Proxmox host ACPI problem

[michu]

Hi,

I'm using Proxmox 1.6 on HP DL320 G6. ACPI is enabled on mainboard and detected by system (/var/log/dmesg).

I installed acpid daemon and acpi-support packages.
It seems that Proxmox host doesn't create file /proc/acpi/event which is used by acpid, so this last can't read acpi events like pushing power button.

I have some Ubuntu 10.04 quests on Proxmox host, they all run with 2.6.32 too and acpid is working fine there (/proc/acpi/event exists).
My pveversion output:

pve01:~# pveversion -v
pve-manager: 1.6-2 (pve-manager/1.6/5087)
running kernel: 2.6.32-3-pve
proxmox-ve-2.6.32: 1.6-14
pve-kernel-2.6.32-3-pve: 2.6.32-14
qemu-server: 1.1-18
pve-firmware: 1.0-7
libpve-storage-perl: 1.0-13
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-7
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.12.5-1
ksm-control-daemon: 1.0-4

I was googling a little and found one thread on Debian forums, maybe this will be helpful:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=462467

Regards,

Michu

 

[tom]

You need to install the acpid from backports. add the following to your /etc/apt/sources.list:

# Lenny backports
deb [url]http://www.backports.org/debian[/url] lenny-backports main contrib non-free

then run:

aptitude update
aptitude install -t lenny-backports acpid acpi-support

 

[michu]

Hi,

Thank you for your reply.

Unfortunately I will not risk with backported packages The server is in production with many quest machines... I think its better to wait for next Proxmox based on Debian Squeeze.


Regards,

Michu

 

[JustaGuy]

Is it still recommended to use acpid from backports since last week's updates to PVE stable?

I notice in my nested PVEs, both 1.6-5121-4 & 1.6-5261-4, that the acpid from backports is ignoring ACPI events, as if it were not installed.
In either nested PVE, acpid from the stable repo functions properly.

The file /proc/acpi/event exists in all 3 PVEs, in the 2 nested PVEs the file remians present after purging acpid.

Although the output of pveversion -v differs somewhat, the physical PVE host is the same as the first nested PVE, both were installed from the 1.6-5121 iso (when it was new), are safe-upgraded to current, and share the same config files except network, storage & grub.

Physical PVE installed from 1.6-5121-4:

Bascule:~# apt-show-versions acpid
acpid/lenny-backports uptodate 1.0.10-5~bpo50+1
Bascule:~# pveversion -v
pve-manager: 1.6-5 (pve-manager/1.6/5261)
running kernel: 2.6.35-1-pve
pve-kernel-2.6.32-4-pve: 2.6.32-24
pve-kernel-2.6.35-1-pve: 2.6.35-6
pve-kernel-2.6.18-4-pve: 2.6.18-8
qemu-server: 1.1-22
pve-firmware: 1.0-9
libpve-storage-perl: 1.0-14
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-8
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
Bascule:~# apt-show-versions pve-qemu-kvm
pve-qemu-kvm/lenny uptodate 0.12.5-2
Bascule:~# apt-show-versions ksm-control-daemon
ksm-control-daemon/lenny uptodate 1.0-4
Bascule:~#

Nested PVE installed from 1.6-5121-4:

UARS:~# apt-show-versions acpid
acpid/lenny-backports uptodate 1.0.10-5~bpo50+1
UARS:~# pveversion -v
pve-manager: 1.6-5 (pve-manager/1.6/5261)
running kernel: 2.6.35-1-pve
proxmox-ve-2.6.35: 1.6-6
pve-kernel-2.6.32-4-pve: 2.6.32-24
pve-kernel-2.6.35-1-pve: 2.6.35-6
qemu-server: 1.1-22
pve-firmware: 1.0-9
libpve-storage-perl: 1.0-14
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-8
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.12.5-2
ksm-control-daemon: 1.0-4
UARS:~#

Nested PVE installed from 1.6-5261-4:

SOFIA:~# apt-show-versions acpid
acpid/lenny uptodate 1.0.8-1lenny2
SOFIA:~# pveversion -v
pve-manager: 1.6-5 (pve-manager/1.6/5261)
running kernel: 2.6.35-1-pve
proxmox-ve-2.6.32: 1.6-24
pve-kernel-2.6.32-4-pve: 2.6.32-24
pve-kernel-2.6.35-1-pve: 2.6.35-6
qemu-server: 1.1-22
pve-firmware: 1.0-9
libpve-storage-perl: 1.0-14
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-8
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.12.5-2
ksm-control-daemon: 1.0-4
SOFIA:~#

 

[tom]

Is it still recommended to use acpid from backports since last week's updates to PVE stable?
...

I think yes, can you test?

 

[JustaGuy]

I think yes, can you test?

I've been, there's definetly something screwy somewhere.
Tasks came up today that will interrupt the process for awhile.
I'll post my findings once it's sorted.


---

Thanks for catching my typo on an other post earlier, glad the error wasn't allowed to remain.