Red Hat Bugzilla – Bug 589900
virt-manager: error cannot creating qcow2 storage volume
Last modified: 2010-05-14 03:03:28 EDT
Description of problem: Tried to add a new storage volume in virt-manager. Clicked on "New Volume" filled the fields but when I click on "finish" nothing gets created. Started virt-manager with --debug and captured the following log: 2010-05-07 10:59:06,342 (connection:212): Using libvirt API for netdev enumeration 2010-05-07 10:59:06,344 (connection:251): Using libvirt API for mediadev enumeration 2010-05-07 10:59:10,714 (engine:628): window counter incremented to 2 2010-05-07 10:59:12,043 (engine:632): window counter decremented to 1 2010-05-07 10:59:13,088 (create:715): Guest type set to os_type=hvm, arch=x86_64, dom_type=kvm 2010-05-07 10:59:19,003 (config:588): get_default_directory(media): returning /extern/ISOs 2010-05-07 10:59:27,849 (config:596): set_default_directory(media): saving /extern/ISOs 2010-05-07 10:59:39,049 (DistroInstaller:119): DistroInstaller location is a local file/path: /extern/ISOs/RHEL6.0-20100428.3-snap2-Workstation-x86_64-DVD1.iso 2010-05-07 10:59:53,153 (createvol:212): Creating volume with xml: <volume> <name>RHEL6.img</name> <capacity>10485760000</capacity> <allocation>0</allocation> <target> <format type='qcow2'/> </target> </volume> 2010-05-07 10:59:53,173 (createvol:244): Starting backround vol creation. 2010-05-07 10:59:53,173 (Storage:1137): Creating storage volume 'RHEL6.img' with xml: <volume> <name>RHEL6.img</name> <capacity>10485760000</capacity> <allocation>0</allocation> <target> <format type='qcow2'/> </target> </volume> 2010-05-07 10:59:53,374 (Storage:1193): Couldn't lookup storage volume in prog thread. Looks like this is qcow2 related, as it works when I select RAW! Version-Release number of selected component (if applicable): virt-manager-0.8.4-1.el6.noarch How reproducible: always Actual results: cannot add new storage volume Expected results: able to add new storage volume Additional info:
This request was evaluated by Red Hat Product Management for inclusion in a Red Hat Enterprise Linux major release. Product Management has requested further review of this request by Red Hat Engineering, for potential inclusion in a Red Hat Enterprise Linux Major release. This request is not yet committed for inclusion.
Are you missing a portion of the error output? I don't see anything here about not being able to create the storage volume. There is the message: 2010-05-07 10:59:53,374 (Storage:1193): Couldn't lookup storage volume in prog thread. But that is harmless in this case and shouldn't have interrupted anything. Can you be more specific?
I turned selinux to "permissive" and after that I got the following message: 2010-05-12 12:55:35,837 (createvol:244): Starting backround vol creation. 2010-05-12 12:55:35,838 (Storage:1137): Creating storage volume 'test.img' with xml: <volume> <name>test.img</name> <capacity>1048576000</capacity> <allocation>0</allocation> <target> <format type='qcow2'/> </target> </volume> 2010-05-12 12:55:35,935 (virt-manager:173): Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/host.py", line 651, in refresh_current_pool cp = self.current_pool() File "/usr/share/virt-manager/virtManager/host.py", line 662, in current_pool return self.conn.get_pool(curruuid) File "/usr/share/virt-manager/virtManager/connection.py", line 661, in get_pool return self.pools[uuid] KeyError: '63dcf4c1-cb5a-d6f1-196d-cccde7615769' None Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/host.py", line 651, in refresh_current_pool cp = self.current_pool() File "/usr/share/virt-manager/virtManager/host.py", line 662, in current_pool return self.conn.get_pool(curruuid) File "/usr/share/virt-manager/virtManager/connection.py", line 661, in get_pool return self.pools[uuid] KeyError: '63dcf4c1-cb5a-d6f1-196d-cccde7615769' 2010-05-12 12:55:36,041 (Storage:1193): Couldn't lookup storage volume in prog thread. Additional note: I've added a second storage pool in virt-manager and the problem only occurs if I try to add a qcow2 image there. Works in the primary storage pool. Adding a qcow2 image to the primary storage pool looks like this: 2010-05-12 13:03:53,807 (createvol:244): Starting backround vol creation. 2010-05-12 13:03:53,808 (Storage:1137): Creating storage volume 'test.img' with xml: <volume> <name>test.img</name> <capacity>1048576000</capacity> <allocation>0</allocation> <target> <format type='qcow2'/> </target> </volume> 2010-05-12 13:03:53,847 (Storage:1160): Storage volume 'test.img' install complete. 2010-05-12 13:03:54,012 (Storage:1193): Couldn't lookup storage volume in prog thread.
Okay, I'm still confused. Can you provide the following info: - How many pools do you have? - The XML output of all the pools (using virsh pool-dumxpml $poolname) - Can you consistently reproduce? - Is the error only harmless backtraces in the logs, or is an actual error dialog popping up? - Does selinux permissive vs. enforcing change the results at all? If you can consistently reproduce, please provide the full --debug output from a run, and the exact steps you take to reproduce, as well as the above info.
If SELinux is running in enforcing mode, then it could be a dup of bug 590713. libvirt uses qemu-img to create the qcow2 file and qemu-img had wrong security labelling preventing it working.
Sounds like that might be the issue, closing as a dup, Stefan please reopen if this issue seems separate. *** This bug has been marked as a duplicate of bug 590713 ***