mikeyosm
-
Posts
651 -
Joined
-
Last visited
Content Type
Profiles
Forums
Downloads
Store
Gallery
Posts posted by mikeyosm
-
-
On 2/19/2023 at 11:27 PM, Shivaess said:
[EDIT: Turning off secure boot fixed my problem immediately.]
Apologies for replying to a dead topic, but figured this would leave the most continuity of the discussion. I am attempting to boot unraid on ESXi 8 and having issues with the usb not being found by vmware efi on boot. Interestingly when I boot into the bios and go to configure boot options -> add a boot option the first entry listed is an Unraid MBR. so obviously it can see the usb stick. What am I missing here?
Not sure if you ever fixed this but since you attached a USB 3.1 controller, the USB drive needs to be 3.0/3.1 otherwise it wont boot. I removed the 3.1 controller, added a USB 2.0 controller and booted off of my USB 2.0 thumb drive, no issue.
-
16 hours ago, blacklight said:
I got the W680-M Ace SE with the IMPI card actually
it has a better pcie layout in my opinion. Unraid works without a problem.
Specs:i9 - 13900KS
128ECC RAM (have to look up the brand if you need it - was quite a pain to find ECC -> its recognized even in VMs /Singe bit ECC)
PCI Layout:
- PCI x1: IPMI Card
- M2 (CPU): empty for future NVME
- PCIe x8 (gen5 - bifur): Quadro K4200 (for VM testing)
- PCIe x8 (gen5 - bifur): RTx3060TI (for VM testing)
- PCIe x4 (gen3): Broadcom 9300-16i (HBA) with 14 hard drives attached
- PCIe x4 (gen3): Mellanox ConnectX-3 Pro for max 20Gbit
- 2x M2 (x4 - chipset): optane 900P (for NAS testing, using a adapter M2 to fullsize PCIe)
For now everything is recognized and booting Unraid was a charm, unfortunately I have problems with virtualizing a NAS, but for now it seems like a software problem. I will update the post if I find a incompatibility (IOMMU problems or similiar). The IOMMU layout seems solid btw, no suprises there for now, just that the two onboard SATA groups (1: 4x sata, 2: SLIM-SAS which has a PCIe mode) are in the same group, but I already thought that this would be the case.
And also the IPMI is a nice bonus in the combo package. Its a little bit expensive compared to a custom one (have no experience there) but it just works.
I am currently on another continent for a year and the access and controlling power over VPN with the IPMI card is a real relieve.
I attached a screenshot of the IOMMU groups, because I had trouble finding them online. Hope they will be useful.
Let me know if you need more information
Great news, thanks for the update. Do you mind showing a screenshot of the complete IPMI sensors please? I would like to see what sensors are detected.
-
On 7/1/2023 at 9:22 PM, NathanR said:
So I don't know why people do what they do.
What I can say is it is unlikely to matter what the mobo says it can support TDP wise.Rationale for the above statement:
1) 125W TDP is just a classification, real metrics will change depending on workload
2) Average power draw will be ~114w. Sauce: https://www.techpowerup.com/review/intel-core-i9-13900k/22.html
3) 8-pin CPU connector can deliver 235w so the mobo isn't power limited. Sauce: https://www.overclock.net/threads/gpu-and-cpu-power-connections.1773088/
4) Power Components will work until their thermal limits are exceeded. Meaning the mosfets will deliver more current (power) than they are designed for until they reach thermal limiting. Sauce: https://www.infineon.com/dgdl/ir3550.pdf?fileId=5546d462533600a4015355cd7c831761 (note, this is a generic VRM, I do not know what VRM the X13 uses)
4A) servers usually have forced air cooling...thus the thermal headroom on the heatsink is very little because it is expected there will be hundreds of CFM moving across the mobo... Home server will cause thermal bottle-necks.
5) Bench/Stress peak power ratings are often meaningless, (albeit fun, interesting, useful to find the limits) for what we would typically use the 13900k For (server, docker, VM, storage)
X13SAE-F mini-review/writeup
https://www.servethehome.com/supermicro-x13sae-f-intel-w680-motherboard-mini-review/3/
Thank you! This is awesome. You probably saved me 20+hrs of work.
I am considering moving my 5900X build to intel for two reasons.
1) iGPU for PLEX transcoding (my parents, wife-parents, sister, etc. all need things transcoded; only direct play is in-house to my NVIDIA shield)
2) lower power consumption (more efficient processors for dockers & server VMs)
Build I'm considering:
- Mobo: MBD-X13SAE-F-O
- 64GB: MEM-DR532MD-EU48
- 13700k or 13900K
- Re-use everything else from my 5900X build
Did you end up getting the Supermicro board? Thinking of parting with my cash on this or the Asus W680-M Ace SE
-
15 minutes ago, SimonF said:
Released 2023.10.06 with IPMI Lan Version option.
Thanks for implementing this so quickly, great work!
-
7 minutes ago, SimonF said:
ipmi-sensors -h myip -u myuser -p mypass -l operator -vv -D LAN_2_0
_D LAN_2_0 is the same as lanplus I think. Will look to see if that can be specified in the plugin.
Yes, that works! Please implement so all of us Asrock owners who could never get this working over LAN, now can.
-
@SimonF success with ipmitool - had to use the following command line...
How do I get this working in the plugin?
root@TEST:~# ipmitool -I lanplus -H 192.168.1.x -U admin -P test sel list
1 | 08/06/2023 | 13:51:01 | Event Logging Disabled #0xcf | Log area reset/cleared | Asserted
2 | Pre-Init |0000003872| Voltage #0x0b | Lower Critical going low | Asserted
3 | Pre-Init |0000003877| Voltage #0x0b | Lower Non-recoverable going low | Asserted
4 | Pre-Init |0000003878| Voltage #0x0b | Lower Non-recoverable going low | Deasserted
5 | Pre-Init |0000003880| Voltage #0x0b | Lower Non-recoverable going low | Asserted
6 | Pre-Init |0000003881| Voltage #0x0b | Lower Non-recoverable going low | Deasserted
7 | Pre-Init |0000003884| Voltage #0x0b | Lower Non-recoverable going low | Asserted
8 | Pre-Init |0000003885| Voltage #0x0b | Lower Non-recoverable going low | Deasserted
9 | Pre-Init |0000003886| Voltage #0x0b | Lower Non-recoverable going low | Asserted
a | Pre-Init |0000003888| Voltage #0x0b | Lower Non-recoverable going low | Deasserted -
57 minutes ago, SimonF said:
I think this is going to be related to you specific hardware. I believe some newer Dell servers dont allow access of the IPMI via IDRAC.
There may be an option in settings or ask the vendor if it is possible.
Can you get info from the host OS?
Yep host OS works perfect, just network doesnt work. Asrock asked me to try ipmitool instead.
root@TEST:~# ipmitool -H 192.168.1.x -U admin -P test -A PASSWORD
Authentication type PASSWORD not supported
Error: Unable to establish LAN session
Error: Unable to establish IPMI v1.5 / RMCP session
Tried all of the password types, same error every time.
-
On 10/3/2023 at 9:00 AM, mikeyosm said:
Sure.
I get -
ipmi-sensors: authentication type unavailable for attempted privilege level
@SimonF any ideas on what could be wrong?
-
Yeah, considering the same. Been messing with ESXi too which has been faultless with vgpu and my p4, too much of a resource hog tho to run vcenter.
-
With debug enabled --
192.168.x: =====================================================
192.168.x: IPMI 1.5 Get Channel Authentication Capabilities Request
192.168.x: =====================================================
192.168.x: RMCP Header:
192.168.x: ------------
192.168.x: [ 6h] = version[ 8b]
192.168.x: [ 0h] = reserved[ 8b]
192.168.x: [ FFh] = sequence_number[ 8b]
192.168.x: [ 7h] = message_class.class[ 5b]
192.168.x: [ 0h] = message_class.reserved[ 2b]
192.168.x: [ 0h] = message_class.ack[ 1b]
192.168.x: IPMI Session Header:
192.168.x: --------------------
192.168.x: [ 0h] = authentication_type[ 8b]
192.168.x: [ 0h] = session_sequence_number[32b]
192.168.x: [ 0h] = session_id[32b]
192.168.x: [ 9h] = ipmi_msg_len[ 8b]
192.168.x: IPMI Message Header:
192.168.x: --------------------
192.168.x: [ 20h] = rs_addr[ 8b]
192.168.x: [ 0h] = rs_lun[ 2b]
192.168.x: [ 6h] = net_fn[ 6b]
192.168.x: [ C8h] = checksum1[ 8b]
192.168.x: [ 81h] = rq_addr[ 8b]
192.168.x: [ 0h] = rq_lun[ 2b]
192.168.x: [ 23h] = rq_seq[ 6b]
192.168.x: IPMI Command Data:
192.168.x: ------------------
192.168.x: [ 38h] = cmd[ 8b]
192.168.x: [ Eh] = channel_number[ 4b]
192.168.x: [ 0h] = reserved1[ 3b]
192.168.x: [ 0h] = get_ipmi_v2.0_extended_data[ 1b]
192.168.x: [ 3h] = maximum_privilege_level[ 4b]
192.168.x: [ 0h] = reserved2[ 4b]
192.168.x: IPMI Trailer:
192.168.x: --------------
192.168.x: [ AAh] = checksum2[ 8b]
192.168.x: =====================================================
192.168.x: IPMI 1.5 Get Channel Authentication Capabilities Response
192.168.x: =====================================================
192.168.x: RMCP Header:
192.168.x: ------------
192.168.x: [ 6h] = version[ 8b]
192.168.x: [ 0h] = reserved[ 8b]
192.168.x: [ FFh] = sequence_number[ 8b]
192.168.x: [ 7h] = message_class.class[ 5b]
192.168.x: [ 0h] = message_class.reserved[ 2b]
192.168.x: [ 0h] = message_class.ack[ 1b]
192.168.x: IPMI Session Header:
192.168.x: --------------------
192.168.x: [ 0h] = authentication_type[ 8b]
192.168.x: [ 0h] = session_sequence_number[32b]
192.168.x: [ 0h] = session_id[32b]
192.168.x: [ 10h] = ipmi_msg_len[ 8b]
192.168.x: IPMI Message Header:
192.168.x: --------------------
192.168.x: [ 81h] = rq_addr[ 8b]
192.168.x: [ 0h] = rq_lun[ 2b]
192.168.x: [ 7h] = net_fn[ 6b]
192.168.x: [ 63h] = checksum1[ 8b]
192.168.x: [ 20h] = rs_addr[ 8b]
192.168.x: [ 0h] = rs_lun[ 2b]
192.168.x: [ 23h] = rq_seq[ 6b]
192.168.x: IPMI Command Data:
192.168.x: ------------------
192.168.x: [ 38h] = cmd[ 8b]
192.168.x: [ 0h] = comp_code[ 8b]
192.168.x: [ 1h] = channel_number[ 8b]
192.168.x: [ 0h] = authentication_type.none[ 1b]
192.168.x: [ 0h] = authentication_type.md2[ 1b]
192.168.x: [ 0h] = authentication_type.md5[ 1b]
192.168.x: [ 0h] = authentication_type.reserved1[ 1b]
192.168.x: [ 0h] = authentication_type.straight_password_key[ 1b]
192.168.x: [ 0h] = authentication_type.oem_prop[ 1b]
192.168.x: [ 0h] = authentication_type.reserved2[ 1b]
192.168.x: [ 0h] = authentication_type.ipmi_v2.0_extended_capabilities_available[ 1b]
192.168.x: [ 0h] = authentication_status.anonymous_login[ 1b]
192.168.x: [ 0h] = authentication_status.null_username[ 1b]
192.168.x: [ 1h] = authentication_status.non_null_username[ 1b]
192.168.x: [ 0h] = authentication_status.user_level_authentication[ 1b]
192.168.x: [ 1h] = authentication_status.per_message_authentication[ 1b]
192.168.x: [ 0h] = authentication_status.k_g[ 1b]
192.168.x: [ 0h] = authentication_status.reserved[ 2b]
192.168.x: [ 0h] = channel_supports_ipmi_v1.5_connections[ 1b]
192.168.x: [ 0h] = channel_supports_ipmi_v2.0_connections[ 1b]
192.168.x: [ 0h] = reserved[ 6b]
192.168.x: [ C1D6h] = oem_id[24b]
192.168.x: [ 0h] = oem_auxiliary_data[ 8b]
192.168.x: IPMI Trailer:
192.168.x: --------------
192.168.x: [ 70h] = checksum2[ 8b]
192.168.x: ipmi-sensors: authentication type unavailable for attempted privilege level
-
11 hours ago, SimonF said:
Try this command, you will need to use your IP etc
/usr/sbin/ipmi-sensors --output-sensor-thresholds --comma-separated-output --output-sensor-state --no-header-output --interpret-oem-data --always-prefix -h '192.168.x.x' -u 'ADMIN' -p 'PASSWORD' --session-timeout=5000 --retransmission-timeout=1000
Sure.
I get -
ipmi-sensors: authentication type unavailable for attempted privilege level
-
@SimonF I am getting connection failed with my Asrock Rack x570D5U mobo. I am running UNRAID inside a VM and wanting to display my Asrock IPMI stats in the dashboard. What am I doing wrong?
-
plugin: installing: nvidia-vgpu-driver.plg Executing hook script: pre_plugin_checks plugin: downloading: nvidia-vgpu-driver.plg ... done plugin: downloading: nvidia-vgpu-driver-2022.09.01.txz ... done +============================================================================== | Installing new package /boot/config/plugins/nvidia-vgpu-driver/nvidia-vgpu-driver-2022.09.01.txz +============================================================================== Verifying package nvidia-vgpu-driver-2022.09.01.txz. Installing package nvidia-vgpu-driver-2022.09.01.txz: PACKAGE DESCRIPTION: Package nvidia-vgpu-driver-2022.09.01.txz installed. -----ERROR - ERROR - ERROR - ERROR - ERROR - ERROR - ERROR - ERROR - ERROR------ ---Can't get latest Nvidia vgpu driver version and found no installed local driver--- plugin: run failed: '/bin/bash' returned 1 Executing hook script: post_plugin_checks
-
Is this project still alive? Would be good to have the latest linux vgpu driver 535.104.06 compiled for 6.12.4.
-
2 hours ago, ich777 said:
Yes, did you enable the service/module like it is mentioned in the description from the plugin?
Sorted, thanks.
-
1
-
-
@ich777 is corefreq supported on 6.12.4? Plugin installed successfully but getting the following error.
:~# corefreq-cli
Daemon connection error code 2
corefreq-ro-shm: 'No such file or directory' @ line 21469 -
7 hours ago, ich777 said:
But this is expected, you basically take the GPU from the Host and pass it to the VM.
What is the benefit from seeing it on the Dashboard? I assume you don‘t use it for anything else then the VM…?
Don‘t know if that‘s easy doable but I think it can also cause issues for other users depending how it’s implemented.
I often switch the gpu between windows and other OS's. On occasions the GPU is bound to the host so I want the dashboard to update accordingly and not have N/A or false stats. Granted, it's more of a cosmetic issue but stil...
-
10 hours ago, ich777 said:
Sorry but this seems like a niche problem...
Wouldn't it be better to not configure it in the GPU Statistics plugin so that it even don't show up on the Dashboard when it's passed through?
No. When the VM is powered down and the GPU is released back to the host, the dashboard re-detects the card and updates, this I like.
-
2 hours ago, SimonF said:
The HDD value is calculated based om the highest drive temp of all drives.
Oh I see. Is it possible to show more than one drive?
-
I might have found a bug in the plugin.
I have 3 x HDDs in total ( 1 x SSD and 2 x NVME)
The plugin settings only shows 1 HDD Temperature that I am able to select to add to the footer/dashboard, I expect to select 3?
-
When a GPU is passed though to a VM the GPU dashboard shows false information. Any chance @SimonF you can modify the behaviour of the dashboard to show "N/A Device Passed through" or similar so it looks better?
-
Beadapanel support would be really welcome too. These screens connect via USB.
-
Using 6.12.4 and keep getting this in the log every every minute or so.
I have an AMD 6600XT connected to the PCIE4.0 X16 slot and a 7900X CPU.
Using radeon top/gpu stats plugin. dGPU and iGPU are not passed through or stubbed.
Any ideas why UNRAID keeps logging this?
----
Sep 23 16:47:12 MOUNRAID01 kernel: [drm] PCIE GART of 512M enabled (table at 0x00000081FEB00000).
Sep 23 16:47:12 MOUNRAID01 kernel: [drm] PSP is resuming...
Sep 23 16:47:12 MOUNRAID01 kernel: [drm] reserve 0xa00000 from 0x81fd000000 for PSP TMR
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: RAS: optional ras ta ucode is not available
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: SMU is resuming...
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: smu driver if version = 0x0000000f, smu fw if version = 0x00000013, smu fw program = 0, version = 0x003b2b00 (59.43.0)
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: SMU driver if version not matched
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: use vbios provided pptable
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: SMU is resumed successfully!
Sep 23 16:47:13 MOUNRAID01 kernel: [drm] DMUB hardware initialized: version=0x0202001E
Sep 23 16:47:13 MOUNRAID01 kernel: [drm] kiq ring mec 2 pipe 1 q 0
Sep 23 16:47:13 MOUNRAID01 kernel: [drm] VCN decode and encode initialized successfully(under DPG Mode).
Sep 23 16:47:13 MOUNRAID01 kernel: [drm] JPEG decode initialized successfully.
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 5 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 6 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 7 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 8 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 9 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 10 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring kiq_2.1.0 uses VM inv eng 11 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 1
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 1
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 1
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: amdgpu: ring jpeg_dec uses VM inv eng 5 on hub 1
Sep 23 16:47:13 MOUNRAID01 kernel: amdgpu 0000:03:00.0: [drm] Cannot find any crtc or sizes
-
Is this working with 6.12 and latest UNRAID stable release?
it has a better pcie layout in my opinion. Unraid works without a problem.
unRAID plugin for iGPU SR-IOV support
in VM Engine (KVM)
Posted · Edited by mikeyosm
I just built a second UNRAID server with a low power 11900K (bit of a contradiction I know) and an Asrock Z490D4U-2lT2 motherboard. I want to split the iGPU between Plex for transcoding and one other VM for some very light remote gaming/emulation.
Is the SRV IO plugin going to enable me to achieve this set up and is there anything I need to be aware of config wise to get this to work?
Just installed the plugin and it says my 11900 is not supported or may be in VF mode. I thought the SRV IO plugin worked with the 11900k and what is VF mode?