Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Added section on snapshots and recovery mode

...

Below is a compilation of issues you may encounter when using Nimbus, and workarounds or solutions where possible. This will be added to as time goes on, so we recommend checking here first if you encounter a problem.

Creating

...

and Accessing Instances

Binding failed for port

When creating an instance through the dashboard, it starts to create the instance, but then fails with the following error message in the top right corner:

Info
iconfalse

...

Error: Failed to perform requested operation on instance "<instance_name>", the instance has an error status: Please try again later [Error: Exceeded maximum number of retries. Exceeded max scheduling attempts 3 for instance <instance_id>. Last exception: Binding failed for port <port_id>, please check neutron logs for more information.].

...

This usually means that you have attempted to to set "public external" as the primary network for your instance. This will always fail, as you should be using a private network for all of your instances (which you have to set up before creating any instances). For information on setting up a private network (and a virtual router for external access), please refer to Nimbus - Launching an Instance

Accessing an Instance

SSH Access - permission denied

Once your instance is created, when attempting to SSH into the instance, you may receive the following error:

Code Blockinfo
iconfalse

Permission

denied

(publickey).

There may be a number of possible reasons for this:

...

  • If you have an SSH key par through the Nimbus dashboard, you may instead want to try creating the SSH key on your local machine, then importing the public key into Nimbus. The SSH key par can be generated locally with the command "ssh-keygen". Be sure to ONLY import the public key on Nimbus; the private key must remain where it is.

Creating and Using Snapshots

Error: Unable to create snapshot

When attempting to create a snapshot, it's possible for the command to fail. This is usually caused by insufficient storage quota.

The total size of the snapshot will include the size of any mounted volumes. If you wish to take a snapshot of your instance without inessential volumes, you will need to first detach them from the instance so that the snapshot only includes the most important system files. If you have a separate system volume, you can alternatively, create a snapshot of just this volume:

  1. From the Nimbus dashboard → Volumes
  2. Select Create Snapshot
  3. To review the snapshot in the dashboard, you will find it under Volumes → Snapshots

Accessing Recovery Mode

If your VM fails to boot or there is some other critical issue with your instance (e.g. due to system misconfiguration), it is possible to access recovery mode in Ubuntu systems. Although the instructions below are given for Ubuntu's recovery mode, the procedure is similar for accessing the emergency mode on CentOS systems.

Warning

The recovery and emergency modes are advanced features intended for users experienced with Linux systems. If you are unfamiliar with the system, we recommend using alternative means to recover your VM, such as rolling back to an earlier snapshot or mounting your root volume to a new instance. Please contact help@pawsey.org.au for more assistance.

To access Ubuntu's recovery mode:

  1. From the Nimbus dashboard → Actions menu to the right of your instance, and select Console. You may need to click on the "Click here to show only console" link and refresh the page before the console appears.
  2. In the top-right corner of your console window, click on the "Send CtrlAltDel" button to reboot your instance
  3. As soon as the console goes blank, hold down the shift key to make the GRUB boot menu appear
  4. From the GRUB boot menu, select "Advanced options for Ubuntu", then select the recovery mode for the latest kernel version listed (usually the second option from the top)