For those wondering (or like mine, where a bunch of BIOS stuff was disabled for no good reason that I can tell.... F2 is how you get into the BIOS, but only after the BMC init process runs (and for me seems to fail).
Dell 3-Node AMD DCS6005
- Thread starter Xeppo
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
Anybody have a reason why I would be getting "BMC Not responding" on all 3 of my nodes?
Anybody else have this problem?
Anybody else have this problem?
Does anyone have any information on the jumpers available? The manuals for the 6105 does not match by a long shot. (specifically on the jumper that disables the BMC)
No, but I've asked this in this thread as well :
https://forums.servethehome.com/index.php?threads/dell-c6105-dcs-how-to-disable-bmc.4332/
more specifically, if someone can take pictures of all the jumpers on the mobo from a fully functional node. It would help immensely.
https://forums.servethehome.com/index.php?threads/dell-c6105-dcs-how-to-disable-bmc.4332/
more specifically, if someone can take pictures of all the jumpers on the mobo from a fully functional node. It would help immensely.
Last edited:
If your just installing some pcie cards, you only need to remove the blank sled, and sled #1, this gives you complete access to both lower sleds, so no need to disconnect at all.
For sake of reference, from the REAR, we'll call the empty sled #0, and then the top system sled #1, the one below that #2, and the sled on the bottom by the power supply #3 (clockwise from the top).
First, you need to power down all 3 sleds (these sleds are NOT hot removable, unlike newer models).
Pull the blank sled out (there is one small screws on the back thats mounted to the bottom of the pcie slot area) remove that screw, then push the the locking lever to the right (again facing the rear of the chassis), and pull. The blank sled should slide out.
Repeat the screw removal with all 3 other sleds.
Now, you need to :
1) disconnect the power cable going to the motherboard,
2) disconnect the SATA cables from the motherboard (take note of which cable goes where for easy reconnect, if that's your intention),
3) disconnect the small black/red two prong wire that goes up near the SATA connectors as well, but take note of where it connects (this is what turns each system on).
Now, push the rear leaver as before, and pull, it should slide out.
Reverse when installing.
Is there anybody in this thread, that can take some pictures of all their jumper positions on a node that fully works (ipmi/NIC @ 1gbps) ?
It would be awesomely appreciated!
Even if you can text them too me that would rock (PM me for my cell #)
It would be awesomely appreciated!
Even if you can text them too me that would rock (PM me for my cell #)
I'm interested in a couple of the DCS-6005 boxes, understanding the inherent limitations. Outside of the complete lack of support/documentation and sporadic NIC/BIOS issues, they seem to be pretty solid for a home virtualization lab. I do have one remaining concern, however. This has been hashed a bit, but I don't think anyone ever confirmed - are there Dot Hill RAID drivers out there for Win2K12R2? If not, it's pretty much a deal breaker for me, unfortunately.
Reading that Supermicro did adopt Dot Hill in 2014 for their A+ motherboards, and doing some quick Googling, I did find this: ftp://ftp.supermicro.com/driver/SATA/DotHill_AMD_SP5100/
Of course, it's the SP5100 instead of the VST5100 (looks like it's labeled rcraid vs. the old bcraid), but wanted to confirm that someone tried these -- lemieuski, Ken or blitzsd, perhaps?
Great thread, thanks everyone!
Reading that Supermicro did adopt Dot Hill in 2014 for their A+ motherboards, and doing some quick Googling, I did find this: ftp://ftp.supermicro.com/driver/SATA/DotHill_AMD_SP5100/
Of course, it's the SP5100 instead of the VST5100 (looks like it's labeled rcraid vs. the old bcraid), but wanted to confirm that someone tried these -- lemieuski, Ken or blitzsd, perhaps?
Great thread, thanks everyone!
I cant answer your driver question, but make sure your buying them from an ebay seller who has sold a bunch of them. Got mine from a private seller and ive had nothing but problems with the nics and ipmi plus some video issues with one node.
