andrijapanicsb commented on PR #13656: URL: https://github.com/apache/cloudstack/pull/13656#issuecomment-5037856846
@rp- — this PR now also adds **Linstor as a destination** for the VMware→KVM import/migration paths (cold staged, cold direct-VDDK, and warm CBT), on top of the Ceph/RBD support. I'd really appreciate a review of the Linstor/DRBD side from you. To make that easy there's a standalone write-up committed in the PR: [docs/vmware-linstor-migration.md](https://github.com/andrijapanicsb/cloudstack/blob/kvm-rbd-vmware-migration/docs/vmware-linstor-migration.md) The Linstor-side design points I'd most like sanity-checked: 1. **Pre-created resources + `qemu-img convert -n`** — since qemu can't create a DRBD device, the agent pre-creates each resource at source capacity via `LinstorStorageAdaptor` and writes with no-create mode (relying on DRBD's size round-up being >= source). 2. **Single-Primary / auto-promote** — the conversion host opens the device exclusively during the copy and the in-place virt-v2v finalization; devices return to Secondary afterwards. Any pitfalls we should guard against? 3. **48-char resource names, no rename** — short deterministic names (`<uuid>-d%02d` for direct imports, `cbt-<mig8>-<diskId>` for warm CBT); imported volumes keep the name as their recorded volume path. `listPhysicalDisks` is now implemented (RD list filtered by the pool's resource group). 4. **Diskful-satellite preference** for the conversion host is recommended in the docs but not enforced — should it be? Validated on a 3-node LINSTOR/DRBD 9 lab: byte-identical disk copies (`qemu-img compare`) for both cold and warm paths, plus a full end-to-end `importVm` producing a CloudStack VM with Linstor-backed volumes, UpToDate on all replicas. Happy to adjust anything on the Linstor side — thanks! -- 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]
