ASUS Z10PA-D8 - Dual Socket 2011-3
- Thread starter gb00s
- Start date
Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.
That's a known issue with these boards because the board needs to initialize 'new' (added) hardware correctly and that's not always going well. You could start to remove all but 1 ram stick from each socket. Be aware of using the correct ram socket for each stick. There's a table in the mainboard manual. Also, if you have one single ram stick with 'minor issues' you will not be able to boot with socket 1 populated, even with this ram stick populated related to socket 0 and working there.
If you need help let me know.
PS:
Start with socket 0 populated only and add 1 ram stick only. Start it, let it boot. Shutdown. Populate socket 1 and add one ram stick. start it, let it boot. Boot errors can be seen in the log checking the IPMI very easily. Even when you add a PCIe card, it's possible you have to go through the whole initialization process again. The least problems are coming from bios 3801 ... The most problems I had with bios 3202, even with the seller recommending it. Yes, the CPU's are running better with 3202, but it's due to the spectr/meltdown bios patches. Otherwise, I had lots of problems with bios 3202.
What ram are you using ... LRDIMM can cause issues as well. RDIMM should be fine. But once you get these mainboards running, they are playing nice.
If you need help let me know.
PS:
Start with socket 0 populated only and add 1 ram stick only. Start it, let it boot. Shutdown. Populate socket 1 and add one ram stick. start it, let it boot. Boot errors can be seen in the log checking the IPMI very easily. Even when you add a PCIe card, it's possible you have to go through the whole initialization process again. The least problems are coming from bios 3801 ... The most problems I had with bios 3202, even with the seller recommending it. Yes, the CPU's are running better with 3202, but it's due to the spectr/meltdown bios patches. Otherwise, I had lots of problems with bios 3202.
What ram are you using ... LRDIMM can cause issues as well. RDIMM should be fine. But once you get these mainboards running, they are playing nice.
Last edited:
Good day to you all,
I also bought the ASUS Z10PA-D8 from a well known ebay dealer and I am curious which BIOS version I will get and if it will run with a Xeon E5 2620 V3 and if an iKVM module will be included.
Maybe you could be so kind and tell me if it was always included?
I also bought the ASUS Z10PA-D8 from a well known ebay dealer and I am curious which BIOS version I will get and if it will run with a Xeon E5 2620 V3 and if an iKVM module will be included.
Maybe you could be so kind and tell me if it was always included?
I assume 3202
yes, yes and yes.
PS: That's my experience with the seller. Never heard anything else.
Last edited:
Just want to clear things up here as I receive a surprising amount of PMs regarding these boards.
I never said these are bad boards. I just had some issues documented here. Why I had to change all of my boards? I don’t know. Maybe I use different and uncommon hardware with the board. Use older used parts due to limited financial resources. Trying different ways to build a cheap but performant system? Maybe.
But, these used boards are 1/3 of the price of a comparable used Supermicro board. So if my calc is right you get 3:1. One of these 3 will always work. Otherwise, the seller always changed my boards with issues without any hassle or complain. All spare boards are working just fine till now.
Also, these boards were used 24/7 for 4 years in a render farm. If rumors are correct, more than 1500 boards are/were on sale. If I just count my boards and the boards with issues here, I’m not passing 50 boards. Not even close to that. So if <50 heavily used boards fail out of 1500, or even 1000, then the failure rate is not worth to be mentioned. I have all my boards running flawlessly now. All PCIe-slots populated. All set up for performance w/o any ‘power saving’ schnick schnack. Optimized bios settings for IO. Build-server for a linux distro compiling 24/7. 2x VM center (Proxmox cluster). 1x docker center. 1x F@H. 1x storage. All running in 3U down to 1U environments with its thermal challenges in a cabinet not far away from your bed (noise vs. thermals).They just run!
Again, it was never my attention to talk these boards down. Yes, they have design issues, bios issues and i believe power delivery issues. But if you buy 3:1 Supermicro and if you are unlucky with one, you have still 2 to tinker with. With a broken Supermicro board, you are out of the business for a moment.
Thanks.
I never said these are bad boards. I just had some issues documented here. Why I had to change all of my boards? I don’t know. Maybe I use different and uncommon hardware with the board. Use older used parts due to limited financial resources. Trying different ways to build a cheap but performant system? Maybe.
But, these used boards are 1/3 of the price of a comparable used Supermicro board. So if my calc is right you get 3:1. One of these 3 will always work. Otherwise, the seller always changed my boards with issues without any hassle or complain. All spare boards are working just fine till now.
