[pve-devel] applied: [PATCH qemu] add revert to work around performance regression when backing up large RBD disk

Thomas Lamprecht t.lamprecht at proxmox.com
Thu May 19 09:36:44 CEST 2022


Am 5/17/22 um 10:29 schrieb Fabian Ebner:
> resulting in QMP timeouts and very slow backups. The plan is to figure
> out (ideally together with upstream) a way to make the implementation
> of bdrv_co_block_status for RBD more efficient. But for now, revert
> the problematic change as a stop-gap measure.
> 
> Upstream bug report:
> https://gitlab.com/qemu-project/qemu/-/issues/1026
> 
> Forum threads:
> https://forum.proxmox.com/threads/109272/
> https://forum.proxmox.com/threads/109448/
> https://forum.proxmox.com/threads/101334/ (partially)
> 
> Signed-off-by: Fabian Ebner <f.ebner at proxmox.com>
> ---
> 
> I simply removed the other two patches already fixing the offending
> commit from the "series" file, to make it less likely that we forget
> to re-add them once we revert the revert and to reduce the git diff.
> Hope that is ok.
> 
>  ...k-rbd-implement-bdrv_co_block_status.patch | 161 ++++++++++++++++++
>  debian/patches/series                         |   3 +-
>  2 files changed, 162 insertions(+), 2 deletions(-)
>  create mode 100644 debian/patches/pve/0053-Revert-block-rbd-implement-bdrv_co_block_status.patch
> 
>

applied, thanks!




More information about the pve-devel mailing list