Tag Info

Hot answers tagged

19

In maas 1.2 through 1.4 the virsh power type requires only the Address and Power ID. The libvirt-bin package needs to be installed to get the virsh command $ sudo apt-get -y install libvirt-bin the Power ID is the name of the virtual machine shown by sudo virsh list --all The address is a normal libvirt connect string: ...


11

I suspect they're just trying to stop people nicking portions of the documentation for their own nefarious purposes but no, by all measures that matter to people like us, that document is not free. But the MAAS packages in the main repos all use the AGPL3 license: MAAS is Copyright 2012 Canonical Ltd. Canonical Ltd ("Canonical") distributes the MAAS ...


10

Scope This document provides instructions on how to install the Metal As A Service (MAAS) software. You have sufficient, appropriate node hardware You will be using Juju to assign workloads to MAAS You will be configuring the cluster network to be controlled entirely by MAAS (i.e. DNS and DHCP) If you have a compatible power-management system, any ...


8

The issue you were running into is with apparmor. 'dmesg' would probably have shown you something like: [ 4822.366235] type=1400 audit(1384973058.254:52): apparmor="DENIED" operation="mount" info="failed type match" error=-13 parent=1272 profile="lxc-container-default" name="/mnt/" pid=1273 comm="mount" fstype="ext4" srcname="/dev/loop0/" flags="ro" ...


7

Preparing MAAS for Juju and OpenStack using Simplestreams When Juju bootstraps a cloud, it needs two critical pieces of information: The uuid of the image to use when starting new compute instances. The URL from which to download the correct version of a tools tarball. This necessary information is stored in a json metadata format called ...


7

What are the minimum system requirements for setting up a MaaS cluster, which is going to be orchestrated with juju-jitsu? For the region controller (the master machine), you don't need anything particularly powerful unless you want to manage 100s or 1000s of machines. Cluster controllers place even fewer demands, but unless you're dealing with more than, ...


7

Not a 100% complete answer since I may need more details. Please comment here if you have anything updated you need me to look at. But, I think this could be your issue. The checklist is dynamic, it will compute again when you reload the page. You will not need to re-install landscape. I suspect you probably did try to reload the page already -- I'm ...


6

Try editting /usr/share/openstack/templates/juju-env/maas.yaml


6

MaaS/juju allow you to utilize the hardware that you have one by one, while openstack (this is oversimplification) also allows you to build virtual machines (usually kvm guest), which in many cases significantly improve the physical hardware utilization. In addition, openstack gives you the following: It provides a self-service portal where end users (not ...


6

The OpenStack Autopilot is built for server-class hardware. If you want to use NUCs you'll have to modify them to act like server-class hardware (add USB keys that act like second disks, add a USB NIC to one of them so it has two wired NICs to use for Neutron, etc). In terms of bootstrapping the orange box, you can use some helper scripts found at ...


6

These are the networking requirements for a successful OpenStack Autopilot run. Internet access MAAS and its nodes will need to be able to reach the internet, or at least these sites (http and https): maas.ubuntu.com cloud-images.ubuntu.com streams.canonical.com an Ubuntu archive (archive.ubuntu.com or a mirror) api.jujucharms.com manage.jujucharms.com ...


6

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.


6

if someone deploys maas+juju for openstack, please run juju status --format=png > architecture.png


6

The DHCP configuration is for the MAAS server to be managing it's own subnet; this is how you would usually have a MAAS region/controller set up. The fields are as follows: IP: This is the IP address of your MAAS controller; you are correct in assuming that it would be 172.16.142.61 if you installed it at this address. Subnet Mask: This is the subnet mask ...


6

DISCLAMIER: I AM NOT AN OPENSTACK EXPERT. THE FOLLOWING IS MERELY AN OPINION AND MAY BE OUTDATED OR COMPLETELY INCORRECT. IN NO WAY CAN I BE HELD RESPONSIBLE FOR CONSEQUENCES IN USING THE CONTENT BELOW. Multi-Hosting Having networking services running on every node will have its advantages and disadvantages. Many different networking redundancy ...


