Proxmox VE 8.1 released!

[nMax]

Will amd-pstate-epp be coming to proxmox kernels?

Kernel 6.5 is the version where it was made default but that does not seem to be the case in proxmox nor can I activate it by adding "amd_pstate=active" into grub.. Maybe there are some reasons the pstate epp driver should not be used in proxmox?

It would be really nice to get this supported to allow for more energy efficiency, or to squeze a little more performance of such amd cpus. In my testing the "epp performance performance" driver outpeforms "ahcpi performance" in peformance and slightly in efficiency and "powersave balance-performance" gets better single thread performance and higher efficiency yet but lower multicore (the rest just increase efficiency further at cost of performance), which seem to correlate with phoronix testing done on the different drivers.

Actually I just realized what the fault was, my proxmox boots with systemd-boot! - so of course adding kernel parameters in grub does not do anything!

I added amd_pstate=active into /etc/kernel/cmdline followed by proxmox-boot-tool refresh and a reboot and now everything works!

Sorry for the noise!

 

[spirit]

Actually I just realized what the fault was, my proxmox boots with systemd-boot! - so of course adding kernel parameters in grub does not do anything!

I added amd_pstate=active into /etc/kernel/cmdline followed by proxmox-boot-tool refresh and a reboot and now everything works!

Sorry for the noise!

what is your cpu model ? on my epyc v3 && v4, it's already enabled. I wonder if not only enabled by default on epyc and not ryzen for example.

 

[spirit]

what is your cpu model ? on my epyc v3 && v4, it's already enabled. I wonder if not only enabled by default on epyc and not ryzen for example.

after some research, it seem than cppc need to be enable in bios, to get amd_pstate=active be by default. ("lscpu|grep cppc" shoud find it is it's correctly enabled)

 

[DerDanilo]

Thanks for the many good features! Are the notification targets easy to extend? Gotify may be a good start but there are many projects were a ping to other (chat/notification) systems might be easier to setup.

To name are a few:
- healthcheck.io (also selfhosted)
- RocketChat
- Telegram
- Mattermost
- Slack
- ...

LibreNMS has quiet a few integrated for example. But others as well.
https://github.com/librenms/librenms/blob/master/doc/Alerting/Transports.md

It would be awesome if the list of supported notification targets was extended, as the list of acme backends is also quiet extensive.

Thanks in advance!

 

[t.lamprecht]

Are the notification targets easy to extend?

Not hard to extend, but there's a commitment for each type as we have to maintain that integration for a long time.

One of the next ones will probably be a more generic "HTTP-POST to user-specified URL", or webhook, target, with that users could proxy it to whatever they want without Proxmox VE having the burden of support for every possible target.

 

[DerDanilo]

Not hard to extend, but there's a commitment for each type as we have to maintain that integration for a long time.

One of the next ones will probably be a more generic "HTTP-POST to user-specified URL", or webhook, target, with that users could proxy it to whatever they want without Proxmox VE having the burden of support for every possible target.

That would be awesome since most integrations allow a simple webhook call.
Please allow to set a "secret" for the HTTP-POST/Webhook, that allows to define a variable, without having to expose the passwords/tokens in the UI. I guess that those would be stored in `/etc/pve/priv/notification/...` or something, right?

Supporting most publicly knows targets should not be an issue though, since they usually don't change their API out of the blue.

 

[Usernet]

UPDATE 8.0.9 -> 8.1.3
No more network access and by switching to the local console the boot freeze on this indication only:

loading Linux 6.5.11-4-pve ...

I rolled back the kernel.Any idea what happened?
And how did I manage to update it?

(Dell Optiplex 3090)
THANKS

 

[Haulman]

Found out that 8.1 does not allow do add empty cdrom, button "add" is grayed out until you select any iso image. Checked 7.4 and 8.0 - no problem.

 

[Dunuin]

Found out that 8.1 does not allow do add empty cdrom, button "add" is grayed out until you select any iso image. Checked 7.4 and 8.0 - no problem.

I can't confirm that. There is still the "do not use any media" option when adding a new CD ROM to an existing VM, editing an existing CD drive or creating a new VM. Which is the proper way to have a CD drive without an ISO.

