Public bug reported:
I noticed if I create an iPV6 only network using DHCPv6 stateful the
qdhcp namespeace does not contain a default route to make recursive DNS
requests possible as they need to route to external networks.
The below shows what I see in the routing table of the qdhcp namespace
Public bug reported:
Release = newton
When launching a Windows instance utilizing an IPv6-only network,
cloudinit fails to post the encrypted (generated) password to the
metadata service due to the absence of the metadata service on non-
dual-stack/IPv4 probably
** Affects: neutron
Public bug reported:
Via the nova compute API I can see it is possible to GET metadata for an
instance but it does not seem to be possible to get system_metadata. It
is useful to be able to GET the system metadata fro an instance if you
want to query the point in time properties that where
Public bug reported:
Version = newton (2:9.2.0-0ubuntu1~cloud0)
DVR (HA mode) with ipv4 and ipv6
l2_population = True
I noticed when using dual stack ipv4 and ipv6 Linux guests are working
fine but windows guests seemed to have a problem with their ipv4
connectivity. Upon investigation I found
Public bug reported:
I have noticed the below in the newutron l3 agent log of compute nodes
running in DVR mode, It seems to log the same message every 60 seconds
or so. Restarting the agent will stop the log until log rotate is called
the next day and then the return in the same fashion.
Public bug reported:
Version = openstack-dashboard3:11.0.1-0ubuntu1~cloud0
Ceph version = 10.2.7
When using ceph RGW swift interface for open stack and the open stack
dashboard version above to create a swift container the dashboard does a
number of curl requests to check if the
Public bug reported:
It appears that when the Lock Instance / Unlock Instance options are
used on a compute instance, nothing is indicated within the dashboard or
action log to suggest this is the case.
Steps to repeat:
Log into dashboard
Click instances
>From the drop down on the right of a
Public bug reported:
OpenStack liberty (Ubuntu 14.04)
When creating a new instance via the dashboard and selecting 'boot from
image (creates new volume)' the below is logged into nova-compute.log on
the host the instance is deployed to:
"Ignoring supplied device name: /dev/vda. Libvirt can't
** Also affects: ubuntu
Importance: Undecided
Status: New
** No longer affects: ubuntu
--
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1528877
Title:
Libvirt
Public bug reported:
When using Ceph as the back end storage a snapshot of a volume is
dependent upon the original volume meaning it is an illegal operation to
delete the volume without first deleting the snapshot.
Should you try and delete such a volume horizon {kilo} reports back the
error
10 matches
Mail list logo
