Dell 3-Node AMD DCS6005

spyrule

Active Member
DCS 6125?

If thats a different model, then you'll need to find someone else with a 6125 where the BMC is working and be able to pull a copy of the BMC code. I have no idea how to do that. A DCS6005 I would guess is NOT compatible with DCS6125... (generally speaking, one BIOS from one hardware model is not compatible with even slightly different hardware BIOS models).
 

beckert

New Member
Oct 30, 2015
23
0
1
42
Drivers can tell me what you put in works for W2008R2 raid5.
Thinks


I have one of these DCS6105/DCS6005 servers and they have been great. I now have a need to upgrade from Windows 2008 R2 to Windows 2012 R2 and have searched the WEB for Dot Hill RaidCore VST 5100 drivers. I have tried everything that Tyan, Dot Hill, Dell and SuperMicro have for embedded Raid Controllers. The current 2008 R2 version is DriverVer = 09/16/2010,5.1.5.00020

ftp://ftp.tyan.com/drivers/Website%20Drivers/AMD%20-%20ATI/Storage/SATA%20RAID/Dot%20Hill/Windows/64-bit/SP5100_SATARAID_DotHill_v5.1.5_Windows_64.zip

ftp://ftp.tyan.com/drivers/Website%20Drivers/AMD%20-%20ATI/Storage/Management%20Utility/Dot%20Hill/Windows/SP5100_RAID-Management-Utility_DotHill_v5.1.5_Win_32-64.zip .

Has anybody figured out how to make the driver work for Server 2012 R2?

Thank you,

blitzsd
 

Thanos

New Member
Jun 16, 2014
16
0
1
54
Hi guys, does anyone have the details on the 30x30x10 fans that some of you (spyrule?) use for the chipset? What's the Voltage for example?
if you have a model you use, can you share a link of that? I'm kinda tired of node2 powering down sometimes (even at "normal" ambient temp!). So, I will definitely use the fans for that.
By the way, flashing the BIOS/BMC gave me access (web too!) to the BMC of the (2) nodes I've "lost" connection with IPMI. The nodes worked fine but I didn't have the option to remotely (via IPMI) turn them on/off. Had to go and push the button..
So, thank you for the postings and files for that as well (that's for you Spyrule!!)

Happy Holidays guys!
 

spyrule

Active Member
[QUOTE="...
By the way, flashing the BIOS/BMC gave me access (web too!) to the BMC of the (2) nodes I've "lost" connection with IPMI. The nodes worked fine but I didn't have the option to remotely (via IPMI) turn them on/off. Had to go and push the button..
So, thank you for the postings and files for that as well (that's for you Spyrule!!)

Happy Holidays guys![/QUOTE]

Hey no problem. That's is the point of community forums, isn't it? Share your solutions, and post your fixes, 'cause sometimes you need to ask for help, and then return that favour when you can.
 

StopDropRoll

New Member
Oct 29, 2014
13
5
3
50
Hey Guys,

I'm having a weird issue. We recently moved so I had to tear down my lab and transport it. Now the power buttons are behaving strangely. Only the first node will turn on, the other two wont. If I hold the power for the second node, it will shut off the entire rack (not sure if it did that before).

The NICs still light up though.

I thought that it may have been jostled during moving so I opened it up to take a look. Everything appeared to be in order, but I wasn't really sure what exactly I was looking for.

Any thoughts?
 

StopDropRoll

New Member
Oct 29, 2014
13
5
3
50
EDIT: Annnnnnd, it stopped working again... Lights are dead, even for the node that is running.

Thanks guys, you fixed it.

Posting this for the next person in this situation with a wife that is complaining that she can't watch her Downton Abbey Christmas Special on Plex.

Apparently all you need to do to resolve the issue is post question on the internet, because now it's behaving correctly... power buttons work and everything.

Thanks ServetheHome!
 
Last edited:

spyrule

Active Member
If after major moves, its always good to pull the cpu's and ram, re-seat them, and re-apply thermal compound. The TC has a tendency to dry out over time, and moving allows cracks and spaces to form between the heatsink and the cpu's, as well as ram sometimes shifts in its seating. This can cause hot spots and improper operation. This is especially true, if you move in cold weather. Its one of the reasons why many servers ship with the CPU's out of the case, and you have to install them on new purchases.
 

StopDropRoll

New Member
Oct 29, 2014
13
5
3
50
If after major moves, its always good to pull the cpu's and ram, re-seat them, and re-apply thermal compound. The TC has a tendency to dry out over time, and moving allows cracks and spaces to form between the heatsink and the cpu's, as well as ram sometimes shifts in its seating. This can cause hot spots and improper operation. This is especially true, if you move in cold weather. Its one of the reasons why many servers ship with the CPU's out of the case, and you have to install them on new purchases.
It's worth a shot. I ordered some thermal paste and I'll give it a try.

Thanks!
 

epiphanyplx

New Member
Jan 27, 2016
7
0
1
33
Hello all, first off this thread has been very helpful with my purchase of a DCS6005 a year or so ago, thank you.

Now, the problem...I am moving soon and the noise this thing generates is going to be a problem. I purchased some fans I saw recommended, the SAN ACE 80s. I put them in today, spliced wires correctly, they were all spinning, etc. I hook the server back up and 2 of them start booting ESXi. The other one just sits there, no video, eventually just starts beeping until I turn it off. It was a constant beep so no useful pattern. I also didn't see any lights blinking in any pattern.
I reseated the RAM, made sure I didn't knock any cables out of it, etc. I know some people had posted issues where replacement fan RPM was too low for the BMC so servers wouldn't start so I checked the IPMI and saw that it wasn't able to read the fans speed anymore. (Temperatures appear fine.)

It seems weird that this would only effect one of the sleds if it was due to unhappy BMC. While I had case open I also unplugged and removed the backplane to allow for better airflow as I am using Synology for datastores so have no drives in the server. This also hasn't seemed to effect the other two servers so don't know if it relevant.

Any ideas?

Thank you!
 
Last edited:

pricklypunter

Well-Known Member
Nov 10, 2015
1,606
470
83
Canada
Are you sure that all the fans are outputting a signal? Try swapping them about and see if the problem moves around with them, if it does, get a couple of extra fans and swap them out one at a time :)
 

epiphanyplx

New Member
Jan 27, 2016
7
0
1
33
Prickly,

Yes, at least I am sure all fans are blowing air. Is that what you mean? My understanding is the fans don't run from one of the servers but rather from the shared BMC.
When in the American Megatrends IPMI it only shows 2 CPU temps, however. Is there a separate IPMI for each server or just one for the whole box?
One of my CPUs is reaching temperature threshold in vSphere and triggering warning.

After work today I may take it all down again and replace heatsink compound for all CPUs to see if that helps at all.

Thanks
 

epiphanyplx

New Member
Jan 27, 2016
7
0
1
33
I have found the culprit; a loose screw. :(
Unfortunately, I believe I have fried one of the nodes.

Does anyone happen to have one for sale?


Also, after changing into the San Ace fans my temperature keeps hitting 80 degrees C and triggering alarm. Did anyone else have this problem? I put fresh heatsink compound on but no change.
In vSphere web client it tells me Memory Module SYS FAN 1-4 are going 2800-2850 RPMs.

Could I have miswired them and caused them to spin too slowly?
 
Last edited: