Notice Installation on HPE Proliant ML30 Gen10

[itgiec]

Good morning to all the community, I want to comment on a problem I have with a new server that I am working with I share their characteristics.

System Name : HPE ProLiant ML30 Gen10 Plus

System ROM : U61 v2.00 (02/01/2024) (Last Version Released by HP)

Redundant System ROM : U61 v1.60 (07/14/2022)

Boot Mode : UEFI Mode (Secure Boot)

Processor 1 : Intel(R) Xeon(R) E-2314 CPU @ 2.80GHz

Total DIMM Capacity : 32 GB

I tried to install the latest version of Proxmox (8.1-2) and I got the following warning.



Then I checked the BIOS and indeed I have Intel VT and VT-d enabled by default.



Maybe it is an incompatibility of the latest version of PROXMOX or the HP ROM? What do you suggest to check or perform?

Thank you very much to all.

 

[zzz09700]

From my experience width recent HP servers, there's a rather *** thing called "Workload Profile" that could mess with your BIOS settings - like disabling your virtualization support without you even knowing.

https://support.hpe.com/hpesc/public/docDisplay?docId=a00016408en_us

That's why most of our servers are Dell. Well I'll be honest we have had enough with Dell's *** and are moving to Supermicro.

 

[itgiec]

From my experience width recent HP servers, there's a rather *** thing called "Workload Profile" that could mess with your BIOS settings - like disabling your virtualization support without you even knowing.

https://support.hpe.com/hpesc/public/docDisplay?docId=a00016408en_us

That's why most of our servers are Dell. Well I'll be honest we have had enough with Dell's *** and are moving to Supermicro.

Hi, thanks for the suggestion I made the change to the "General Power Efficient Compute" and the message persists. I will try installversion 7.4.1 to see what happens and let you know.

 

[itgiec]

I have done the installation of Proxmox 8.1.4 and I could see in the log before starting Proxmox the following errors which from what I understand are security-driven issues in the BIOS if I am wrong please your help is always well appreciated.




I would greatly appreciate your guidance and/or comments regarding this case.

Best Regards.

 

[itgiec]

Please find the following log:

:~# dmesg --level=err,warn
[ 0.031681] Warning: node 0 [mem 0x00000000-0x77efffff] overlaps with itself [mem 0x00100000-0x1bd07fff]
[ 0.151655] x86/cpu: VMX (outside TXT) disabled by BIOS
[ 0.151661] x86/cpu: SGX virtualization disabled due to lack of VMX.
[ 0.151663] x86/cpu: SGX Launch Control is locked. Disable SGX.
[ 0.180854] Invalid PCCT: 0 PCC subspaces
[ 0.180854] ENERGY_PERF_BIAS: Set to 'normal', was 'performance'
[ 0.209457] sysfs: cannot create duplicate filename '/firmware/acpi/tables/data/BERT'
[ 0.209459] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 6.5.11-8-pve #1
[ 0.209461] Hardware name: HPE ProLiant ML30 Gen10 Plus/ProLiant ML30 Gen10 Plus, BIOS U61 02/01/2024
[ 0.209462] Call Trace:
[ 0.209463] <TASK>
[ 0.209465] dump_stack_lvl+0x48/0x70
[ 0.209469] dump_stack+0x10/0x20
[ 0.209471] sysfs_warn_dup+0x5b/0x80
[ 0.209474] sysfs_add_bin_file_mode_ns+0xc0/0xf0
[ 0.209475] sysfs_create_bin_file+0x6d/0xb0
[ 0.209477] acpi_bert_data_init+0x37/0x60
[ 0.209479] acpi_sysfs_init+0x20c/0x270
[ 0.209480] acpi_init+0x19b/0x590
[ 0.209483] ? __pfx_acpi_init+0x10/0x10
[ 0.209485] do_one_initcall+0x5b/0x340
[ 0.209488] kernel_init_freeable+0x31d/0x480
[ 0.209490] ? __pfx_kernel_init+0x10/0x10
[ 0.209492] kernel_init+0x1b/0x200
[ 0.209494] ret_from_fork+0x44/0x70
[ 0.209496] ? __pfx_kernel_init+0x10/0x10
[ 0.209497] ret_from_fork_asm+0x1b/0x30
[ 0.209500] </TASK>
[ 0.961728] i8042: Can't read CTR while initializing i8042
[ 0.961734] i8042: probe of i8042 failed with error -5
[ 0.965634] device-mapper: core: CONFIG_IMA_DISABLE_HTABLE is disabled. Duplicate IMA measurements will not be recorded in the IMA log.
[ 0.965732] platform eisa.0: EISA: Cannot allocate resource for mainboard
[ 0.965733] platform eisa.0: Cannot allocate resource for EISA slot 1
[ 0.965734] platform eisa.0: Cannot allocate resource for EISA slot 2
[ 0.965735] platform eisa.0: Cannot allocate resource for EISA slot 3
[ 0.965735] platform eisa.0: Cannot allocate resource for EISA slot 4
[ 0.965736] platform eisa.0: Cannot allocate resource for EISA slot 5
[ 0.965737] platform eisa.0: Cannot allocate resource for EISA slot 6
[ 0.965737] platform eisa.0: Cannot allocate resource for EISA slot 7
[ 0.965738] platform eisa.0: Cannot allocate resource for EISA slot 8
[ 1.380342] i801_smbus 0000:00:1f.4: Transaction timeout
[ 1.588341] i801_smbus 0000:00:1f.4: Transaction timeout
[ 1.792335] i801_smbus 0000:00:1f.4: Transaction timeout
[ 1.996341] i801_smbus 0000:00:1f.4: Transaction timeout
[ 3.170154] spl: loading out-of-tree module taints kernel.
[ 3.208776] zfs: module license 'CDDL' taints kernel.
[ 3.208779] Disabling lock debugging due to kernel taint
[ 3.208793] zfs: module license taints kernel.
[ 7.691551] ERST: [Firmware Warn]: too many record IDs!
[ 8.788986] pstore: backend 'erst' already in use: ignoring 'efi_pstore'
[ 9.701878] power_meter ACPI000D:00: Ignoring unsafe software power cap!
[ 9.701880] power_meter ACPI000D:00: hwmon_device_register() is deprecated. Please convert the driver to use hwmon_device_register_with_info().
[ 10.052840] ipmi_si IPI0001:00: The BMC does not support clearing the recv irq bit, compensating, but the BMC needs to be fixed.
[ 14.632755] kauditd_printk_skb: 9 callbacks suppressed
[ 28.624711] kvm[1079]: memfd_create() called without MFD_EXEC or MFD_NOEXEC_SEAL set

 

[zzz09700]

Hi, thanks for the suggestion I made the change to the "General Power Efficient Compute" and the message persists. I will try installversion 7.4.1 to see what happens and let you know.

From what I gathered you need one of the modes with the word "Virtualization" on its name.
The ez way to test it is by using an install ISO of Ubuntu. Instead of going through the normal install process, you should select the option "try Ubuntu without installing". After you get into the system use this command
`lscpu | grep "Virtualization"`

There you can see if KVM is detected by Ubuntu. I would assume the detection result is the same as in PVE.

 

[gfngfn256]

I made the change to the "General Power Efficient Compute" and the message persists.

On this site - I see 2 possible options either "Virtualization - Power Efficient" or "Virtualization - Max Performance" that you may want to choose.

I do not have an HPE server - so that's all I have for you.

 

[itgiec]

Hello everyone, I will tell you what I did and I solved the messages.

1) Reset the BIOS settings to factory.
2) Activate SGX and SGX Launch Control in Unlocked.
3) Leave the default Workprofile.

I no longer have the above-mentioned errors:

# dmesg --level=err,warn
[ 0.034850] Warning: node 0 [mem 0x00000000-0x77efffff] overlaps with itself [mem 0x00100000-0x1bd07fff]
[ 0.183917] Invalid PCCT: 0 PCC subspaces
[ 0.183917] ENERGY_PERF_BIAS: Set to 'normal', was 'performance'
[ 0.940268] i8042: Can't read CTR while initializing i8042
[ 0.940275] i8042: probe of i8042 failed with error -5
[ 0.944240] device-mapper: core: CONFIG_IMA_DISABLE_HTABLE is disabled. Duplicate IMA measurements will not be recorded in the IMA log.
[ 0.944303] platform eisa.0: EISA: Cannot allocate resource for mainboard
[ 0.944305] platform eisa.0: Cannot allocate resource for EISA slot 1
[ 0.944306] platform eisa.0: Cannot allocate resource for EISA slot 2
[ 0.944307] platform eisa.0: Cannot allocate resource for EISA slot 3
[ 0.944307] platform eisa.0: Cannot allocate resource for EISA slot 4
[ 0.944308] platform eisa.0: Cannot allocate resource for EISA slot 5
[ 0.944309] platform eisa.0: Cannot allocate resource for EISA slot 6
[ 0.944310] platform eisa.0: Cannot allocate resource for EISA slot 7
[ 0.944310] platform eisa.0: Cannot allocate resource for EISA slot 8
[ 1.428331] i801_smbus 0000:00:1f.4: Transaction timeout
[ 1.632330] i801_smbus 0000:00:1f.4: Transaction timeout
[ 1.836331] i801_smbus 0000:00:1f.4: Transaction timeout
[ 2.040339] i801_smbus 0000:00:1f.4: Transaction timeout
[ 3.137319] spl: loading out-of-tree module taints kernel.
[ 3.175697] zfs: module license 'CDDL' taints kernel.
[ 3.175699] Disabling lock debugging due to kernel taint
[ 3.175714] zfs: module license taints kernel.
[ 6.807379] ERST: [Firmware Warn]: too many record IDs!
[ 7.760979] pstore: backend 'erst' already in use: ignoring 'efi_pstore'
[ 8.349571] power_meter ACPI000D:00: Ignoring unsafe software power cap!
[ 8.349573] power_meter ACPI000D:00: hwmon_device_register() is deprecated. Please convert the driver to use hwmon_device_register_with_info().
[ 8.577894] ipmi_si IPI0001:00: The BMC does not support clearing the recv irq bit, compensating, but the BMC needs to be fixed.
[ 13.088181] kauditd_printk_skb: 9 callbacks suppressed
[ 27.413649] kvm[1089]: memfd_create() called without MFD_EXEC or MFD_NOEXEC_SEAL set

I think trying to put some security to the BIOS and the processor a configuration conflicted with the kernel.

Thank you all for your attention.