Tag Info

New answers tagged

0

Hope someone finds an answer for this, I'm having the same problem.


1

To find the resilience of your data, try sudo ceph osd dump | less In the beginning of the output, it will show you the pools and the replicated size or the erasure coding count depending upon your setup. Instance live migration depends upon the type of volume storage used. If its a volume backed storage, then you can use live migration. (And yes, those ...


0

Ensure that within neutron-api charm, "neutron-security-groups" is set to True.


0

I have specified use-floating-ip: true and the juju-openstack-machine-0 (the bootstrap state machine) does get a public IP. That's fine and the juju client can talk to it in the first stages of the bootstrap process. The problem comes in the stage Bootstrapping Juju machine agent My agent-metadata-url points to a swift container endpoint My ...


0

When you start an instance manually, via horizon for example, what do you have to do to access it? If allocating a floating ip is enough, then you can add this to your juju environments.yaml file in the section for your cloud: use-floating-ip: True With that, juju will ask the cloud to allocate a floating IP for the instance.


0

Adding neutron-api and neutron-openvswitch charms - followed by reboot of cloud controller and rabbit servers - did the trick. (There is a new problem, i.e., Error: Unable to retrieve security groups; however, will open a separate thread on that.)


1

You can enable the Ubuntu Cloud Archive for OpenStack Kilo on Ubuntu 14.04 installations by running the following commands: sudo add-apt-repository cloud-archive:kilo sudo apt-get update For more info check the following source: https://wiki.ubuntu.com/ServerTeam/CloudArchive For a step by step installation: ...


0

You can start simple, install juju in the same server where you have MaaS. What you want now is to get the latest juju from the ppa: apt-get install python-software-properties sudo add-apt-repository ppa:juju/stable sudo apt-get update && sudo apt-get install juju-core Now the juju configuration: juju generate-config Follow the documentation ...


-2

Does anyone know how to login to landscape, or what is default port (as I am unable to login to the server that landscape may have been installed. )


0

Looks like the MAAS nodes never left the "pending" state. Can you select a node in MAAS and click "start", and see if it reaches the "Deployed" state? I suggest to configure juju to use MAAS (https://jujucharms.com/docs/stable/config-maas) and then try to bootstrap and deploy the simple ubuntu charm to all nodes. This is a good pre-flight check. Once you ...


0

Try to follow the steps provided on this site: OpenStack Install guide


1

You have more locale environment variables set than you really need. LANG is a fallback for when no LC* variable is set, and LC_ALL overrides all the others. So in your case, LC_ALL=C is overriding LANG=en_US.UTF-8 and all the others. The installer needs UTF-8. I'm not actually sure which specific LC_* variable(s) it uses, probably several of them. You ...


2

I've seen this issue when the compute nodes can't resolve their own hostname when sudo commands are executed, resulting in stderr output which breaks most of the command parsing across nova and neutron. You can check this by running: sudo su - on any compute node - if you get an 'unable to resolve' type message, this is the problem. This normally ...


1

After you get access to the inner cloud environment (see http://askubuntu.com/a/598035/7056), you have to enable these settings in the nova-compute service (one command line): juju set nova-compute enable-live-migration=True enable-resize=True migration-auth-type=ssh Then watch juju debug-log until it quiets down, and your nova resize command should work: ...


1

To manage/administer your openstack environment, you will need to use Juju to get access. You can see the steps here: http://askubuntu.com/a/598035/7056 Once you do that, you will need to debug this service like any other. Landscape and Maas at this point are just guiding the installation. Certainly if you identify what is going on here, it would be ...


2

You'll want to access the underlying juju environment as described in your previous question: Add node to existing landscape openstack From there you'll want to run the following: juju set nova-cloud-controller "console-access-protocol=novnc" That'll enable the console via Horizon dashboard.


3

There is no official way through the Landscape Autopilot to do this right now. This functionality is still under development. Juju underneath supports expanding capacity of services and the nova compute charm also supports expanding capacity, that said, you could run into issues doing this, and it's entirely unsupported, dirty, messy and all that. I would ...


0

With the default image - you can't. Most cloud images don't have username/password authentication enabled, and rightfully so - using key pairs is much safer. But the console will not allow you to log in with a key-pair. For development/debugging purposes, it is a good idea to add a 'debug' user to your image, with a configured password. Just don't use this ...


0

I solved it by not using openstack key but by creating my own and imported to openstack


0

I re-installed the cloud and applied this fix on nova compute node charm in the config-flag box vif_plugging_is_fatal=false,vif_plugging_timeout=0. After the restarted DHCP, things appear to be working error free.


1

I've fixed the problem by connecting to the Computing Module and: Adding the following line at /etc/nova/nova.conf: libvirt_type=qemu And performing a restart of the Nova VM. HTH


0

Our experimental ppa contains a plugin for heat that you can access via the add unit dialog (keypress 'A'). If using our stable package you can do the following to add heat to your stack: # Obtain the container name sudo lxc-ls # Attach to container sudo lxc-attach <name> # become ubuntu user and set JUJU_HOME sudo su - ubuntu export ...


0

1st search for rejoin-stack.sh on your machine and copy and save its path on your machine. 2nd use the terminal by pressing CTRL+ALT+T on your keyboard. Compute Nodes: 3rd type sudo service openstack-nova-api restart sudo service openstack-nova-cert restart sudo service openstack-nova-consoleauth restart sudo service openstack-nova-scheduler ...


0

don't know if you still have this issue but maybe for others. I had to configure the parameter "network-manager (string)" in my nova-cloud-controller charm. You have to set the right network manager. In my case the following string: Neutron|Quantum


1

MAAS will be on its own node, and most people usually run the installer from the MAAS machine. You'll still need 7 machines as stated in the requirements section from that document.


2

We figured it out. We needed to add configuration to neutron-openvswitch to enable LBaaS agent. juju ssh neutron-openvswitch/0 Install lb agent sudo apt-get install neutron-lbaas-agent Modify /etc/neutron/neutron.conf by adding the following lines service_provider = ...


0

Because I've done the openstack-install from my MAAS machine, I had to ssh to it and then follow these steps: # 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 ...


7

Please read the following question: Ubuntu landscape openstack installer troubleshooting help It should be exactly what you need to start troubleshooting/exploring your environment. Good luck!


1

It could be several things indeed. So let's go through the generic steps of getting logs from a failed deployment. First let's show how to get access to the different juju environments in play here: Landscape environment: where the Autopilot was deployed Autopilot environment: the cloud environment created by Autopilot Taken from Landscape openstack juju ...


1

This launchpad bug is the same issue, seems better to handle it there probably?



Top 50 recent answers are included