Public bug reported:
Currently zookeeper's log files grow without limit which can cause a
machine to eventually run out of disk space.
The zookeeper package should include a logrotate script or configure
log4j (whichever is most appropriate) to rotate the log files.
We currently use the
** Tags added: canonistack
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova in Ubuntu.
https://bugs.launchpad.net/bugs/1161338
Title:
python-nova should depend on ebtables
To manage notifications about this bug go to:
Public bug reported:
After upgrading my installation from Folsom to Grizzly I received the
below error message when attempting to create a volume using the EC2 API
(I did not verify with the Nova API).
I have configured two regions; one each with a cinder endpoint. This
works for all other
*** This bug is a duplicate of bug 1087735 ***
https://bugs.launchpad.net/bugs/1087735
** This bug has been marked a duplicate of bug 1087735
Attaching volume fails if keystone has multiple endpoints of Cinder
--
You received this bug notification because you are a member of Ubuntu
Public bug reported:
When attempting to change a quota for one of my customers, with the
command below, I receive an error message instead of the command
succeeding.
What I see:
$ sudo nova-manage project quota --project=
--key=instances --value=15
Public bug reported:
Currently python-boto does not verify SSL certificates by default. This
is unacceptable as this exposes users to man in the middle attacks. This
can be worked around by the user (see below).
Unfortunately after enabling verification, python-boto uses it's own
cacerts.txt
@Dan
The default_character_set_name is set to 'latin1'. This system has been
upgraded through various lifecycles of Openstack which may explain it (a
more recent installation that I looked at is set to 'utf8').
I would argue that setting the character set explicitly on table
creation is the
Public bug reported:
The upgrade from Precise+Essex to Precise+Folsom using the ubuntu-cloud
archive fails while running the 092_add_instance_system_metadata
migration. The migration fails while creating the
`instance_system_metadata` table (see below).
The '1005' MySQL error indicates that a
** Patch added: 092_add_instance_system_metadata.patch
https://bugs.launchpad.net/bugs/1062277/+attachment/3375835/+files/092_add_instance_system_metadata.patch
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova in Ubuntu.
I upgraded today and found that I had this issue.
Not all of my instances had duplicate entries in the
`instance_id_mappings` table, however I did have more entries within the
`instance_id_mappings` than within the `instances` table.
It seems that the `id` column in the `instances` table is
The loose relationship between two auto_incrementing `id` columns of
different tables is insanely brittle. I am not convinced that the
intention was for the relation to work this way. It may simple be a
problem with part of the code referencing `id` instead of `uuid`.
The uuid's in the
Public bug reported:
Running `ceph-authtool keyringfile --print` throws an error claiming
that it does not know the option however, the help indicates that this
is the valid long-option for '-p'.
Looking at the source code, the command expects '--print-key' and using
this works as expected.
** Patch added: Correct help message to provide correct long option
'--print-key'
https://bugs.launchpad.net/bugs/1056462/+attachment/3343155/+files/ceph_authtool.patch
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to ceph in
@Rafi,
Thanks for the suggestion. What you describe used to happen to me when
we were using Oneiric's libvirtd (LP#903212). Since upgrading to Precise
I haven't experienced the exact problem you're having.
My issue is that libvirtd seems to temporarily stop responding, enough
to block
@Serge,
As mentioned in #1 the cluster is running Precise + Essex. Unfortunately
I cannot consistently reproduce the problem, however the problem has
been recurring approximately once per month.
As mentioned in #5 I think that perhaps nova-compute is attempting to
query libvirtd and libvirtd is
*** This bug is a duplicate of bug 1033934 ***
https://bugs.launchpad.net/bugs/1033934
** This bug has been marked a duplicate of bug 1033934
Attempting to change regions in the dashboard does not display correctly
--
You received this bug notification because you are a member of Ubuntu
*** This bug is a duplicate of bug 1033920 ***
https://bugs.launchpad.net/bugs/1033920
** This bug has been marked a duplicate of bug 1033920
Dashboard raises a ServiceCatalogException when attempting to download juju
settings
--
You received this bug notification because you are a
@Adam:
Bug #1006898 seems unlikely IMHO. We are not running with 'multi_host'
nor with VLANS. So there is one dnsmasq process.
