The rescue extension lets you put a server into a “rescue” status when the virtual instance will be replaced with a “rescue” image and have its existing drive attached as a second disk so that a root user can log in and troubleshoot problems with the virtual server.
To put a server into “rescue” state, you do not have to be an administrator. The only requirement is that the base image used to build your instance must still be available.
To obtain current information the extensions available to you, issue an EXTENSION query on the OpenStack system where it is installed, such as http://mycloud.com/v1.1/tenant/extensions.
As shown below, responses to an EXTENSION query in XML or JSON provide basic information about the extension.
Extension Query Response: XML:: N/A
Extension Query Response: JSON:
{"extensions": [{"updated": "2011-08-18T00:00:00+00:00", "name": "Rescue", "links": [], "namespace": "http://docs.openstack.org/ext/rescue/api/v1.1", "alias": "os-rescue", "description": "Instance rescue mode"}]}
Revision Date | Summary of Changes |
2011-09-16 | Initial draft |
This extension to the OpenStack Compute API enables rescue of running instances.
To support these new actions, the extension also issues new states.
rescue unrescue
None
None
None
RESCUING UNRESCUING
A new action added to the 4.3 Server Actions section.
Verb | URI |
POST | /servers/id/rescue |
Normal Response Code(s): 202
Error Response Code(s): computeFault (400, 500, …), serviceUnavailable (503), unauthorized (401), forbidden (403), badRequest (400), badMethod (405), overLimit (413), itemNotFound (404), badMediaType (415), buildInProgress (409)
Status Transition: ACTIVE -> RESCUING -> ACTIVE
This operation places the server into RESCUING status.