Also, these boards were used 24/7 for 4 years in a render farm. If rumors are correct, more than 1500 boards are/were on sale. If I just count my boards and the boards with issues here, I’m not passing 50 boards. Not even close to that. So if <50 heavily used boards fail out of 1500, or even 1000, then the failure rate is not worth to be mentioned. I have all my boards running flawlessly now. All PCIe-slots populated. All set up for performance w/o any ‘power saving’ schnick schnack. Optimized bios settings for IO. Build-server for a linux distro compiling 24/7. 2x VM center (Proxmox cluster). 1x docker center. 1x F@H. 1x storage. All running in 3U down to 1U environments with its thermal challenges in a cabinet not far away from your bed (noise vs. thermals).They just run!
Again, it was never my attention to talk these boards down. Yes, they have design issues, bios issues and i believe power delivery issues. But if you buy 3:1 Supermicro and if you are unlucky with one, you have still 2 to tinker with. With a broken Supermicro board, you are out of the business for a moment.
Thanks.
Last edited:
Hello all,
Recently got the motherboard from eBay, came as revision 1.02 with BIOS 3202 and and ASMB8-iKVM. Flashed BIOS to 3801.
1) I don't know if it is a conspiracy but I can't find an I/O shield in Europe either. Dust will become an issue. I guess I will need to find and reuse an I/O shield from a different motherboard or order one without openings from China and wait several months for it to arrive :-/
2) There is no CD (e.g. for the iKVM) to be found anywhere (?) on the Internet. I am not sure if it is necessary or things can be done with the downloadable support drivers/utilities.
3) I am a newbie at this but I can't make the ASMB8-iKVM work. I tried entering static IPs in BIOS and using every LAN port, no ping. I downloaded the latest (version 1.16.1, 2021-01-06 !) firmware from Asus on a bootable FreeDOS USB stick but running flash*.bat returns "Please update the last version of BIOS". Running ASMC8.exe gives me
In BIOS->Server Mgmt I get:
Am I doing something wrong ?
Recently got the motherboard from eBay, came as revision 1.02 with BIOS 3202 and and ASMB8-iKVM. Flashed BIOS to 3801.
1) I don't know if it is a conspiracy but I can't find an I/O shield in Europe either. Dust will become an issue. I guess I will need to find and reuse an I/O shield from a different motherboard or order one without openings from China and wait several months for it to arrive :-/
2) There is no CD (e.g. for the iKVM) to be found anywhere (?) on the Internet. I am not sure if it is necessary or things can be done with the downloadable support drivers/utilities.
3) I am a newbie at this but I can't make the ASMB8-iKVM work. I tried entering static IPs in BIOS and using every LAN port, no ping. I downloaded the latest (version 1.16.1, 2021-01-06 !) firmware from Asus on a bootable FreeDOS USB stick but running flash*.bat returns "Please update the last version of BIOS". Running ASMC8.exe gives me
Code:
Detect Motherboard -> (Z10PA-D8 Series)
Can't find PCI SMIC base address
Can't find SMIC/KCS/BT interface for IPMI
Code:
BMC Self Test Status: PASSED
BMC Device ID: Unknown
BMC Device Revision: Unknown
BMC Firmware Revision: Unknown
IPMI Version: Unknown
Last edited:
Regarding #3 (iKVM),
OK after resitting the module and power cycling the computer several times I got it to work, then updated to firmware 1.16.1.
All that remains is to find a way to redirect video, all I get is black screen using either jviewer.jnlp or sol.jnlp.
OK after resitting the module and power cycling the computer several times I got it to work, then updated to firmware 1.16.1.
All that remains is to find a way to redirect video, all I get is black screen using either jviewer.jnlp or sol.jnlp.
Last edited:
hey guys, my mb just arrived (form ebay seller too, also with the ikvm..), about the cooler... there's been some suggestions on this thread, and I´m going for the cheap one.. the artic freezer 34 (for dual 2650L v4) anyone using them? found any stresses?... I haven't bought them yet.. so I´m open for more suggestions? some photos of your builds? 
thanks
thanksI went with Noctuas (NH-U9S) because I already had one sitting around, even though I am with mere 2630L v3 cpus for a start. Might think of upgrading in the future. Temperatures are not just about the coolers or the fan speed but also about the server's placement and air spacing inside the case. I managed to squeeze everything in an old ATX case (which by the way was very hard to find second hand as I needed the extra 5.25 slots for a RAID, I guess I got lucky) but at the same time I have the luxury of running my server outdoors (in the balcony), so I can compensate.
Hum.. I still need to test mine.. havent receive the cpu yet.. about the qvl list.. pardon my ignorance.. why is one of those modules highlighted in yellow?
