Qemu-img Differences between Snapshots: Difference between revisions
Jump to navigation
Jump to search
No edit summary |
|||
Line 18: | Line 18: | ||
4. Revert the original filesystem to the state before the event, by reverting to snapshot, with [[Qemu-img#Revert_the_Disk_to_a_Saved_State|qemu-img snapshot -a <''snapshot-tag''> <''volume-name''>]]. | 4. Revert the original filesystem to the state before the event, by reverting to snapshot, with [[Qemu-img#Revert_the_Disk_to_a_Saved_State|qemu-img snapshot -a <''snapshot-tag''> <''volume-name''>]]. | ||
5. Optionally, delete the snapshot with [[Qemu-img#Delete_the_Snapshot|qemu-img snapshot -d <''snapshot-tag''> <''volume-name''>]]. |
Revision as of 18:46, 18 December 2017
Internal
Overview
This is a procedure that can be used to list differences between qemu-img snapshots. It is probably not optimal, and there are probably tools that do this better, but at the time of the writing, I did not know about them.
Procedure
1. Take a snapshot of the filesystem before the event whose effects you want to analyze: qemu-img snapshot -c ...
2. Trigger the event.
3. Save the state of the filesystem after the event, by cloning the filesystem, with virsh vol-clone, after the VM guest has been shut down:
virsh vol-clone --pool main-storage-pool --prealloc-metadata build-guest.qcow2 build-guest-after.qcow2
4. Revert the original filesystem to the state before the event, by reverting to snapshot, with qemu-img snapshot -a <snapshot-tag> <volume-name>.
5. Optionally, delete the snapshot with qemu-img snapshot -d <snapshot-tag> <volume-name>.