Red Hat Bugzilla – Bug 759392
Ubuntu 10.04 guest loses network connection if big files are downloaded in the guest
Last modified: 2013-01-09 19:35:03 EST
Description of problem:
I am using Virtual Manager and have created a virtual machine running Ubuntu 10.04. However, the guest loses the network quite often. I think that it is related to downloading big files, or maybe high network traffic. I was able to reproduce it almost all the times by running 'apt-get upgrade' in the Ubuntu guest, which causes a big network traffic (with 200 updates to be applied :).
After the network is lost, the only way I could find to recover is to reboot the guest, which is far from desirable in a production environment.
The interesting thing is that the Ubuntu 10.04 guest run without problems with Fedora 14 host, but after I upgraded the host to Fedora 15, it started giving problems. However, the use under Fedora 14 was scarce, and therefore I cannot tell whether it is a regression or simply I didn't spot the problem before.
Just for the record, I have seen similar problems but with a Ubuntu 10.04 host and a Debian Squeeze guest.
Maybe it is related to Bug #734731, but in principle they seem different bugs.
I attach the qemu.conf used in libvirt and the Ubuntu-10.04.xml which defines the virtual machine
Version-Release number of selected component (if applicable):
Most of the cases
Steps to Reproduce:
1.Install a Ubuntu 10.04 guest with Virtual Manager
2.Run apt-upgrade in the guest
This is not however the way I did it, since I installed the guest under Fedora 14 and later on I upgraded the host to Fedora 15.
Created attachment 539537 [details]
definition of virtual manchine by libvirt
Created attachment 539538 [details]
qemu.conf used by libvirt
This package has changed ownership in the Fedora Package Database. Reassigning to the new owner of this component.
Given that F15 is end of life in less than a month, it's unlikely this will get fixed before hand. Enrique, if you can reproduce this issue with a later Fedora version, please reopen and we can go from there.