 

[tconnors]

I'm trying to convert a host from grub legacy boot to uefi, and have enabled uefi only boot in the bios, but proxmox 8.1 won't boot. OK, so let's try a download of the 8.1 disk and run a rescue. Alas, I get:

error: no such device: rpool ERROR: unable to find boot disk automatically.

Is the 8.1 CD expected to be able to import a system that was rpool upgraded recently, perhaps as new as the 6.5 kernel with zfs 2.2.1?

 

[t.lamprecht]

Is the 8.1 CD expected to be able to import a system that was rpool upgraded recently, perhaps as new as the 6.5 kernel with zfs 2.2.1?

Yes, but the rescue boot is a very trivial method of trying to find a correct kernel/initrd combo.

You might be better off using a debug mode from the installer's grub menu, as the second stage of that resembles much more a live system that can be used to repair an existing one.

 

[tconnors]

Yes, but the rescue boot is a very trivial method of trying to find a correct kernel/initrd combo.

You might be better off using a debug mode from the installer's grub menu, as the second stage of that resembles much more a live system that can be used to repair an existing one.

I wasn't able to deal with the AZERTZ keyboard that the console in debug mode defaults to! zpool doesn't like to mount onto / unlike all other filesystems, so I overrode (with difficulty, because it was so hard to find a working '=' key!) the mountpoint, and chrooted into it, but something about proxmox-boot-tool still failed on me (I didn't go to the effort of mounting /sys and /proc etc because I broke the keyboard even further when trying to find a program to remap the keyboard). It's already tomorrow and I haven't eaten dinner yet, so I've booted my host back up and might try again later.

 

[Haulman]

I can't confirm that. There is still the "do not use any media" option when adding a new CD ROM to an existing VM, editing an existing CD drive or creating a new VM. Which is the proper way to have a CD drive without an ISO.

I wrote nothing about "do not use any media" option existance, i wrote about add button disabled, when you choose "Use physical CD/DVD" or "Do not use any media". Rechecked on fresh installation.

 

[Dunuin]

I wrote nothing about "do not use any media" option existance, i wrote about add button disabled, when you choose "Use physical CD/DVD" or "Do not use any media". Rechecked on fresh installation.

Ah ok. Yes, here that "Add" button is disabled too. Then it really should be a bug. You might want to add a bug tracker ticket at https://bugzilla.proxmox.com in case it's not already tracked there (and I can't see such a tricket).

 

[coNemniatEND]

I know gotify was added but doesn’t support iOS. Can you add support for ntfy.sh ( supports both)

 

[debitlagos]

UPDATE 8.0.9 -> 8.1.3
No more network access and by switching to the local console the boot freeze on this indication only:

loading Linux 6.5.11-4-pve ...

I rolled back the kernel.Any idea what happened?
And how did I manage to update it?

(Dell Optiplex 3090)
THANKS

Probably similar to the issue I have with my 6 3080's, the realtek NIC drivers. If you did the 8168 driver fix back in the day, then that's the culprit. If you remove the blacklist for the 8169 drivers, you should be back up and running. This is exactly what I did. I just nano'd /etc/modprobe.d/r8168-dkms.conf and #'d out the two lines and rebooted.

https://forum.proxmox.com/threads/a...3-network-not-work-anymore.136831/post-608874

 

[iprigger]

Probably similar to the issue I have with my 6 3080's, the realtek NIC drivers. If you did the 8168 driver fix back in the day, then that's the culprit. If you remove the blacklist for the 8169 drivers, you should be back up and running. This is exactly what I did. I just nano'd /etc/modprobe.d/r8168-dkms.conf and #'d out the two lines and rebooted.

https://forum.proxmox.com/threads/a...3-network-not-work-anymore.136831/post-608874

I doubt that... we have 10g intel nics in the servers...

Tobias

 

[Usernet]

Probably similar to the issue I have with my 6 3080's, the realtek NIC drivers. If you did the 8168 driver fix back in the day, then that's the culprit. If you remove the blacklist for the 8169 drivers, you should be back up and running. This is exactly what I did. I just nano'd /etc/modprobe.d/r8168-dkms.conf and #'d out the two lines and rebooted.

https://forum.proxmox.com/threads/a...3-network-not-work-anymore.136831/post-608874

THANKS
So it's still the same problem with Realtek drivers...
Again, again... I have 150km between the two machines.

