-
#1
Hi, this is my first post connected with new hardware.
I’m starting to build up a server.
I have such a devices:
MB: SuperMicro X11SSL-F
CPU: e3-1230v6
RAM: Samsung 4x16GB 2400 UDIMM ECC.
I had build up the PC, run it and I hear 5 short beeps and one long.
Then I entered to the IMPI and I see following Hardware list.
which is completely different, because Now I have v6 cpu and only one Ram on the board.
Here is stated bios 2.0c, but on the other site there is an 2.1a
On monitor I see something like that:
0x09 mrc data time centering 1d error
What should I do?
-
#2
- place one stick of ram in slot DIMMA1
- reset your ipmi back to factory default
- reboot the system
-
#3
did it,
now there is other issue
the system works only with 2 memory slots occupied — so only 32GB ram.
If I put 3.rd or 4th ram stick the system can not boot.
I get such error exactly:
Code:
0x09=mrcDataTimeCentering1DErr
I can only occupy the 2 first slots from the top (black and blue)
-
#5
Are all four of the ram modules part#M391A2K43BB1-CRC ???
Test each stick (one at a time) in the first ram slot (if one of the
sticks fails to boot) you may have a bad stick of ram.
-
#6
Disable the jumpers of SMB to PCI Slots (JI2C1/JI2C2) — from 1-2 to 2-3 and all will be fine
DrKK
FreeNAS Generalissimo
-
#7
Disable the jumpers of SMB to PCI Slots (JI2C1/JI2C2) — from 1-2 to 2-3 and all will be fine
really. What does that do?
-
#8
This is due to hardware limitation as the result of SMBus address conflict between DIMMA1 and the PCIe devices
-
#9
This is due to hardware limitation as the result of SMBus address conflict between DIMMA1 and the PCIe devices
How do you arrive at that conclusion?
-
#10
my system (X11-SSM) had same problems with 4pcs of 16GB modules (beeps at startup, 0x09=mrcDataTimeCentering1DErr, etc). I got information from Supermicro support about hardware limitations for SSM/SSL/SSH mainboards. And after setting jumpers to disable SMBus — system can start normally, w/o beeps and with all 64GB.
Very easy to check, but i don’t know where is TS — yacenty
Last edited: May 13, 2018
-
#11
Interesting. I assume those DIMMs are not on the QVL?
-
#12
yes ) HPE with SK Hynix chips
-
#13
my system (X11-SSM) had same problems with 4pcs of 16GB modules (beeps at startup, 0x09=mrcDataTimeCentering1DErr, etc). I got information from Supermicro support about hardware limitations for SSM/SSL/SSH mainboards. And after setting jumpers to disable SMBus — system can start normally, w/o beeps and with all 64GB.
Very easy to check, but i don’t know where is TS — yacenty
RSA- I was having this exact issue whenever I tried putting in 4DIMMS… your post saved me a ton of wasted time and heartache!!! Thank you!!!!
-
#14
RSA- I was having this exact issue whenever I tried putting in 4DIMMS… your post saved me a ton of wasted time and heartache!!! Thank you!!!!
i’m really happy to help you
|
1 / 1 / 0 Регистрация: 10.10.2019 Сообщений: 9 |
|
|
1 |
|
|
18.03.2021, 16:20. Показов 11588. Ответов 11
Доброго времени суток. Есть сервер со следующей конфигурацией:
Проблема заключается в том,что при включении сервера на экране появляется логотип SuperMicro на белом фоне,внизу слева следуют надписи: Код PEI—IPMI Initialization... PEI—SB Initialization... (появляется на миллисекунду) PEI—Intel MRC Execution... (справа код 2F) После этого ничего не происходит. В чём может быть проблема?
__________________
0 |
|
4448 / 2135 / 432 Регистрация: 09.09.2017 Сообщений: 7,742 |
|
|
18.03.2021, 16:38 |
2 |
|
I_FarEvil_I, Не поддерживает мат. плата регистровую (буферизированную) память. Для проца и мат. платы заявлена только поддержка ECC-памяти (с коррекцией ошибок).
0 |
|
1 / 1 / 0 Регистрация: 10.10.2019 Сообщений: 9 |
|
|
18.03.2021, 16:39 [ТС] |
3 |
|
ibarg, то есть нужно идти покупать UDIMM?
0 |
|
4448 / 2135 / 432 Регистрация: 09.09.2017 Сообщений: 7,742 |
|
|
18.03.2021, 16:41 |
4 |
|
покупать UDIMM? По ходу да.
0 |
|
1 / 1 / 0 Регистрация: 10.10.2019 Сообщений: 9 |
|
|
18.03.2021, 16:47 [ТС] |
5 |
|
Хорошо, спасибо за информацию. В ближайшее время куплю UDIMM память, если будет подобное-напишу.
