Hi! There seem to be two issues here in the one bug -- DHCP refreshes
are causing problems, and pain with windows instances because of how
they handle timezones and that causing extra DHCP refreshes. I'll
talk about each of those separately.
Windows timezones
=================
I'm sorry to hear you're having pain with Windows instances.
Unfortunately I don't know a lot about windows, but it sounds as though
if we could change the bios timezone for Windows instances to the
timezone of the compute node that would alleviate your issue. Is that
correct?
Looking at the code, each instance has an os_type field, which is
inherited from the image the instance is based off (there is an image
property called os_type). Xen expects this value to be "windows" for
windows instances. So, libvirt could do the same thing here and treat
the timezone for instances with os_type == "windows" differently.
I'm going to create a new bug to track this windows issue, so that I
can leave this bug focussed on the DHCP timeout issue. This isn't
because Windows is less important as a guest, but because I don't want
it to be missed in the DHCP discussion that's happening here.
DHCP refreshes
==============
It also sounds as if we could bear in increase the default DHCP interval,
although you can configure that yourself as well. It might also make
sense to ping an IP before we delete its lease.
Cheers,
Michael
** Changed in: nova
Importance: Undecided => High
** Changed in: nova
Status: Confirmed => Triaged
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1026621
Title:
nova-network gets release_fixed_ip events from someplace, but the
database still keeps them associated with instances
To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1026621/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
