VM Creation Fails on First Use of New Image Due to Missing Cinder Image Cache in OpenStack with NetApp ONTAP – “Volume did not finish being created” Error
Applies to
- OpenStack (RHOSP18/Antelope and similar)
- Cinder with NetApp ONTAP Block/iSCSI/NFS backend
- Glance multistore using NetApp NFS backend
Issue
When launching a VM from a newly uploaded image (not previously cached) in OpenStack, the instance creation fails with an error similar to:
Error: failed to perform the requested operation on instance, The instance has an error status. Please try again later... Volume <id> did not finish being created even after we waited for 187 seconds or 61 attempts, and its status is creating.
In the Cinder volume logs, the following is observed:
No accessible image volume for image <image_id> found....Downloaded image <image_id> to volume <volume_id> successfully....The NetAppCmodeISCSIDriver volume driver's get_volume_stats operation ran for 232.3 seconds. This may indicate a performance problem with the backend which can lead to instability.
- The first VM creation from a new image fails or takes excessive time (often exceeding default Nova/Cinder timeouts).
- Subsequent VM/volume creations from the same image succeed quickly after the image cache is created in the Cinder backend.
