I'd agree and work on adding the rule upstream and into Ubuntu, but what I need to to do is help to understand "why this triggers for you".
I run libvirt locally and in many tests, but so far have never seen this apparmor denial.
Although if it is a non fatal bug it is easier to miss ...
The linked Debian bug (thanks paride) has a bit more details how to trigger.
But the bug also is almost a year old and no one else has hit this, ... that is odd.
I've set up a L1 guest with an extra disk as scsi disk
44 <disk type='file' device='disk'>
45 <driver name='qemu' type='qcow2'/>
46 <source file='/var/lib/uvtool/libvirt/images/testguest-scsi-ephem-00.qcow'/>
47 <target dev='sda' bus='scsi'/>
48 <address type='drive' controller='0' bus='0' target='0' unit='0'/>
49 </disk>
...
100 <controller type='scsi' index='0' model='virtio-scsi'>
101 <address type='pci' domain='0x0000' bus='0x0a' slot='0x01' function='0x0'/>
102 </controller>
In the guest that appears as scsi disk, here from lshw:
*-scsi
description: SCSI storage controller
product: Virtio SCSI
vendor: Red Hat, Inc.
physical id: 1
bus info: pci@0000:07:01.0
version: 00
width: 64 bits
clock: 33MHz capabilities: scsi msix bus_master cap_list configuration: driver=virtio-pci latency=0
resources: irq:23 ioport:c000(size=64) memory:fc000000-fc000fff memory:fe000000-fe003fff
*-disk
description: SCSI Disk
product: QEMU HARDDISK
vendor: QEMU
physical id: 0.0.0
bus info: scsi@0:0.0.0
logical name: /dev/sda
version: 2.5+
size: 4GiB (4294MB) capabilities: 5400rpm configuration: ansiversion=5 logicalsectorsize=512 sectorsize=512
*-sata
description: SATA controller
product: 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA Controller [AHCI mode]
vendor: Intel Corporation
physical id: 1f.2
bus info: pci@0000:00:1f.2
version: 02
width: 32 bits
clock: 33MHz capabilities: sata msi ahci_1.0 bus_master cap_list configuration: driver=ahci latency=0
resources: irq:41 ioport:d060(size=32) memory:fd41b000-fd41bfff
Using that to define another guest:
<disk type='block' device='disk'> <driver name='qemu' type='raw'/> <source dev='/dev/sda'/> <target dev='sda' bus='scsi'/>
</disk>
<controller type='scsi' index='0' model='virtio-scsi'/>
But with that the guest starts fine and no apparmor denial shows up.
Could you help by outlining how you configure your host and guest so that this issue triggers.
Only then we have a use case that we can tie to the new apparmor rule to allow this.
I'd agree and work on adding the rule upstream and into Ubuntu, but what I need to to do is help to understand "why this triggers for you".
I run libvirt locally and in many tests, but so far have never seen this apparmor denial.
Although if it is a non fatal bug it is easier to miss ...
The linked Debian bug (thanks paride) has a bit more details how to trigger.
But the bug also is almost a year old and no one else has hit this, ... that is odd.
I've set up a L1 guest with an extra disk as scsi disk var/lib/ uvtool/ libvirt/ images/ testguest- scsi-ephem- 00.qcow' /> virtio- scsi'>
44 <disk type='file' device='disk'>
45 <driver name='qemu' type='qcow2'/>
46 <source file='/
47 <target dev='sda' bus='scsi'/>
48 <address type='drive' controller='0' bus='0' target='0' unit='0'/>
49 </disk>
...
100 <controller type='scsi' index='0' model='
101 <address type='pci' domain='0x0000' bus='0x0a' slot='0x01' function='0x0'/>
102 </controller>
In the guest that appears as scsi disk, here from lshw:
capabilities: scsi msix bus_master cap_list
configuration: driver=virtio-pci latency=0 c000(size= 64) memory: fc000000- fc000fff memory: fe000000- fe003fff
capabilities: 5400rpm
configuration: ansiversion=5 logicalsectorsi ze=512 sectorsize=512
capabilities: sata msi ahci_1.0 bus_master cap_list
configuration: driver=ahci latency=0 d060(size= 32) memory: fd41b000- fd41bfff
*-scsi
description: SCSI storage controller
product: Virtio SCSI
vendor: Red Hat, Inc.
physical id: 1
bus info: pci@0000:07:01.0
version: 00
width: 64 bits
clock: 33MHz
resources: irq:23 ioport:
*-disk
description: SCSI Disk
product: QEMU HARDDISK
vendor: QEMU
physical id: 0.0.0
bus info: scsi@0:0.0.0
logical name: /dev/sda
version: 2.5+
size: 4GiB (4294MB)
*-sata
description: SATA controller
product: 82801IR/IO/IH (ICH9R/DO/DH) 6 port SATA Controller [AHCI mode]
vendor: Intel Corporation
physical id: 1f.2
bus info: pci@0000:00:1f.2
version: 02
width: 32 bits
clock: 33MHz
resources: irq:41 ioport:
Using that to define another guest:
<driver name='qemu' type='raw'/>
<source dev='/dev/sda'/>
<target dev='sda' bus='scsi'/> virtio- scsi'/>
<disk type='block' device='disk'>
</disk>
<controller type='scsi' index='0' model='
But with that the guest starts fine and no apparmor denial shows up.
Could you help by outlining how you configure your host and guest so that this issue triggers.
Only then we have a use case that we can tie to the new apparmor rule to allow this.