Thanks for the tip, spyrule. The ebay seller I'm looking at seems to have sold tons of these and was referred to in previous posts in this thread. While I'm not confident I won't have the same NICs and or BIOS/IPMI, the seller at least seems to be willing to swap out problematic nodes.
Here are some pictures I just took of one of my nodes. ipmi is working fine and Gigabit too.
The only thing I noticed is that the nodes randomly power on by themselves... Everything is off and a couple of days later one node powers on. I turn it off, then a few hours later another node powers on and so on... I have no idea what's causing this...
Let me know if you need anything else.
First off, HUGE thank you for taking the pictures. Greatly appreciated. Turns out one of my jumpers was backwards on all 3 nodes. Sadly, it seems to have made no difference.
As for your nodes turning on, sounds like WOL, and you've got magic packets floating around your network (coming from something). That would be my first guess. Second guess, would be a bad PSU, that's spiking, causing shorts, and making the controller think you've pushed a button. Another possible cause would be a dying FAN control circuit, or lastly... REALLY bad dust (do you smoke) ?
My nodes do the same, as far as powering themselves on. I even put them on a separate vlan / subnet and they still do the same thing. My guess is the NSA got some code in the firmware, and they have to boot up to talk back to the mothership.
I really should figure out a pattern or something, but oh well, many other things to do.
I've been struggling to get these servers to work.
I ordered a second chassis with just mobo/cpu's included, primarily to get ipmi functioning.
Unfortunately, 4 of the 6 cpus don't post, I then swapped my good cpu's into the newer motherboards, and at least ipmi works, only now what's happening is after they've been powered on for a 3-5 minutes I get a constant tone from each motherboard.
Now, I've temporarily swapped to some less noisy (and only 3 pin fans), so I'm wondering if maybe the ipmi is telling the motherboard that they cannot detect any fans.
Gonna swap back to stock fans tomorrow and see what happens.
I ordered a second chassis with just mobo/cpu's included, primarily to get ipmi functioning.
Unfortunately, 4 of the 6 cpus don't post, I then swapped my good cpu's into the newer motherboards, and at least ipmi works, only now what's happening is after they've been powered on for a 3-5 minutes I get a constant tone from each motherboard.
Now, I've temporarily swapped to some less noisy (and only 3 pin fans), so I'm wondering if maybe the ipmi is telling the motherboard that they cannot detect any fans.
Gonna swap back to stock fans tomorrow and see what happens.
Ok, this is annoying... but awesome at the same time.
Fixed my BMC error saying "BMC NOT RESPONDING"
So, while I was looking around for some command line options for KCSFLASH for pulling a copy of the existing BMC on a good/working sled, I came across a thread that had
mentioned that the user had gone to ASPEED's website (the BMC creator). I then downloaded the SOCFLASH tool from here : Driver Downloads：ASPEED 信驊科技
The trick is to download the LATEST version of the SOC tool. Install it on a bootable USB key. Also, if you havn't already, you MUST also flash your
BIOS to the same/matching version (the two files came together previously).
Also, while at it, copy the BMCBACK.ROM file from the link posted near the start of this thread. Once, I flashed that rom (which is a known good BMC copy), powered the sled off, rebooted, suddenly my BMC started to work perfectly again! Woohoo!
I'll be linking a copy of the BIOS/BMC and tools from my dropbox shortly, I just have to wait for a large file to finish syncing first.
Fixed my BMC error saying "BMC NOT RESPONDING"
So, while I was looking around for some command line options for KCSFLASH for pulling a copy of the existing BMC on a good/working sled, I came across a thread that had
mentioned that the user had gone to ASPEED's website (the BMC creator). I then downloaded the SOCFLASH tool from here : Driver Downloads：ASPEED 信驊科技
The trick is to download the LATEST version of the SOC tool. Install it on a bootable USB key. Also, if you havn't already, you MUST also flash your
BIOS to the same/matching version (the two files came together previously).
