RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 1267154 - libvirt tries to change ownership of ISOs even though dynamic_ownership is set to 0
Summary: libvirt tries to change ownership of ISOs even though dynamic_ownership is se...
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: libvirt
Version: 7.2
Hardware: All
OS: All
unspecified
high
Target Milestone: rc
: ---
Assignee: Michal Privoznik
QA Contact: Virtualization Bugs
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2015-09-29 07:36 UTC by Jiri Denemark
Modified: 2015-11-19 06:55 UTC (History)
15 users (show)

Fixed In Version: libvirt-1.2.17-12.el7
Doc Type: Bug Fix
Doc Text:
Clone Of: 1266628
Environment:
Last Closed: 2015-11-19 06:55:38 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2015:2202 0 normal SHIPPED_LIVE libvirt bug fix and enhancement update 2015-11-19 08:17:58 UTC

Description Jiri Denemark 2015-09-29 07:36:36 UTC 
+++ This bug was initially created as a clone of Bug #1266628 +++

[root@adam libvirt]# grep dynamic /etc/libvirt/qemu.conf
# Whether libvirt should dynamically change file ownership
dynamic_ownership = 0

yet, with libvirt 1.2.17-10, when I start a VM, it tries to change ownership of the ISO file attached to it (and fails, as it's on a network share). This prevents me running any VMs, in my setup.

Downgraded to 1.2.17-9, it works fine.

I kinda suspect this change:

https://libvirt.org/git/?p=libvirt.git;a=commitdiff;h=f4c60dfbf2ec606a5fc148b4c6ff1cd17ffd79ec

which adds a virSecurityDACSetOwnership call which is not protected by a block like this:

    if (!priv->dynamicOwnership)
        return 0;

which several of the others are. But not *all* of them are, and I'm not 100% sure exactly what that virSecurityDACDomainSetDirLabel() is for or what it operates on, so IMBW.

--- Additional comment from Cole Robinson on 2015-09-28 23:55:07 UTC ---

Thanks for the report... you were in the right ballpark but the breakage is a bit more subtle. Patch posted upstream:

https://www.redhat.com/archives/libvir-list/2015-September/msg01000.html


This bug is caused by patches for bug 1124841.
Comment 1 Michal Privoznik 2015-09-29 13:58:47 UTC 
Moving to POST:

http://post-office.corp.redhat.com/archives/rhvirt-patches/2015-September/msg00676.html
Comment 3 zhenfeng wang 2015-10-12 05:38:26 UTC 
Reproduce this bug with libvirt-17.2-10

1.Set dynamic_ownership=0 in qemu.conf
#cat /etc/libvirt/qemu.conf
dynamic_ownership=0
user=qemu
group=qemu

#systemctl restart libvirtd

2.Prepare a guest with a cdrom and iso file inserted
#virsh dumpxml vm1
--
    <disk type='file' device='cdrom'>
      <driver name='qemu' type='raw'/>
      <source file='/var/lib/libvirt/images/virtio-win-1.7.4.iso'/>
      <target dev='hda' bus='ide'/>
      <readonly/>
      <address type='drive' controller='0' bus='0' target='0' unit='0'/>
    </disk>

# ll /var/lib/libvirt/images/virtio-win-1.7.4.iso -Z
-rw-r--r--. root root system_u:object_r:virt_content_t:s0 /var/lib/libvirt/images/virtio-win-1.7.4.iso

3.Start the guest, guest could start successfully and libvirt have change the
ownership of isos and vmdisk even though dynamic_ownership is set to 0

# virsh start vm1
Domain vm1 started

# ps -efZ|grep vm1
system_u:system_r:svirt_t:s0:c122,c695 qemu 11614  1 88 13:23 ?        00:00:14 /usr/libexec/qemu-kvm -name vm1