The following is just conjecture as the event has rotated out of the logs and I
have not done any further testing. These events happen rather infrequently, but
are a
Public bug reported:
When attempting to download the juju settings file from the dashboard I
receive an exception¹ which displays an ugly, blank 500 error message.
It seems that the juju module is expecting an S3 endpoint in my service
catalog, not finding it and throwing an exception which is
Public bug reported:
After logging in to the dashboard in one region I attempted to change
regions by selecting the other region in the top right corner. After
doing this, login credentials form elements appear on the bottom left-
hand side and does not look correct (I have attached screenshots
** Attachment added: Tarball containing images illustrating the display
problem.
https://bugs.launchpad.net/bugs/1033934/+attachment/3251203/+files/region-displays-incorrectly.tar.gz
** Description changed:
After logging in to the dashboard in one region I attempted to change
regions by
Public bug reported:
Keystone's `token` table grows unconditionally with expired tokens.
Keystone should provide a backend-agnostic method to find and delete
these tokens. This could be run via a periodic task or supplied as a
script to run as a cron job.
An example SQL statement (if you're
** Description changed:
Keystone's `token` table grows unconditionally with expired tokens.
Keystone should provide a backend-agnostic method to find and delete
these tokens. This could be run via a periodic task or supplied as a
script to run as a cron job.
An example SQL
** Description changed:
A client of mine attempted to rescue an instance by using the `nova
- rescuecl` command which failed putting the instance in the `ERROR`
- state.
+ rescue` command which failed putting the instance in the `ERROR` state.
I am running Openstack Essex on Ubuntu 12.04
Public bug reported:
A client of mine attempted to rescue an instance by using the `nova
rescuecl` command which failed putting the instance in the `ERROR`
state.
I am running Openstack Essex on Ubuntu 12.04 with KVM as my hypervisor
and I do not have `rescue_image_id`, `rescue_kernel_id` nor
** Description changed:
A client of mine attempted to rescue an instance by using the `nova
rescuecl` command which failed putting the instance in the `ERROR`
state.
I am running Openstack Essex on Ubuntu 12.04 with KVM as my hypervisor
and I do not have `rescue_image_id`,
We're running nova with nova.network.manager.FlatDHCPManager. Grepping
the nova-network machine's syslog shows that the last DHCP lease request
was on Jul 16 14:04:29
Sample from the log file:
Jul 16 14:04:29 dziban dnsmasq-dhcp[30249]: DHCPREQUEST(br100) 10.55.60.141
fa:16:3e:11:c5:37
Jul 16
** Description changed:
- Running proposed on one of our clusters, I see the following with
- instances started via juju. I have been unable to re-create the problem
- with raw ec2 commands.
+ Running Precise proposed on one of our clusters, I see the following
+ with instances started via juju.
Public bug reported:
What happens:
I have two regions and one availability zone (AZ) for my Essex cluster.
When creating a volume I accidentally used the region name instead of
the AZ name. This resulted in a volume being stuck in the creating
state which I cannot delete.
What I expect:
I
Fortunately (or not) this has just recently occurred. We do not have
debug symbols installed.
It looks to me to be stuck on virDomainGetInfo(). Interestingly,
libvirtd seems to be responding when I query it. Perhaps there is a
missing timeout somewhere?
Backtrace from python:
(gdb) bt
#0
I do not believe that we've seen this issue in the later releases of
Diablo and in Essex. I think that we can close this bug unless someone
objects or still experiences this issue.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova
*** This bug is a duplicate of bug 897140 ***
https://bugs.launchpad.net/bugs/897140
** This bug has been marked a duplicate of bug 897140
unassociated floating IPs not visible to admin
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is
Public bug reported:
Occassionally I need to deallocate floating IP addresses that my clients
have allocated but are not using. Attempting to do this throws a
NotAuthorized even if my user has the global role of cloudadmin.