0 |
|
1376 / 508 / 99 Регистрация: 11.03.2015 Сообщений: 1,735 |
|
|
19.03.2021, 13:18 |
6 |
|
I_FarEvil_I,
1 |
|
0 / 0 / 0 Регистрация: 15.06.2021 Сообщений: 3 |
|
|
15.06.2021, 09:36 |
7 |
|
В итоге, с какой именно запустилось?
0 |
|
1 / 1 / 0 Регистрация: 10.10.2019 Сообщений: 9 |
|
|
15.06.2021, 12:08 [ТС] |
8 |
|
dompro, UDIMM ECC DDR4, у меня — Samsung M391A4G43MB1-CTDQY
1 |
|
0 / 0 / 0 Регистрация: 15.06.2021 Сообщений: 3 |
|
|
24.06.2021, 12:35 |
9 |
|
Попробовал Kingston KSM32ES8/8ME , та же ошибка. Это на 3200 частоты, но подстраивается под 2666. Но похоже плате нужно жестко конкретно 2666?
0 |
|
1 / 1 / 0 Регистрация: 10.10.2019 Сообщений: 9 |
|
|
24.06.2021, 13:23 [ТС] |
10 |
|
плате нужно жестко конкретно 2666? Да.
0 |
|
0 / 0 / 0 Регистрация: 15.06.2021 Сообщений: 3 |
|
|
06.07.2021, 13:17 |
11 |
|
Какая-то хрень, опять ждал пока доедет другая плашка, купил Samsung [M391A1K43BB2-CTD] 8 ГБ, поставил и то же самое. Может дело не в памяти? Уже двадцать раз пожалел что эту материнку выбрал.
0 |
|
1376 / 508 / 99 Регистрация: 11.03.2015 Сообщений: 1,735 |
|
|
12.07.2021, 13:01 |
12 |
|
dompro, Error code 2F means memory initialization. It is generated thru the BIOS. You may have memory issue. Please check your memory DIMMs and motherboard’s DIMM sockets. Скачайте мануал от материнки с этой страницы:
0 |
|
IT_Exp Эксперт 87844 / 49110 / 22898 Регистрация: 17.06.2006 Сообщений: 92,604 |
12.07.2021, 13:01 |
|
12 |
I recently built a new server which I’ve been testing successfully without peripherals. I picked up a PERC H200 that I intend to flash to IT mode (but is currently in IR mode) that I intended to use for my HBA. Unfortunately, when I install the card, the system refuses to boot.
The relevant components of the system in question are:
-
Supermicro X11SSM-F (BIOS 2.0a)
-
Xeon E3-1240 v6
-
Kingston ValueRAM 16GB 2133MHz DDR4 ECC x4
Before failing to POST, it beeps 5 shorts and 1 long, indicating «No memory detected in system». It also displays «6=mrcReadMPRErr» on the screen.
I’ve tried reseating the card and tried both CPU and PCH PCI-E slots. I don’t want to try it in my existing production server, but I tried it in two desktop systems. The newer system gives the same result, a POST error code that indicates that no memory is installed. The older system (still has PCI slots) POSTs fine and I can go into the PERC configuration utility without issue. It shows MPT Firmware Revision 2.15.63.00-IR and Package Version 7.01.33.00.
I’d appreciate any input on how I should proceed.
Update 1
I flashed the device to IT mode but, unfortunately, it had no effect on the POSTing problem.
Update 2
The H310 tape mod actually worked, well mostly. It allowed the system to POST:
All is not perfect, however, it takes 3 minutes from the LSI copyright notice before the Initializing… & Ctrl-C prompts show, and then another 2 minutes after hitting Ctrl-C for the configuration utility to show. (And not just on the first boot after flashing.) I know these will be gone once I flash to IT mode, but I worry that this card might have problems. I’m unsure if I should trust it. Any thoughts?
Update 3
I’ve decided to try to return it to the seller and ordered another one from elsewhere. It just seems to be a problematic card. Thanks for the input all.
Update 4
A new H200 from a different seller arrived, and so far it’s doing the same thing! I’m really perplexed. I’m going to try the H310 tape mod and see what happens.
Update 5
With the H310 tape mod and flashed to IT mode P20, the device and connected drives are showing up in ESXi. I’m going to hook up the remaining drives and breakout cable and then setup the PCI passthrough and begin burn-in testing.
I’m not certain I trust this situation yet and I also worry a bit about the longevity of the tape mod, but I don’t have a lot of options unless I want to keep buying cards. I’ll report back one way or the other.
Update 6
I completed a full badblocks burn-in and SMART reported no errors. So I think the H200 + H310 tape mod + IT mode flash is working okay.