Bug 1099827
| Summary: | "Create Virtual Machine"-wizard: Step 5 of 5 broken | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | [Community] Virtualization Tools | Reporter: | Daniel Krämer <d.kraemer> | ||||||||||||
| Component: | virt-manager | Assignee: | Cole Robinson <crobinso> | ||||||||||||
| Status: | CLOSED UPSTREAM | QA Contact: | |||||||||||||
| Severity: | unspecified | Docs Contact: | |||||||||||||
| Priority: | unspecified | ||||||||||||||
| Version: | unspecified | CC: | acathrow, berrange, crobinso, d.kraemer, gscrivan, kdex | ||||||||||||
| Target Milestone: | --- | ||||||||||||||
| Target Release: | --- | ||||||||||||||
| Hardware: | x86_64 | ||||||||||||||
| OS: | Linux | ||||||||||||||
| Whiteboard: | |||||||||||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||||||||||
| Doc Text: | Story Points: | --- | |||||||||||||
| Clone Of: | Environment: | ||||||||||||||
| Last Closed: | 2014-06-02 16:25:16 UTC | Type: | Bug | ||||||||||||
| Regression: | --- | Mount Type: | --- | ||||||||||||
| Documentation: | --- | CRM: | |||||||||||||
| Verified Versions: | Category: | --- | |||||||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||||||
| Embargoed: | |||||||||||||||
| Attachments: |
|
||||||||||||||
|
Description
Daniel Krämer
2014-05-21 09:50:00 UTC
Thanks for the report. Please reproduce with virt-manager --debug and post the output here Created attachment 898030 [details]
full debug output
I can confirm the same behavior and was just about to report this. I will provide my debug output as well. Created attachment 899355 [details]
Debug output in recreating bug
This makes VM creation impossible for me, as step 5 is essential for my setup.
Both of you have a similar root issue:
File "/usr/share/virt-manager/virtManager/create.py", line 1341, in page_changed
self.populate_summary()
File "/usr/share/virt-manager/virtManager/create.py", line 931, in populate_summary
self.netdev_changed(None)
File "/usr/share/virt-manager/virtManager/create.py", line 1086, in netdev_changed
obj = self.conn.get_net(key)
File "/usr/share/virt-manager/virtManager/connection.py", line 685, in get_net
return self.nets[uuid]
KeyError: 'e195702a-4650-2fe0-4227-f408225fc369'
For one, virt-manager obviously wasn't reporting this error and the UI behaved strangely. I've now pushed a fix for that upstream.
But it doesn't explain what exactly is going wrong here, we shouldn't ever see that key error.
I've pushed an additional debugging patch to git. Can someone reproduce with latest git and post the --debug output here?
git clone git://git.fedorahosted.org/virt-manager.git
cd virt-manager
./virt-manager --debug
Created attachment 901133 [details]
Debug output from git repository
Yes, I was still able to reproduce the bug. This time, on the other hand, I got stuck on page 4 of 5 (I wasn't able to move forward in the wizard).
See the attachment for more details.
Created attachment 901134 [details]
debugging patch
Thanks for the info. Still can't quite tell what the issue is... Can you apply the attached debugging patch to git, reproduce the issue, and post the new --debug output? Thanks a bunch Created attachment 901152 [details]
Debug output, patched git version
Sure, here's the output of the patched git version.
Thank you very much for the testing and prompt responses. I've fixed the issue upstream now:
commit d3e9af832899088a17aeb44441e8316177f563ae
Author: Cole Robinson <crobinso>
Date: Mon Jun 2 12:17:21 2014 -0400
pollhelpers: Use correct dictionary keys for old pool/net polling (bz 109982
We were inadvertently using the object name rather than UUID when libvirt
didn't support new style polling APIs.
The reason I wasn't seeing it is because all libvirtd instances I was testing against use new style polling APIs, and our fallback path was broken in this case.
Thank you very much for your quick intervention. I can't reproduce the bug anymore and thus consider the issue resolved. |