On 2020/6/7 下午7:50, Andrei Borzenkov wrote: > 07.06.2020 13:09, Goffredo Baroncelli пишет: >>> >>> Unallocated: >>> /dev/mapper/btrfs1 1.02MiB >>> /dev/mapper/btrfs2 930.49GiB >> >> The old disk is full. And the fact that Metadata has a raid1 profile >> prevent further metadata allocation/reshape. >> The filesystem goes RO after the mount ? If no a simple balance of >> metadata should be enough; pay attention to select >> "single" profile for metadata for this first attempt. >> >> # btrfs balance start -mconvert=single <mnt-point> >> >> This should free about 4G from the old disk. Then, balance the data >> > > E-h-h ... how btrfs knows it should free 4G from the old and not form > the new disk? I guess it is old request to allow explicitly select > "victim" and "target" devices during balance. Btrfs always choose the disk with the most unallocated space. That's why btrfs balance would always "balance" the usage when possible. Thanks, Qu > > IOW can we rely on btrfs doing "the right thing" always or is just > coincidence? >
Attachment:
signature.asc
Description: OpenPGP digital signature