5

I am afraid this is currently not a supported configuration for MAAS (but i would be happy to be wrong on this). I also have to use MAAS behind a proxy, and here is how i dealt with it: importing the pxe images: sudo http_proxy=http://my.proxy:3128/ maas-import-pxe-files Edit /usr/share/maas/preseeds/enlist_userdata set apt_proxy #we cannot use maas as ...


5

I think you are probably hitting this bug: https://bugs.launchpad.net/ubuntu/raring/+source/ceph/+bug/1218852 Right now the version of Ceph in 13.04 and the Ubuntu Cloud Archive for Grizzly can't deal with uppercase letters in hostnames. Your hostname appears to be 'CS1'.


5

If you verified there are no obvious errors, you can always retry running the hook to see if it resolves the issue: juju resolved --retry keystone/0 If the problem still persists, you could try: juju debug-hooks keystone/0 config-changed, which will SSH into the machine and wait for the config-changed to run. You need to manually trigger the hook outside ...


5

You may refer to Manual DHCP configuration in the following link http://maas.ubuntu.com/docs/configure.html#manual-dhcp There is also a nice document by DELL (Deploying workloads with Juju and MAAS) where you can find a section for using an existing DNS/DCHP server and you can find it here ...


5

Yes, for one machine MAAS is absolutely overkill. However Juju can (as of version 1.18) deploy to any Ubuntu Server with OpenSSH via manual provisioning, we designed this feature for cases like yours where you want to deploy services quickly without needing a cloud: https://juju.ubuntu.com/docs/config-manual.html The basic process is you register your ...


5

Consider a higher time-out. bootstrap-timeout: 1800 In your environments.yaml should work.


5

Using --to flag without containerization is a really bad idea. We've likened this "Hulk Smashing". Basically you're layering a ton of services on top of each other that all expect to own the machine. So, what can you do to achieve isolation and still keep everything on one machine? Containerization! The --to flag has a finesse about it which allows you ...


5

The reason for this is that OpenStack is not really meant for a 2 node cluster. It is meant to scale to thousands of nodes. It has many disconnected components, including mysql, rabbitmq, several API services, etc. For your case of "just trying openstack out" you can simply use the local provider to install everything on one box rather than MaaS which wants ...


5

The failure that you see is because this is a virtual machine. The ipmi_si related stuff will only work on real hardware, so the error doesn't really affect your case scenario and you don't really need a fix for it. A good idea would be if you were able to post the output of /proc/cpuinfo so that we can make sure that the module above is not loaded on ...


5

Each node that MAAS prepares for you has your ssh key on it, so you can automate these tasks using any method that can use ssh as transport, so that certainly includes Juju. That pushes you into the realm of customising your charms to do these set-up tasks, which might fit your expectations, but is going to be awkward if you're using off-the-shelf charms. ...


5

Ubuntu Cloud and MAAS are different things - different layers if you like. MAAS is a low-level layer which deals with physical hardware. It is for deploying the base level software to a rack of servers quickly and easily. It can then work in conjunction with juju to deploy particular services (e.g. web servers, databases) to that hardware. This could be for ...


5

--to is what you looking for: juju deploy wordpress --to 0 juju deploy mysql --to 0 Will deploy to the same 0 node. References and examples: https://juju.ubuntu.com/docs/charms-deploying.html#deploying-to-specific-machines-and-containers


5

If you're using the install resources on the ISO and not the OpenStack Autopilot, then there are no licensing restrictions at all and you can provision and install on as many servers as you want. If you want to use the Autopilot and/or manage those machines with Landscape post-install, you'll need to purchase additional seats to cover your needs. Hope that ...


5

Ubuntu OpenStack: the Canonical distribution deploys OpenStack in Canonical's reference architecture. The mechanism through which you install OpenStack uses Landscape to talk to MAAS and Juju and deploy the charms in that architecture. You get added benefit from Landscape doing the work because it helps you monitor your OpenStack setup after it's installed. ...


5

LDS stands for "Landscape Dedicated Server". From landscape.ubuntu.com: The Landscape systems management tool helps you monitor, manage and update your entire Ubuntu infrastructure from a single interface. Part of Canonical’s Ubuntu Advantage support service, Landscape brings you intuitive systems management tools combined with world-class support. ...



Only top voted, non community-wiki answers of a minimum length are eligible