Also, while at it, copy the BMCBACK.ROM file from the link posted near the start of this thread. Once, I flashed that rom (which is a known good BMC copy), powered the sled off, rebooted, suddenly my BMC started to work perfectly again! Woohoo!
I'll be linking a copy of the BIOS/BMC and tools from my dropbox shortly, I just have to wait for a large file to finish syncing first.
Btw,
For anybody that has installed VMware ESXi 5.5U2, did you find that it takes FOREVER to get past the sb.v00 and s.v00 modules when loading?
For anybody that has installed VMware ESXi 5.5U2, did you find that it takes FOREVER to get past the sb.v00 and s.v00 modules when loading?
Sorry, as a quick update in case anybody needs this (this is in my public-Dropbox folder) here's a copy of the files needed to recover the Dell DCS C6005 BMC (where your getting "BMC not responding")
Get the BMC Recovery tools and BIOS/BMC .rom files
Get the BMC Recovery tools and BIOS/BMC .rom files
Small extra note for anybody trying to tame these dcs C6005 servers, if you replace the fans, order the Noctua NF-R8 regular fans, DO NOT bother with the pwm. If these fans go into lower rpm mode, they don't push enough air to cool this server. I pulled the pwm pin (4th pin) to get the fans to run full speed. By doing so, I was able to get the hottest sled #2, to run at CPU0 - 58', and CPU1 - 48', and the SR9050 - 85'c (ordered a 30x30x7mm fan to keep it cooler).
Also, if anybody needs (and for sake of record), you can adjust the lower critical fan sensor alarm with the following:
On a linux box (I used centos 7):
yum install ipmitool (let it install with defaults)
then run:
ipmitool -I lan -H {IP_ADDRESS_OF_BCM} -U %Username% -P %password% sensor thresh "SYS FAN 1" lcr 800
(this allows you to adjust the Lower CRitical to 800 rpm, adjust the 800 as you need).
This second command allows you to adjust the upper warning of the CPUs (should you need it):
ipmitool -I lan -H {IP_ADDRESS_OF_BCM} -U %Username% -P %password% sensor thresh "CPU0_TEMP" ucr 80
This set the Upper CRitical to 80'
Taken from another forum thread:
"The thresholds listed are, in order: lnr, lcr, lnc, unc, ucr, unr
These are acronyms for:
Lower Non-Recoverable
Lower Critical
Lower Non-Critical
Upper Non-Critical
Upper Critical
Upper Non-Recoverable"
In case you want to adjust any other settings.
The only thing I havn't figured out is how to adjust the logoff timer for the ipmi web interface (just for testing for the next few days, I'd like to keep them open, if anybody knows how to adjust this, that would be awesome!).
Also, if anybody needs (and for sake of record), you can adjust the lower critical fan sensor alarm with the following:
On a linux box (I used centos 7):
yum install ipmitool (let it install with defaults)
then run:
ipmitool -I lan -H {IP_ADDRESS_OF_BCM} -U %Username% -P %password% sensor thresh "SYS FAN 1" lcr 800
(this allows you to adjust the Lower CRitical to 800 rpm, adjust the 800 as you need).
This second command allows you to adjust the upper warning of the CPUs (should you need it):
ipmitool -I lan -H {IP_ADDRESS_OF_BCM} -U %Username% -P %password% sensor thresh "CPU0_TEMP" ucr 80
This set the Upper CRitical to 80'
Taken from another forum thread:
"The thresholds listed are, in order: lnr, lcr, lnc, unc, ucr, unr
These are acronyms for:
Lower Non-Recoverable
Lower Critical
Lower Non-Critical
Upper Non-Critical
Upper Critical
Upper Non-Recoverable"
In case you want to adjust any other settings.
The only thing I havn't figured out is how to adjust the logoff timer for the ipmi web interface (just for testing for the next few days, I'd like to keep them open, if anybody knows how to adjust this, that would be awesome!).
Ok, so it seems as though the above ipmi commands don't commit after a full shutdown/pull powercord (but not resetting the bmc, just pulling the powercord reset back to original values...)