Unfortunately the 'nova-manage' commands do not seem to provide a means
I ran the following SQL query to work around the issue:
sql UPDATE floating_ips SET deleted_at = NULL, fixed_ip_id = NULL,
project_id = NULL, host = NULL where floating_ips.deleted = 0 AND
floating_ips.auto_assigned = 0 and host is NULL;
--
You received this bug notification because you are a
@James:
It looks like the --help output doesn't list --region and lists
--os_region_name and --region_name instead. That said, my tests seem to
show that the --region option still works (for now).
Retesting:
$ for opt in region region_name os_region_name; do
for region in doesnotexist
The symptoms are similar to what we experienced in LP#903212, however I
can confirm that libvirtd seems to be responding correctly in Precise.
Is there further information that we can provide?
$ dpkg-query --show nova-*
nova-api2012.1~e4~20120210.12574-0ubuntu1
nova-common
This has happened again.
Process listing:
$ ps auxwwwf | grep [n]ova-compute
nova 25735 0.0 0.0 48040 4 ?Ss Apr16 0:00 su -s /bin/sh
-c exec nova-compute --flagfile=/etc/nova/nova.conf
--flagfile=/etc/nova/nova-compute.conf nova
nova 25746 1.2 0.1 1725088 32604 ?
Public bug reported:
I am performing some tests with Keystone and multiple regions and have
discovered that nova client does not respect the --region argument
correctly.
Setup:
==
I have defined services within two regions:
- regionOne
- regionTwo
Tests:
==
# Ask for all endpoints
$
A little more clarification:
When setting NOVACLIENT_DEBUG in my shell's environment I can confirm
that Keystone returns the full catalog for all regions to the nova
client. Assuming this is the correct behavior for Keystone, then this is
a client-side filtering bug.
** Tags added: canonistack
Public bug reported:
The euc2ools commands do not support the --region flag for anything
other than EC2. I would like to define multiple regions and use these
commands against Eucalyptus and Openstack clouds (among others).
Currently the endpoint URL is hardcoded (see the function below).
Further testing shows that nova list commands respect --region
correctly. So this may only affect a subset of commands. I've updated
the bug description to reflect this.
** Summary changed:
- nova client does not respect regions
+ nova client does not respect regions for a subset of commands
--
@Adam:
I definitely had to update the Glance database and update the
image_properties table. My memory is a little fuzzy on the exact
details, but I believe that the nova.s3_images table did not have the
correct UUID for ramdisks and kernels as well. I only noticed this much
later however and I
suggested in order to see the IP address.
--
Regards.
Andrew Glen-Young
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to euca2ools in Ubuntu.
https://bugs.launchpad.net/bugs/933656
Title:
euca-describe-instances --ipv4 does
Public bug reported:
The package python-nova should include openssh-client as a dependency as
it includes a call to the ssh-keygen binary. Specifically, the file
/usr/share/pyshared/nova/crypto.py includes a shell out to this binary.
= Testing =
$ grep -n ssh-keygen
*** This bug is a duplicate of bug 943031 ***
https://bugs.launchpad.net/bugs/943031
** This bug has been marked a duplicate of bug 943031
After MySQL connection failure (or timeout), first request reports MySQL
Server has gone away, following requests work
--
You received this bug
Public bug reported:
I have configured Keystone to use SQL backends (MySQL in my case) for all
services running on Precise.
After restarting MySQL, Keystone does not to attempt to reconnect to the
database, but instead throws a traceback.
Tests:
--
1. Start MySQL
$ sudo start mysql
** Attachment added: Keystone traceback
https://bugs.launchpad.net/bugs/953985/+attachment/2862542/+files/keystone.traceback
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to keystone in Ubuntu.
https://bugs.launchpad.net/bugs/953985
Considering Glance and Nova use sqlite as default, it seems reasonable
that Keystone should do the same. This would a) prevent interactive
prompts and b) provide a consistent experience between Openstack
components.