So, I preferred to update the package
https://forum.proxmox.com/threads/proxmox-ve-8-1-released.136960/page-5#post-609522
@wolpi

Is it possible to add this repository DEFINITELY to check for updates during an apt update ?

 

[proteus]

Got this last night, lost connection to server. The server is mostly idle, one vm doing nothing.
Network is 2 x 1Gb/s i210 onboard.

Dec 04 03:03:30 PVE-02 kernel: igb 0000:07:00.0 eno1: PCIe link lost
Dec 04 03:03:30 PVE-02 kernel: ------------[ cut here ]------------
Dec 04 03:03:30 PVE-02 kernel: igb: Failed to read reg 0xc030!
Dec 04 03:03:30 PVE-02 kernel: WARNING: CPU: 24 PID: 2007428 at drivers/net/ethernet/intel/igb/igb_main.c:745 igb_rd32+0x93/0xb0 [igb]
Dec 04 03:03:30 PVE-02 kernel: Modules linked in: veth 8021q garp mrp ebtable_filter ebtables ip_set ip6table_raw iptable_raw ip6table_filter ip6_tables iptable_filter bpfilter nf_tables sunrpc bonding tls softdog nfnetlink_log nfnetlink binfmt_misc ipmi_ssif intel_rapl_msr intel_rapl_common amdgpu snd_sof_amd_rembrandt amd64_edac snd_sof_amd_renoir snd_sof_amd_acp edac_mce_amd snd_sof_pci snd_sof_xtensa_dsp snd_hda_codec_realtek snd_sof kvm_amd snd_hda_codec_generic snd_sof_utils ledtrig_audio snd_hda_codec_hdmi snd_soc_core kvm amdxcp snd_hda_intel iommu_v2 snd_compress snd_intel_dspcfg drm_buddy ac97_bus snd_intel_sdw_acpi gpu_sched snd_pcm_dmaengine drm_suballoc_helper snd_hda_codec snd_pci_ps irqbypass drm_ttm_helper crct10dif_pclmul snd_rpl_pci_acp6x snd_hda_core ttm snd_acp_pci polyval_clmulni polyval_generic snd_hwdep snd_pci_acp6x ghash_clmulni_intel drm_display_helper aesni_intel snd_pcm snd_pci_acp5x cec ast snd_rn_pci_acp3x crypto_simd snd_timer snd_acp_config rc_core cryptd acpi_ipmi drm_shmem_helper snd snd_soc_acpi
Dec 04 03:03:30 PVE-02 kernel: ipmi_si drm_kms_helper soundcore rapl snd_pci_acp3x ccp pcspkr ipmi_devintf k10temp joydev input_leds ipmi_msghandler mac_hid vhost_net vhost vhost_iotlb tap efi_pstore drm dmi_sysfs ip_tables x_tables autofs4 rndis_host cdc_ether usbnet mii hid_generic usbmouse usbhid hid zfs(PO) spl(O) btrfs blake2b_generic xor raid6_pq libcrc32c simplefb xhci_pci nvme xhci_pci_renesas crc32_pclmul igb ahci xhci_hcd nvme_core i2c_algo_bit i2c_piix4 libahci dca nvme_common video wmi
Dec 04 03:03:30 PVE-02 kernel: CPU: 24 PID: 2007428 Comm: kworker/24:2 Tainted: P O 6.5.11-6-pve #1
Dec 04 03:03:30 PVE-02 kernel: Hardware name: Supermicro AS -1015A-MT/H13SAE-MF, BIOS 1.1a 10/19/2023
Dec 04 03:03:30 PVE-02 kernel: Workqueue: events igb_watchdog_task [igb]
Dec 04 03:03:30 PVE-02 kernel: RIP: 0010:igb_rd32+0x93/0xb0 [igb]
