4.1 from a single 500GB 2.5" Hitachi HDD, attached to the internal SATA controller. If it runs, it runs, but I have two other problems causing pain and I hope you guys have a hint for me. I read the whole threat, learned a lot but still have the same problems on both systems. 1st: If I power up the system it needs 15 - 20 tries to receive a video signal and it seems that the OS is not booting as the Broadcom NICs do not answer PINGs. 2nd: If I have the luck to receive a video signal it can occur, that the system is switching off itself within the first 2 - 3 minutes. If I pass these traps, system runs and runs without any further issue. It makes me suspicious that both systems show the same effect. I assume the switch off effect has nothing to do with the OS, because it also happened when I left the system showing up the BIOS Menu. Any help is appreciated, hope you have a hint for me as this is very frustrating. Thanks! Regards, Tom
It sure sounds like your CPU's are overheating. Remove and clean your CPU's and heatsinks (rubbing with alcohol) and re-apply fresh thermal material. Personally I like Artic Silver Ceramique because it cleans up easily, but anything is probably better than what might be dried up under your heatsinks.
Hi all, No unfortunately I still have the same problems, wondering about this "no video" problem. UPDATE: They company who sold the systems to me highlighted in a mail, that I have to switch to "Customized" in the power savings options in the BIOS. I tried a couple of time and it seems to work. No "power offs" anymore. Now I left with this no video problem which is bad enough. Regards, Tom