Tag Info

New answers tagged

0

I'd suggest looking over this question. It will direct you how to access the inner juju environment where you can start looking over things. After you follow the above, you can explore your juju environment at a high level with juju status. I wouldn't suggest attempting to poke through machines and lxcs by hand without that high level tool. It would be ...


0

Are you talking about network errors on LXC containers that you created inside VMs that you spawned in the cloud? If yes, then it's an MTU problem and there is no automatic fix yet. You need to lower the MTU of the container interfaces to 1454. If you look at the VM interfaces, they already have that MTU.


1

OpenStack is a set of software tools for building and managing cloud computing platforms for public and private clouds. Backed by some of the biggest companies in software development and hosting, as well as thousands of individual community members, many think that OpenStack is the future of cloud computing.What is OpenStack? Basically ...


1

Assuming this is a Single install -- Log into the container sudo lxc-ls -f sudo lxc-attach -n <container> # usually something like uoi-bootstrap Once in the container sudo su - ubuntu export JUJU_HOME=~/.cloud-install/juju Now you can access your juju environment as normal juju status juju ssh nova-cloud-controller/0 ... As far as bringing ...


5

I believe you will have to redeploy the service on that machine. Even in MAAS, it's usually best to recommission the node if its hardware changed.


2

If you use the latest Landscape Autopilot, LDS 15.01, then you need at least: one machine for MAAS one machine for LDS 3 machines for OpenStack Details about the networking setup can be found here: https://insights.ubuntu.com/2015/04/10/maas-network-layouts-for-the-landscape-autopilot/ Or at this askubuntu q&a: How should I setup MAAS so that it ...


2

1) Could be a possible bug, the MAAS UI should support multiple entries. 2) You'll need to make your changes within /etc/bind/maas/named.conf* as those files are included into the top level named options.


0

One instance where this is required is when configuring Neutron OpenVSwitch within Autopilot as you will need to select your 'provider' network. That information is pulled from MAAS, therefore, it needs to be defined within MAAS.


1

The solution to my problem lay in the fact that I'm running my whole Openstack setup in a virtualised environment, in particular each of my compute nodes is a VM. By default my set up did not allow nested VMs, i.e. my compute node VMs could not themselves create instance VMs inside of themselves. I found this article very helpful: Configure DevStack with ...


1

https://launchpad.net/cirros (which http://cirros-cloud.net/ redirects to) says to "download images for cirros at http://download.cirros-cloud.net/". You can find the image you wish to download there, which should be http://download.cirros-cloud.net/0.3.3/cirros-0.3.3-x86_64-disk.img.


0

The default rules that come with the Openstack security policy appear to make everything wide open but I am finding that's just not the case. I had to manually add all the Egress and Ingress ports I needed before everything was working properly. So in this case, it appears to be 100% related to the openstack security policy I was using on the instance


2

The MySQL server cannot start for some reason, this reason should be detailed in the MySQL error log, try tailing /var/log/mysql.err for further information. I've had this happen when some sort of post upgrade action is required. Without seeing the corresponding log entries however, this would be hard to diagnose.


0

I'll add a general answer here that could help others. When encountering such issues, where it's not clear what is failing, the general suggestion is to go simple. In this case, try to provision nodes in MAAS directly with juju instead of going through the cloud installer. It should be much easier and faster to debug. This URL has instructions on to use ...


1

Lots of units deployed just fine according to your log, just landscape-msg/0 failed: [ERROR] deployer.env: The following units had errors:\n unit: landscape-msg/0: machine: 0/lxc/3 agent-state: error details: hook failed: "install" It would be very helpful if you could get the unit-landscape-msg-0.log file from /var/log/juju in the bootstrap node. You ...


3

Similarly to: Landscape openstack juju management You can use the following: # If you used the install-openstack tool, you want to do this first: export JUJU_HOME=~/.cloud-install/juju juju status # Get an interactive shell with the right ENV sourced on the landscape # server itself. juju ssh landscape/0 sudo ...


2

UPDATE: 06 Apr 2015: As of this date, the official instructions are now installing LDS 15.01 that gets around this problem. I believe you are running into github issue 374 which has been addressed in LDS 15.01. Notice in the release notes that upgrading the juju deployed LDS is not directly supported (though it can be done if you follow the ...


1

Our experimental ppa contains support for 'contributed' plugins to the installer. So far we have Heat, Ceilometer, Mongo, and a few others. Once we release 1.0 the plugins will be available in our stable repo. You can install the experimental packages to mess around with them: https://launchpad.net/~cloud-installer/+archive/ubuntu/experimental We'd love ...


1

You can customize the sizes of the VM's via a configuration file containing your services placements and their constraints. install_type: Single placements: controller: assignments: LXC: - nova-cloud-controller - glance - glance-simplestreams-sync - openstack-dashboard - juju-gui - keystone - mysql ...


0

The slightly older version of python-django in your output indicates that you may need to refresh your package lists. To make sure your packaging system is always working with current information, it is advisable to do this before package installations: sudo apt-get update -y


1

The official method is to use the Landscape OpenStack Autopilot. If not using the autopilot, the next recommended method is to use MAAS to commission the machines, then use Juju to deploy OpenStack. In these scenarios, MAAS handles DHCP for the broadcast domain behind your router ("LAN") and neutron handles DHCP for the instances in your tenant ...


0

The bleeding edge PPA's are hosted by the OpenStack Ubuntu testing team - the following are actively maintained: ppa:openstack-ubuntu-testing/icehouse ppa:openstack-ubuntu-testing/juno ppa:openstack-ubuntu-testing/kilo You can enable these using the add-apt-repository command: Example: sudo add-apt-repository ppa:openstack-ubuntu-testing/kilo Testing ...


1

The PPA should be ppa:cloud-installer/stable


3

The metadata commands you used will generate the index and other files. What seems to be the problem is a mismatch between the cloud/region name in those files, and what you have in your juju environments.yaml file. Could you compare them? Specifically the keystone (auth) url needs to match, byte for byte.


0

I re-ran the Landscape Autopilot step and the install completed successfully this time.


0

The guide you are referring to is for setting up a production or PoC instance. MAAS is only needed if you are going to provision multiple physical servers. For a single-node scenario, check out Adam's guide at http://astokes.org/ubuntu-openstack-installer/ for a very simple walkthrough of the installation.


0

It seems you have to sync-tools before you bootstrap. If you can sync during the bootstrap, there is not reason to do it after, as you are up to date (short of an update coming out). The problems I had were around not being able to sync-tools during bootstrap, because my cluster didn't have external access. My work around to do what IO alluded to in my ...


0

Never mind. You are clearly having problems before I was, and my answer won't help you. Your bootstrap machine does not have access to the internet, right? This is what worked for me. I named my environment maas, and my bootstrap machine jujuBS. I made a sub directory under my home called .juju to hold the tools and charms. I copied what I needed locally ...


5

I don't think it means you have to go to a website named maas.ip, but rather that you have to go to http://ww.xx.yy.zz where ww.xx.yy.zz is the IP address of the MAAS server.


0

It was a bios Issue, I have 4 NICs in each node and only two were connected. The timeout for the installer kicked in while the NIC's kept trying to PXE boot.



Top 50 recent answers are included