@kubernation90 send me a pm I have to adjust my quick answer. If you see a 'b7' boot error at first and then solved it but the screen stays black, please try to get into the bios and change the graphic adapter option from external to internal. External graphic adapters are graphical PCIe cards. The internal is the onboard graphical adapter. I hope this helps. If I'm not mistaken, then the bios changes automatically to external graphic adapters. Please keep in mind they are coming from a render farm
Im wondering why the seller states in german that “voll funktionsfähiges Motherboard”
I know they want to get rid of them, but at least they should indicate that faulty boards and glitches with headache can happen....
I know they want to get rid of them, but at least they should indicate that faulty boards and glitches with headache can happen....
I assume they are working with their specific settings and we are kind of inexperienced with them and do some trial/error. Just the b7 error together with a black screen lets you assume they are not working. I you reset the bios and put the graphic adapter just to internal mode they might just work fine.
In 90% of all IT related issues, the real issue is always sitting in front of the 'problem'.
In 90% of all IT related issues, the real issue is always sitting in front of the 'problem'.
I found a way. For some reason it doesn't work reliably every time though, sometimes it just refuses to work without rebooting or power cycling but I can live with that.
Last edited:
I just noticed one thing: when turned off the server still consumes almost 10w, I don't know if it is the ASMB8-iKVM module but I would still like to try to remove it to see if anything changes on power consumption ... the question is: the card works well same if I take it off? and then: if I can remove it, does anyone know how to turn this card back on via wol? because at the moment I can only turn the server back on by passing via ASMB8-iKVM, if I send the classic magic packet nothing happens, both on the IP of the network card and on the IP of the service port ...
Hi all, we bought few of these boards a while ago (same seller) and 2 are in production with XCP-ng running 24/7.
One issue we had was with Shared_IKVM, with shared IPMI LAN, system sometimes system loose the connectivity with internal Intel I210AT LAN. When we disabled the Share_LAN for IPMI and used the dedicated IPMI LAN, system worked fine.
Now we tried to deploy Truenas with this board and where we encounter another issue.
TrueNas can get the dhcp ip lease (pfsense) but no other connectivity. Ping only work between pfsense and truenas but nothing else.
We even tried to connect the truenas to direct WAN, it get ip from ISP but still no ping/connectivity to e.g. google.com
IPMI GUI is accessable.
any idea what could be the issue or direct us to right direction.
One issue we had was with Shared_IKVM, with shared IPMI LAN, system sometimes system loose the connectivity with internal Intel I210AT LAN. When we disabled the Share_LAN for IPMI and used the dedicated IPMI LAN, system worked fine.
Now we tried to deploy Truenas with this board and where we encounter another issue.
TrueNas can get the dhcp ip lease (pfsense) but no other connectivity. Ping only work between pfsense and truenas but nothing else.
We even tried to connect the truenas to direct WAN, it get ip from ISP but still no ping/connectivity to e.g. google.com
IPMI GUI is accessable.
any idea what could be the issue or direct us to right direction.
Check your pfSense or TrueNAS config.
Here we have the IPMI interface in a management VLAN and can ping and connect to it from in-/outside. We also have this board as storage with TrueNAS and Proxmox and have no network issues. You could also add a PCIe network card and go from there if you think the network ports are broken. But then, why ping to pfSense is possible?
Has nothing to do with the hardware and this board.
Here we have the IPMI interface in a management VLAN and can ping and connect to it from in-/outside. We also have this board as storage with TrueNAS and Proxmox and have no network issues. You could also add a PCIe network card and go from there if you think the network ports are broken. But then, why ping to pfSense is possible?
Has nothing to do with the hardware and this board.
@gb00s we install the new i350 lan and it had same issue, then we knew some thing else is wrong, we already eliminated the pfsense issues, cables were already changed, thus the switch was the last variable. we were using an old netgear switch for testing and it turn out there were lagg turn on at 4 ports (old test configuration) and Asus Z10PA-D8 was connected to one of these ports. unlucky when we were changing the cables & ports, those ports were still these 1 of 4 lagg ports
yes, nothing wrong with the board just silly oversight
these board are out of stock now but we will buy few more for backup when and if they turn up again
all of our boards are in 2U chenbro chassis and using following heat sink coolers, but cooling is still not the best what wanted
2x Intel E62476-001 CNFN6022J1
2x Supermicro SNK-P0048AP4 with Noctua A6 fans
2x Noctua NH-L9x65
yes, nothing wrong with the board just silly oversight
these board are out of stock now but we will buy few more for backup when and if they turn up again
all of our boards are in 2U chenbro chassis and using following heat sink coolers, but cooling is still not the best what wanted
2x Intel E62476-001 CNFN6022J1
2x Supermicro SNK-P0048AP4 with Noctua A6 fans
2x Noctua NH-L9x65