GutoVeronezi commented on PR #8041: URL: https://github.com/apache/cloudstack/pull/8041#issuecomment-1748900767
> Thanks @GutoVeronezi for raising the PR. > > I've a question here. As I understood with this PR we are now backing up the consolidated snapshot file which is good. How is this PR different from reverting the original PR #5297 where I see lot more changes wrt to the behaviour that had introduced. Do we also need to address any code in other places ? @harikrishna-patnala Before #5297, we were doing a **full VM snapshot** and extracting the volume we wanted to back up to the secondary storage. However, the full VM snapshot is a longstanding process that affects the VM's memory and disks; while this process is running, the VMs get frozen due to memory snapshot, and users are unable to manipulate their VM for a long time (if the amount of memory is huge). The proposal #5124 intention was to improve this process by taking an **external disk-only snapshot**; this process would **still produce a consolidated snapshot**; therefore, the final result would be the same. However, as explained in https://github.com/apache/cloudstack/issues/8034#issuecomment-1746725748, due to a misconception on the implementation of the specific creating snapshot workflow, we have reference to the volume template on the snapshot. This reference only exists if we take snapshots of `ROOT` volumes that were not migrated yet; if we take a snapshot of a `ROOT` volume that was already migrated at some point in time, the migration process would already have consolidated it (the volume), and we would have a consolidated snapshot. Also, `DATADISK` volumes do not have a reference to a template; therefore, we would not have a problem in this situation as well. This PR is fixing this specific case by using `qemu-img convert` instead of copying the base volume to anothery directory in this specific workflow, which is different from reverting the PR (issue #5124 presents in detail the full scope of PR #5297). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