# ll /var/lib/libvirt/images/vm924.qcow2 -Z
-rw-r--r--. qemu qemu system_u:object_r:svirt_image_t:s0:c122,c695 /var/lib/libvirt/images/vm924.qcow2
# ll /var/lib/libvirt/images/virtio-win-1.7.4.iso -Z
-rw-r--r--. qemu qemu system_u:object_r:virt_content_t:s0 /var/lib/libvirt/images/virtio-win-1.7.4.iso


Verify this bug with libvirt-1.2.17-13.el7

SCENARIO 1
1.Rexcute step 1~2 in reproduce steps
2.start guest, guest will fail to start since qemu process didn't have
permission to use the disk 
# virsh start vm1
error: Failed to start domain vm1
error: internal error: process exited while connecting to monitor: 2015-10-10T07:11:56.035219Z qemu-kvm: -drive file=/var/lib/libvirt/images/vm924.1444384712,if=none,id=drive-virtio-disk0,format=qcow2: Could not open '/var/lib/libvirt/images/vm924.1444384712': Permission denied

3.chown the guest's disk to qemu:qemu, then start guest, guest could start successfully
# chown qemu:qemu vm924.qcow2
-rw-r--r--. 1 qemu qemu  9329508352 Oct 10 15:16 vm924.qcow2

# virsh start vm1
Domain vm1 started

# ll -Z /var/lib/libvirt/images/virtio-win-1.7.4.iso 
-rw-r--r--. root root system_u:object_r:virt_content_t:s0 /var/lib/libvirt/images/virtio-win-1.7.4.iso

# ll -Z /var/lib/libvirt/images/vm924.qcow2 
-rw-r--r--. qemu qemu system_u:object_r:svirt_image_t:s0:c100,c368 /var/lib/libvirt/images/vm924.qcow2

SCENARIO 2
1.set dynamic_ownership=1
user=qemu
group=qemu

2.start a guest, guest could start successfully
# virsh start vm1
Domain vm1 star

# ll /var/lib/libvirt/images/virtio-win-1.7.4.iso  -Z
-rw-r--r--. qemu qemu system_u:object_r:virt_content_t:s0 /var/lib/libvirt/images/virtio-win-1.7.4.iso
[root@rhel73zhwang ~]# ll /var/lib/libvirt/images/vm924.qcow2 -Z
-rw-r--r--. qemu qemu system_u:object_r:svirt_image_t:s0:c917,c940 /var/lib/libvirt/images/vm924.qcow2

3.Set per-vm dac in guest's xml, then start guest

#virsh dumpxml vm1
--
  </devices>
  <seclabel type='static' model='dac' relabel='yes'>
    <label>test1:test1</label>
  </seclabel>
</domain>
--
# virsh start vm1
Domain vm1 started

# ps -efZ|grep qemu
system_u:system_r:svirt_t:s0:c800,c867 test1 7844  1 87 11:04 ?        00:00:11 /usr/libexec/qemu-kvm -name vm1 

# ll /var/lib/libvirt/qemu/domain-vm1/ -Zd
drwxr-x---. test1 test1 system_u:object_r:svirt_image_t:s0:c800,c867 /var/lib/libvirt/qemu/domain-vm1/

# ll -Z /var/lib/libvirt/images/vm924.qcow2 
-rw-r--r--. test1 test1 system_u:object_r:svirt_image_t:s0:c800,c867 /var/lib/libvirt/images/vm924.qcow2

# ll -Z /var/lib/libvirt/images/virtio-win-1.7.4.iso 
-rw-r--r--. test1 test1 system_u:object_r:virt_content_t:s0 /var/lib/libvirt/images/virtio-win-1.7.4.iso

SCENARIO 3
Do regression test for bug 1124841, all verify steps in that 
bug could pass

According to upper steps, mark this bug verifed
Comment 5 errata-xmlrpc 2015-11-19 06:55:38 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://rhn.redhat.com/errata/RHBA-2015-2202.html
Note You need to log in before you can comment on or make changes to this bug.