Dec 04 03:03:30 PVE-02 kernel: Code: c7 c6 03 f4 56 c0 e8 1c bd 7f d7 48 8b bb 28 ff ff ff e8 a0 86 35 d7 84 c0 74 c1 44 89 e6 48 c7 c7 f8 00 57 c0 e8 dd fd ba d6 <0f> 0b eb ae b8 ff ff ff ff 31 d2 31 f6 31 ff e9 69 70 b4 d7 66 0f
Dec 04 03:03:30 PVE-02 kernel: RSP: 0018:ffffa97c01afbd98 EFLAGS: 00010246
Dec 04 03:03:30 PVE-02 kernel: RAX: 0000000000000000 RBX: ffff8fd90461cf18 RCX: 0000000000000000
Dec 04 03:03:30 PVE-02 kernel: RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
Dec 04 03:03:30 PVE-02 kernel: RBP: ffffa97c01afbda8 R08: 0000000000000000 R09: 0000000000000000
Dec 04 03:03:30 PVE-02 kernel: R10: 0000000000000000 R11: 0000000000000000 R12: 000000000000c030
Dec 04 03:03:30 PVE-02 kernel: R13: 0000000000000000 R14: 0000000000000000 R15: ffff8fd9170feb40
Dec 04 03:03:30 PVE-02 kernel: FS: 0000000000000000(0000) GS:ffff8ff798800000(0000) knlGS:0000000000000000
Dec 04 03:03:30 PVE-02 kernel: CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
Dec 04 03:03:30 PVE-02 kernel: CR2: 00007f5ba5afd0a0 CR3: 0000000d28034000 CR4: 0000000000750ee0
Dec 04 03:03:30 PVE-02 kernel: PKRU: 55555554
Dec 04 03:03:30 PVE-02 kernel: Call Trace:
Dec 04 03:03:30 PVE-02 kernel: <TASK>
Dec 04 03:03:30 PVE-02 kernel: ? show_regs+0x6d/0x80
Dec 04 03:03:30 PVE-02 kernel: ? __warn+0x89/0x160
Dec 04 03:03:30 PVE-02 kernel: ? igb_rd32+0x93/0xb0 [igb]
Dec 04 03:03:30 PVE-02 kernel: ? report_bug+0x17e/0x1b0
Dec 04 03:03:30 PVE-02 kernel: ? handle_bug+0x46/0x90
Dec 04 03:03:30 PVE-02 kernel: ? exc_invalid_op+0x18/0x80
Dec 04 03:03:30 PVE-02 kernel: ? asm_exc_invalid_op+0x1b/0x20
Dec 04 03:03:30 PVE-02 kernel: ? igb_rd32+0x93/0xb0 [igb]
Dec 04 03:03:30 PVE-02 kernel: ? igb_rd32+0x93/0xb0 [igb]
Dec 04 03:03:30 PVE-02 kernel: igb_update_stats+0x89/0x830 [igb]
Dec 04 03:03:30 PVE-02 kernel: igb_watchdog_task+0x12d/0x880 [igb]
Dec 04 03:03:30 PVE-02 kernel: process_one_work+0x23b/0x450
Dec 04 03:03:30 PVE-02 kernel: worker_thread+0x50/0x3f0
Dec 04 03:03:30 PVE-02 kernel: ? __pfx_worker_thread+0x10/0x10
Dec 04 03:03:30 PVE-02 kernel: kthread+0xef/0x120
Dec 04 03:03:30 PVE-02 kernel: ? __pfx_kthread+0x10/0x10
Dec 04 03:03:30 PVE-02 kernel: ret_from_fork+0x44/0x70
Dec 04 03:03:30 PVE-02 kernel: ? __pfx_kthread+0x10/0x10
Dec 04 03:03:30 PVE-02 kernel: ret_from_fork_asm+0x1b/0x30
Dec 04 03:03:30 PVE-02 kernel: </TASK>
Dec 04 03:03:30 PVE-02 kernel: ---[ end trace 0000000000000000 ]---

 

[Dark26]

I use proxmox on my laptop at work ( hp elitebook), and i think since the new version( kernel related ?) , my vm is crashing, like one time per day in average. Before no problem at all.

On the next crash, i will boot on the 6.2 kernel to see if it's the problem...