The dashboard is open to the public Internet, however various mechanisms are in place to protect the service. If you are having difficulty accessing the dashboard, please contact us.
Our APIs are currently exposed to customers only (not open to the public Internet). We ask for your IP address during sign up, so we can provide you with access to the APIs. If your IP address has changed or you would like to add more IPs, please contact us to change the information provided.
Please note you can access the Catalyst Cloud from a dynamic IP by proxying your connectivity through one of your cloud instances. When it comes to your compute instances, you are in full control of your own firewall rules and can expose them to the Internet.
Delete errors are often caused by dependencies preventing you from deleting a resource. For example:
Trying to delete a volume, when a snapshot depends on it.
Trying to delete a network, when a port is still connected to it (such as a router’s interface)
Error messages provided by the dashboard tend to be brief and sometimes lack specific details about what caused the error. If you are repeatedly being presented with a delete error, please check whether resource dependencies could be the cause. The command line tools are designed to be more verbose and tend to give more information about what is causing a delete error.
Snapshots of instances that were previously created from snapshots need to be booted with the correct options in order to allow further bootable images to be created.
The default option (on the Dashboard, Compute -> Images -> Launch) to launch an instance is to ‘boot from image’. This is correct, unless the original instance also created a new volume at start-up (the instance will boot, but not itself be able to take a snapshot without additional metadata).
In the latter case, where the initial instance was created along with a new volume, the snapshot needs to be booted with the Instance Boot Source set to “Boot from snapshot (creates a new volume)”.
An instance accidentally booted with incorrect options can be corrected by either of the following:
Shut down the existing instance and re-launch it as above.
Alternatively, use the Glance API client to update the metadata for the image that backs the existing instance:
$ glance image-update <image-name-or-id> --container-format bare --disk-format raw