[pve-devel] applied: [PATCH v2 qemu-server] update disk size before local disk migration

[Thomas Lamprecht]

On 12/9/19 2:08 PM, Stefan Reiter wrote:
> Split out 'update_disksize' from the renamed 'update_disk_config' to
> allow code reuse in QemuMigrate.
> 
> Remove dots after messages to keep style consistent for migration log.
> 
> After updating in sync_disks (phase1) of migration, write out updated
> config. This means that even if migration fails or is aborted in later
> stages, we keep the fixed config - this is not an issue, as it would
> have been fixed on the next attempt anyway, and it can't hurt to have
> the correct size instead of a wrong one either way.
> 
> Signed-off-by: Stefan Reiter <s.reiter at proxmox.com>
> ---
> 
> @Thomas: Is this what you had in mind with the local disk migrate bug?

not all, there's still the issue were a 128K source gets drive-mirrored to
a 4M aligned target, and QEMU chokes on it.

> 
> Also, I'm not 100% sure about that last paragraph, if it really doesn't
> matter...
> 
> v1 -> v2:
> * remove $log_fn from update_disksize and return formatted sizes to caller
>   instead
> * include both volid (local-lvm:vm-100-disk-1) and drive bus (efidisk0) in
>   message
> 
>  PVE/QemuMigrate.pm | 15 +++++++++++++++
>  PVE/QemuServer.pm  | 37 +++++++++++++++++++++++++++++--------
>  2 files changed, 44 insertions(+), 8 deletions(-)
> 

applied, thanks!