I also experienced this issue within 4 hours of a host upgrade to kernel 3.8.0-27 yesterday, and migrating ~16 guests in.
There were two observed guest types which lost networking: Ubuntu 12.10 Quantal, and Ubuntu 10.04 Lucid LTS. Rebooting guest was the only way to re-establish bridged network connection to the guests. brctl outputs, ifconfig, and virsh commands all looked proper, and matched working Guests. No errors in host or guest syslogs, or consoles. On the final frozen guest, this morning, we attempted to migrate a network-down guest OS to an IDENTICAL hardware+3.8.0-27 kernel host in hopes of magically regaining networking; and the source host kernel- panicked and froze! Sadly the console display size was too small to see the source of Panic, nor was detail logged to the syslog. We did see a lot of net-tx/rx detail in the stack listing. We hard-booted this panicked host into previous, 3.8.0-26 kernel, 4 hours uptime and no guest connectivity loss (yet?). -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1208532 Title: put_page failures with 3.8.0-27.40 To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1208532/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
