Open zachbthomas opened 1 year ago
Thanks @zachbthomas for creating this issue. What exactly are you expecting to see with multiple disk images? A snapshot for all of them so the base image doesn't get modified during an experiment?
Yeah exactly. Each disk image gets a snapshot, and pressing the “create a backing image” button in phenix uses each snapshot file and creates a new .qc2 file for each.
On Thu, Aug 3, 2023 at 4:01 PM Bryan Richardson @.***> wrote:
Thanks @zachbthomas https://github.com/zachbthomas for creating this issue. What exactly are you expecting to see with multiple disk images? A snapshot for all of them so the base image doesn't get modified during an experiment?
— Reply to this email directly, view it on GitHub https://github.com/sandialabs/sceptre-phenix/issues/145#issuecomment-1664697932, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKCMJNWZX3WQQB2K5J3NIGTXTQNSTANCNFSM6AAAAAA3DL4EBQ . You are receiving this because you were mentioned.Message ID: @.***>
Currently the behavior is such that the second disk is always being written to directly whereas the _snapshot file from the first disk is the diff file from the first drive. In other words, we are not preserving the second disk in any way.
@bjwrigh I thought minimega created a snapshot disk in /tmp/minimega for all disks connected to a VM. Is that not the case?
When specifying multiple disk images (multiple volumes) for a VM in a topology, the minimega script is incorrectly generated. Only the first disk image is snapshotted. Subsequent disk images do not get a snapshot file, and as a result, in subsequent disk images the original image file is specified in the experiment's minimega file rather than snapshot files for each disk image