--
You received this bug notification because you are a member of Ubuntu
Server
Public bug reported:
The keystone package creates a keystone user who's primary (and only)
group membership is 'nogroup'. This results in files/directories created
by the keystone user having 'nogroup' group ownership. This is
considered a bad thing.
The keystone user should have it's primary
** Patch added: keystone-group.patch
https://bugs.launchpad.net/bugs/941905/+attachment/2787690/+files/keystone-group.patch
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to keystone in Ubuntu.
https://bugs.launchpad.net/bugs/941905
New patch attached. Removes my redundant setting of permissions on
files/directories.
** Patch added: keystone-group.patch
https://bugs.launchpad.net/ubuntu/+source/keystone/+bug/941905/+attachment/2787789/+files/keystone-group.patch
--
You received this bug notification because you are a
@Thierry: I believe that new(er) Precise packages should be uploaded
today (Friday). I'll test them early next week and let you know if I see
this error.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova in Ubuntu.
@Adam:
The upgrade was an upgrade of the Ubuntu packages from the Precise/Essex
(i.e. _not_ between Openstack releases). There were instances running at
the time.
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova in Ubuntu.
Public bug reported:
This morning I upgraded my Openstack hosts to the latest packaged
version of the nova packages. After the upgrade some of the compute node
machines were rather unresponsive.
After investigation it became apparent that nova-compute was nearly
exhausting the node's memory. I
Public bug reported:
After upgrading the Openstack packages in Precise this morning to the
latest version, I have found that nova-manage doesn't support the
fixed (fixed-ip) catagory.
The traceback:
$ sudo nova-manage fixed list
No handlers could be found for logger nova.virt.libvirt.firewall
Public bug reported:
After upgrading nova-compute this morning to the latest nova-compute
package in Precise I have noticed a non-fatal traceback in /var/log/nova
/nova-compute.log.
Traceback:
2012-02-22 15:04:28,413 ERROR nova.manager [-] Error during
ComputeManager._sync_power_states: 'dict'
Public bug reported:
I have recently upgraded my Openstack installation from the version packaged in
Oneiric to the one packaged in Precise.
After the installation, no floating ip addresses were available to instances.
When running euca-allocate-address returns a NoMoreFloatingIps
exception.
Public bug reported:
I am using euca2ools with Openstack Nova in Precise Pangolin.
euca-describe-instances helpfully includes an --ipv4 option to display
ip address information of my running instances. Unfortunately, it does
not seem to include the instances public ip address when one is
I forgot the software version.
$ dpkg-query --show euca2ools
euca2ools 2.0.0~bzr495-0ubuntu1
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to euca2ools in Ubuntu.
https://bugs.launchpad.net/bugs/933656
Title:
Public bug reported:
I have recently upgraded from Oneiric to Precise and noticed that 'euca-
get-console-output instance' no longer returns data.
Some investigation reveals that there may be a regression where the
libvirt template is set to log to a log file, but nova-compute is
reading a fifo.
Public bug reported:
After upgrading Openstack from the packaged version in Oneiric to the
packaged version in Precise I have found that some images are being
reported as missing, but not all.
I am using nova-api, glance and swift with deprecated_auth.
Please let me know if you need any further
On Wed 15-02-2012 19:02 ›, Andrew Glen-Young wrote:
glance client:
$ glance show c9d4eed5-2ecc-4fa5-9ae1-0da4a0f0c066
URI: http://0.0.0.0:9292/v1/images/c9d4eed5-2ecc-4fa5-9ae1-0da4a0f0c066
Id: c9d4eed5-2ecc-4fa5-9ae1-0da4a0f0c066
Public: Yes
Protected: No
Name: None
Status: active
Public bug reported:
nova-objectstore's logrotate script uses 'dpkg-vendor' to try determine
how to restart nova-objectstore. This requires 'dpkg-dev' to be
installed which is not desirable.
The script really should use /etc/lsb-release.
Current logrotate script
Public bug reported:
After starting an AMD64 Oneiric instance (downloaded from uec-images.u.c
2011-11-29) using LXC as the nova-compute libvirt backend, I am
occasionally seeing the instance fail to start and the kernel triggering
a backtrace (attached).
There are no libvirtd logs for this
** Attachment added: Kernel backtrace
https://bugs.launchpad.net/bugs/900378/+attachment/2620718/+files/lxc-failed-instance-kern.log
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova in Ubuntu.
** Attachment added: Nova-compute log file
https://bugs.launchpad.net/ubuntu/+source/nova/+bug/900378/+attachment/2620719/+files/nova-compute.log
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to nova in Ubuntu.
Public bug reported:
This is similar to LP#851860
nova-common 2011.3~rc~20110909.r1155-0ubuntu1
The nova package creates a nova user who's primary (and only) group
membership is 'nogroup'. This results in files/directories created by
the nova user having 'nogroup' group ownership. This is
Public bug reported:
For forensic purposes and general good practice, rabbitmq's starup_log
and shutdown_log should contain a date and time that the service was
started/shutdown. Having to use the logfile's mtime seems sub-optimal.
** Affects: rabbitmq-server (Ubuntu)
Importance: Undecided
@Thierry:
The last line of the upstart script in your branch is not quoted
correctly, however I have been using this script for sometime (and
across reboots) with success since I filed LP#820360
--
You received this bug notification because you are a member of Ubuntu
Server Team, which is
Steps to reproduce
==
Initial Install and Tests
-
$ sudo apt-get install squid squid-client
$ grep MAX /etc/default/squid·
SQUID_MAXFD=1024
$ sudo status squid
squid start/running, process 2671
$ squidclient -p 3128 -l 127.0.0.1 -h 127.0.0.1 mgr:info |
@Dave
I have not tried getting this into upstream. AFAIK the project has not
has not been actively maintained since 2005, which is a terrible shame.
I have previously tried contacting the author (a few years back) but
never received a response.
--
analog logfile analyser fails to identify
@chris, @Thierry Thanks, this makes sense to me now.
--
'euca-describe-availability-zones verbose' should show vm types for non-admin
users
https://bugs.launchpad.net/bugs/537464
You received this bug notification because you are a member of Ubuntu
Server Team, which is subscribed to eucalyptus
@chris, could you please explain the rationale behind this behavior?
Since this information is often needed by developers and users when they
want to create or run images on EUC - I am curious as to why this
command is limited to the admin user only?
--
'euca-describe-availability-zones verbose'
Public bug reported:
Running 'euca-describe-availability-zones verbose' as a non-admin user
shows the availability zone status _only_ and does not show what vm
instances are available.
I don't think that displaying the available vm types should be limited
to the admin user only. I cannot think
Updated my patch to include support for Windows 7. Correlating current
web server logs and an MSDN article[1] I am pretty sure that this is
correct.
[1]: http://blogs.msdn.com/ie/archive/2009/01/09/the-internet-explorer-8
-user-agent-string-updated-edition.aspx
--
analog logfile analyser fails
** Attachment added: UA string matching for Windows Vista and Windows 7
http://launchpadlibrarian.net/33781882/analog-6.0-ua.patch
--
analog logfile analyser fails to identify Windows Vista
https://bugs.launchpad.net/bugs/222813
You received this bug notification because you are a member of
Public bug reported:
Binary package hint: apache2-common
Problem:
Disabling the default virtual host enables 'AllowOverride Any' for the root
('/') directory and disables 'FollowSymlinks'.
This effects (at least) Hardy and Intrepid's versions of Apache2.
Overview:
The default Apache virtual
** Description changed:
Binary package hint: apache2-common
Problem:
- Disabling the default virtual host enables 'AllowOverride Any' for the root
('/') directory and disables 'FollowSymlinks'.
+ Disabling the default virtual host enables 'AllowOverride All' for the root
('/')
78 matches
Mail list logo